Family business agency conflict model: a study of Brazilian firms

DALTON PENEDO SARDENBERG FAMILY BUSINESS AGENCY CONFLICT MODEL – A STUDY OF BRAZILIAN FIRMS – A Thesis Presented for the degree of Doctor of Philosophy BIRMINGHAM BUSINESS SCHOOL 2012 FAMILY BUSINESS AGENCY CONFLICT MODEL – A STUDY OF BRAZILIAN FIRMS – by Dalton Penedo Sardenberg A thesis submitted to the University of Birmingham in fulfilment of the requirements of the Degree of Doctor of Philosophy Supervisor: Professor John Child Second Supervisor: Professor Suzana B. Rodrigues Birmingham Business School University of Birmingham June, 2012 University of Birmingham Research Archive e-theses repository This unpublished thesis/dissertation is copyright of the author and/or third parties. The intellectual property rights of the author or third parties in respect of this work are as defined by The Copyright Designs and Patents Act 1988 or as modified by any successor legislation. Any use made of information contained in this thesis/dissertation must be in accordance with that legislation and must be properly acknowledged. Further distribution or reproduction in any format is prohibited without the permission of the copyright holder. ABSTRACT The aim of this thesis is to deepen understanding of how agency conflict arises in the governance of family businesses. The constructs were proposed arising from empirical research, involving four case studies of different Brazilian familyowned companies, which were analysed inductively. Next, data from the literature was used to validate those constructs, which became part of the proposed Family Business Agency Conflict Model and to suggest articulated hypotheses. This model identifies factors that can potentially increase, and those which can mitigate, tensions between the actors involved representing family, ownership and management within family-owned firms. It was subsequently tested through a survey of 152 family-owned firms of varying size, which belonged to different economic sectors and regions of Brazil. Structural equation modelling was used to test the empirical validity of the model, with reference to hypotheses concerning the variables which influence agency conflict. The model was found to predict 38 per cent of the variance in agency conflict. Four variables were found which explain directly the agency conflict in family businesses: strategic alignment, social alignment, trust and impartiality. Therefore, the findings of this research besides focusing on the usual procedural aspects of governance, such as disclosure and control, include behavioural elements which constitute a step ahead in the study of corporate governance in family business. Keywords Agency conflict, corporate governance, family business. ii My joy in learning is partly that it enables me to teach. Seneca iii ACKNOWLEDGMENTS In the continuous learning process that I underwent throughout the development of this thesis, many have made decisive contributions, and I should like to record my indebtedness to them. In the first place I would like to put on record the fundamental role of my supervisor, Prof. John Child, who, always in stimulating fashion, posed questions that challenged the way the study was being carried out, thereby drawing out new insights. His wisdom and serenity will remain forever a source of inspiration in my activities as a professor and researcher. Prof. Suzana Braga Rodrigues, from whom I received important guidance, also deserves my thanks. She was initially my main supervisor, and even after her departure for the Erasmus University Rotterdam, we remained in close contact. Her objectivity, lucidity and academic rigour were essential for laying the foundations necessary for the preparation of this dissertation. Emerson de Almeida, Dean of the Fundação Dom Cabral, who was the first to suggest to me the challenge of doing a doctorate outside Brazil, really started the whole process. I record my debt to the Fundação Dom Cabral for sponsoring the doctorate, and to all my colleagues in this institution who have supported me with encouragement and suggestions. I would like to mention especially, those that distributed and gathered the completed survey questionnaires: Adriana Netto, Antônio de Pádua S. Teixeira, Gilmar Mendes, José Helio Jóia, Juliana Costa Gonçalves, Luzia Frönlich, Marcelo Pelágio Bonfim, Maria Eliane Longo, Maria Teresa Roscoe, Paulo Emílio Carreiro, Ricardo Langanke, Seniria Maria Silva, Valdemar Barros Filho and Vinícius Vieira de Freitas. To my friend José Antônio de Sousa Neto who generously introduced me to the University of Birmingham, where he himself had obtained his Master’s and Doctoral degrees, I record my obligation. Marina Araújo and Rafael Diogo Pereira, researchers who through extensive exchanges of ideas contributed so much to the positive evolution of this study deserve special mention. iv I would like to thank Roger Arthur Gough for our English classes, where we went over this thesis paragraph by paragraph, with the aim of producing a final version that would be in the best British style. Prof. Plínio Rafael Monteiro gave me his invaluable aid and support in the use of statistical packages. Likewise, my thanks to all the professors of the University of Birmingham, whose classes have helped me to augment my understanding of the content relevant to the development of this thesis. In the same way I thank the professors of FACE – the Faculty of Administration and Economic Sciences of the Federal University of Minas Gerais who allowed me, exceptionally, to complete the courses relative to the credits recommended by my supervisors. My sincere thanks go to all those who have kindly answered my survey, forming the data base necessary for doing the statistical analyses, and also, especially, the family, and nonfamliy members of the four companies studied, who have not been cited here for reasons of confidentiality, and who received me with open arms, readily sharing with me their views of the family and business issues, and the tensions arising out of the interaction between the two. My father and mother, Carlos Paiva Sardenberg (in memoriam) and Maria Nazareth Penedo Sardenberg, must be mentioned, for inculcating in me a spirit of perseverance and commitment to the reaching of one’s personal goals. My sons Arthur and Miguel Sardenberg have made their contribution in the joy they have provided every moment we have been together, and for showing me that curiosity and involvement with the object that awakens it are central elements in the learning process. This mission would have been an impossible one without the special encouragement and companionship of my wife Andréa Tamiozzo da Costa Sardenberg, whose untiring dedication to our children minimized the effects of the inescapable lack of the attention that they deserved, due to the time that I had to devote to professional and academic activities. To her go out my love and gratitude. And finally I would like to thank God for the gift of life, and the inspiration to develop the talent that I have been blessed with, so that at the end of my earthly sojourn, I will have been able to return to others a little of everything I have received. v TABLE OF CONTENTS I. INTRODUCTION I.1. Background ………………………………………………………………………… 01 I.2. Research motives …………………………………………………………………. 02 I.3. Research objectives …............................................................... ……………... 04 I.4. Contribution of the research ……………………….……………………………... 05 I.5. Structure of the thesis …………….…….….……………………………………... 07 I.6. Format of this document ………………………………………………………….. 08 II. PRELIMINARY LITERATURE REVIEW II.1. Family-owned company…………………………………………………………... II.1.1. Family business definitions ………………………………………………. II.1.2. Family business characteristics …………………………………………. II.1.3. Family business conflicts …………………………………………………. II.2. Corporate Governance and Agency Theory ………………………………..…. II.2.1. The real meaning of Corporate Governance …………………………… II.2.2. Agency Theory within the Corporate Governance studies …………… II.2.3. Agency Theory applied to a Family Business ……………………..…… II.2.4. Criticisms of Agency Theory ……………………..………………………. 09 09 15 19 23 23 27 31 34 III. QUALITATIVE RESEARCH METHODOLOGY III.1. Research methodology approach ……………………..……………………..… 39 III.2. Grounded Theory ……………………..……………………………………….… 39 III.2.1. Initial data collection …………………….……………………………..… 43 III.2.2. Application of the method …………………….………………..……..… 45 III.2.3. Results of non-participative observation …………………....……..… 47 III.3. Case Studies ……………………..…………………………………………….… 49 III.3.1. Interviews .…..………………..………………………………………...… 50 III.3.2. Interview guide ………………..………………………………….…….… 51 III.3.3. Unit of analysis ……………..………………………………….…….…... 54 III.3.4. Secondary data gathering ……………..………………………………... 55 IV. INTRODUCTION OF CASE STUDIES IV.1. The Milk & Candies Case ………………..………………...………...……….… IV.2. The Bental Case .……………..………………...………...................……….… IV.3. The Prize Case .……………..………………...………...................………...… IV.4. The Case of Minas Agro Group ………………..………………...………...….. V. THEORETICAL MODEL AND HYPOTHESES V.1. Building a theoretical model ………………….………………..………………... V.2. The use of Atlas.ti ………………….…………………………..………………... V.3. The process described …………….…………………………..………………... V.4. Proposed theoretical model ………………….………………..………………... V.5. Validation of hypotheses and constructs through a comparison of the …….. literature vi 57 75 104 122 135 140 141 145 146 V.5.1. The Alignment Construct ……..…….………………..………………..... 147 V.5.2. The Level of Independence and Efficiency of the Board Construct … 153 V.5.3. The Family and Business Interaction Construct ….……..…….……… 160 V.5.4. The Management Control Construct …...…………………..…….…..... 167 V.5.5. The Trust Construct ...…………………….…………………..………..... 175 V.5.6. The Disclosure and Transparency Construct ………………………..... 179 V.5.7. The Legitimacy Construct …...…..……….…………………..………..... 188 V.5.8. The Competence Construct ………………...…………………………... 197 V.5.9. The Opportunism Construct …………………………………………...... 201 V.5.10. The Impartiality Construct ……………………………………….…...... 207 V.5.11. The Altruism Construct ……………………………………….….......... 212 V.5.12. The Comprehension of Social Roles Construct …...……..………..... 217 V.6. Development of the Questionnaire ……………………………………………... 226 V.6.1. Problem definition …………………………………………………………. 226 V.6.2. Preparation of a questionnaire draft …………………………………….. 226 V.6.3. Pilot test …………………………………………………………………..... 228 V.6.4. Final revision …………………………………………………………...….. 229 V.6.5. Questionnaire application ……………………………………………..…. 235 VI. QUANTITATIVE RESEARCH METHODOLOGY VI.1. Data analysis …………………………………..………..……………………..… 236 VI.1.1. Exploratory Analysis …………………..…………..………..….…….… 236 VI.2. Analysis of the quality of the measurement ..………………..…………….… 247 VI.2.1. Analysis of the dimensionality …………………….…..…………….… 247 VI.2.2. Reliability analysis and overall measurement quality …..…………... 252 VI.2.3. Convergent and discriminant validity ….……………..………….….… 257 VI.3. Nomological validity – Test of the model ….……………..………………....... 261 VI.4. Summary of the quantitative results ….…………..…………..………..……... 264 VI.5. The Family Business Agency Conflict Model - validated version ….…….… 265 VI.6. Findings …………………………………..………..…………………………..… 266 VI.6.1. Control variables for conflicts in family businesses ………..……..… 266 VI.6.2. Analysis of the Family Business Agency Conflict Model as a whole 268 VI.6.3. Analysis of each construct and comparison with qualitative findings 269 VII. DISCUSSION VII.1. Closing remarks ………..………..……………………………..………..……... 281 VII.2. Contributions to knowledge …....……………………………….……..……... 283 VII.2.1. Theoretical and empirical contributions …….………………………. 285 VII.2.2. Practical implications ….……………………………………………….. 288 VII.4. Limitations and directions for further research …………………………….… 298 REFERENCES ……………………………………………………………………….… 303 APPENDIX A: Questionnaire for quantitative survey .………………………….….. 339 vii LIST OF ILLUSTRATIONS Figure 1.1 Scheme representing the development of the structure of the thesis 07 Figure 2.1 Three-circle-model 18 Figure 2.2 Three levels of individual and collective conflict in the entrepreneurial environment 21 Figure 3.1 The grounded theory 'data dance' 42 Figure 3.2 Evolution of the Brazilian capital market 48 Figure 4.1 Haag Family Tree 60 Figure 4.2 Mills Family Tree 61 Figure 4.3 Milk & Candies’ executive team 61 Figure 4.4 Bental time line 76 Figure 4.5 Corporate structure of the Talma Group 86 Figure 4.6 Structure of the Talma Family in 2008 97 Figure 4.7 Governance model of the Bental Group in 2008 98 Figure 4.8 Schedule of activities of Bental’s Board of Directors 99 Figure 4.9 Organization chart of Prize 111 Figure 4.10 Enterprises of the Minas Agro Group 122 Figure 5.1 The grounded theory analytic process 137 Figure 5.2 Open coding sequence 138 Figure 5.3 Theoretical model construct on the basis of Grounded Theory 145 Figure 5.4 Categories of the Alignment construct 147 Figure 5.5 Categories of the Level of Independence and Efficiency of the Board construct 154 Figure 5.6 Categories of the Family and Business Interaction construct 161 viii Figure 5.7 Categories of the Management Control construct 167 Figure 5.8 Categories of the Trust construct 175 Figure 5.9 Categories of the Disclosure and Transparency construct 182 Figure 5.10 Categories of the Legitimacy construct 190 Figure 5.11 Conflict Across Generations 192 Figure 5.12 Conflict and the Shadow Effect 193 Figure 5.13 Comparison of entry strategies for succession in family business 193 Figure 5.14 Categories of the Competence construct 198 Figure 5.15 Categories of the Opportunism construct 202 Figure 5.16 Categories of the Impartiality construct 208 Figure 5.17 Categories of the Altruism construct 214 Figure 5.18 Categories of Comprehension of Social Roles construct 219 Figure 6.1 Range of annual revenues of the company 237 Figure 6.2 Level of economic and financial performance of the respondent company 237 Figure 6.3 Number of permanent employees maintained by the company 237 Figure 6.4 Constitution of the capital 238 Figure 6.5 Lifetime of the companies surveyed 238 Figure 6.6 % of respondents from companies with formal board of directors 238 Figure 6.7 % of respondents from companies with independent directors of those who adopt a board 239 Figure 6.8 % of respondents from companies with Chairman from the controlling family 239 Figure 6.9 Family, nonfamliy CEO? 239 Figure 6.10 To which generation does the family CEO belongs 240 Figure 6.11 % of respondents from companies with Family Council 240 ix Figure 6.12 % of respondents members of the owner family 240 Figure 6.13 % of respondents belonging to each generation 241 Figure 6.14 % of respondents working in the company 241 Figure 6.15 % of respondents who are shareholders of the company 241 Figure 6.16 % of respondents members of the board 242 Figure 6.17 Scree Plot of the Agency Conflict Construct 251 Figure 6.18 Theoretical model modified after EFA 261 Figure 6.19 Research model tested via PLS 263 Figure 6.20 Reduced level of agency conflict 266 x LIST OF TABLES Table 1.1 Specific research questions 05 Table 2.1 Conflict in Family Business Studies 22 Table 3.1 Results of non-participative observation 47 Table 3.2 Origin of interview guide and its main questions 51 Table 4.1 The Milk & Candies’ capital structure 59 Table 4.2 Milk & Candies’ business performance 59 Table 4.3 Milk & Candies’ profile of the interviewees 60 Table 4.4 Bental’s profile of the interviewees 75 Table 4.5 Prize’s business performance 106 Table 4.6 Prize’s profile of the interviewees 106 Table 4.7 The Prize’s capital structure 119 Table 4.8 Minas Agro’s profile of the interviewees 124 Table 5.1 Coding scheme for all relevant categories and subcategories and the correlated quotations 142 Table 5.2 Source of measurement of constructs 230 Table 6.1 Data descriptive analysis 243 Table 6.2 Factor analysis 249 Table 6.3 Descriptive Statistics of the Agency Conflict Construct 250 Table 6.4 Correlation Matrix(a) of the Agency Conflict Construct 250 Table 6.5 KMO and Bartlett's Test of the Agency Conflict Construct 251 Table 6.6 Communalities of the Agency Conflict Construct 251 Table 6.7 Total Variance Explained of the Agency Conflict Construct 251 Table 6.8 Component Matrix(a)of the Agency Conflict Construct 252 xi Table 6.9 Quality of the measurement of the scales 256 Table 6.10 Convergent validity tests 258 Table 6.11 Discriminant validity test 260 Table 6.12 Results of the hypotheses of the model proposed by the dissertation 263 Table 6.13 Results of hypotheses testing 264 Table 6.14 Summary of findings 267 xii LIST OF ABBREVIATIONS AVE - Average Variance Extracted BNDES - Brazilian National Bank for Economic and Social Development BMF&BOVESPA - Bolsa de Valores de São Paulo (São Paulo Stock Exchange) CEO – Chief Executive Officer CR - Composite Reliability CVM – Comissão de Valores Mobiliários (Securities and Exchange Commission of Brazil) EFA - Exploratory Factor Analysis FBN – Family Business Network FDC – Fundação Dom Cabral GDP - Gross Domestic Product HR – Human Resource IBGC – Instituto Brasileiro de Governança Corporativa (Brazilian Institute of Corporate Governance) INDI - Integrated Institute of Industrial Development of Minas Gerais State IPO – Initial Public Offering PLS - Partial Least Squares UNCTAD - United Nations Conference on Trade and Development xiii CHAPTER I INTRODUCTION I.1. Background The concept and application of the principles of corporate governance have been consolidated in a more structured way from the end of the last century as a means of curbing opportunistic action of executives of large publicly held corporations, especially in Anglo-Saxon markets. In such companies, the shareholders have virtually no control of the company, which is effectively in charge of the executives. The natural consequence of the separation of ownership and control is that shareholders delegate to the executives the right to determine the fate of the company and the power to make day-to-day decisions. This situation favours the emergence of agency conflict which is a consequence of these individuals within a firm acting in the service of their own goals, at the expense of the goals of the firm’s owners. For Child and Rodrigues (2003) the role of corporate governance is precisely to minimize the agency conflict that arises from the clash of these different interests between ownership and control, ensuring alignment between shareholders and management. Much of the available literature related to the study of agency conflict is devoted to public companies, neglecting its effect of increased costs – called agency costs - in other ownership structures, especially those in family-owned companies. The apparent lack of separation of ownership and control in family-owned firms is responsible for most scholars not addressing this subject in such firms. However, some recent thinking has increasingly drawn attention to the fact that only the firstgeneration entrepreneurial firm with a sole founder would be exempt from agency conflict (Ang et al. 2000). Family businesses achieving higher levels of complexity can choose to hire nonfamliy managers, thus becoming exposed to agency conflict. But even if they do not contract outsiders, typical family-business agency conflicts may arise with the arrival of a new generation in management, where a family manager can develop conflicts of interest with other family owners. Some authors even claim that in these cases, the agency conflict is potentially greater than in public 1 companies (Dyer, 2006; Herrero, 2011; Morck & Yeung, 2004; Schulze et al 2001; Schulze et al, 2003b). The fact is that in the existing studies that deal with this theme, a lack of consensus is evident, indicating an opportunity for research on the reduction or increase of agency costs by family control (Bartholomeusz & Tanawski, 2006). Additionally it is important to understand that in Brazil good corporate governance practices are becoming increasingly important for raising company efficiency, for facilitating access to capital and avoiding the flight of investors, and for guaranteeing stakeholder confidence. Mallin (2004) defends the position that good practices of corporate governance are essential for emerging countries eager to attract foreign direct investments at more appropriate rates of return. Research carried out by the consultancy Mckinsey & Company (2002) showed that Brazil is among the main targets of investors interested in paying a premium for the shares of companies that practise good corporate governance. The Korn Ferry International and Mckinsey & Company study (2001) entitled ‘Panorama of Corporate Governance in Brazil’ showed that the perception of the tie between corporate governance and creation of value varies in accordance with the type of property. According to the study, the more family the company is, the more it is believed that corporate governance may contribute to the generation of value for it. This belief diminishes greatly, however, when control is shared between investors, or is in the hands of multinationals. In this way, this research becomes relevant to argue and understand the evolution of governance models and practices adopted by Brazilian family-owned companies, especially in one of its key contributions: how to manage and diminish their agency conflicts. I.2. Research motives This study gains in importance when one recognizes the present relevance of family businesses in the world scenario, and that this category of firm will remain an important economic influence in the future. Indeed, family-owned companies represent the predominant form of business undertaking worldwide. According to the 2 HSM Management magazine (2003), the organizations controlled and administered by families represent 80% of the business universe and their operations account for almost half of the world GDP - Gross Domestic Product. In the North-American economy, family-owned companies generate 50% of the GDP, are responsible for 60% of the jobs in the country and for 78% of the new jobs created, and moreover, represent around 35% of the companies listed by Fortune 500 (Perman, 2006, p.9). In Brazil the absolute majority of business undertakings are also family controlled, representing more than 90% of the total, according to Moreira Júnior and Altheman (2004). These enterprises are of extreme importance for the development of the country as they exercise great influence in the generation of jobs and participation in the GDP. A research carried out by the Brazilian National Bank for Economic and Social Development – BNDES (2000, cited in Machado, 2006, p. 28) shows that familyowned companies generate around two million jobs, contribute with a share in the GDP of 12% in the agribusiness segment, 34% in industry and 54% in services, and the majority of them are already second generation. In relation to micro and small companies, Pádua (2004) carried out studies with the objective of characterizing the profile of these companies in Brazil and revealed that 73% of the total are controlled and managed by families. Based on this data, the fundamental role of family-owned companies in economic development and the growth of societies is undeniable. Another fundamentally important point is the better understanding of the life spans of family-owned companies. Statistics show that many of these companies do not have a long life and there are few that survive generation changes. Independently of the country or of the culture only a small minority get to the second generation and far less to the third. In accordance with Gallo (1998) of every 100 family-owned companies, 30 pass through to the second generation and only 5 to the third. Therefore, facing this scenario, family controlled organizations have sought to equip themselves for survival and development in the midst of the challenges related to the generation transition, ensuring succession of the property and management, 3 and the creation of value for their shareholders. And it is in this that corporate governance consists, using a series of concepts that regulate the form of interaction between management, shareholders and the market, conferring greater transparency on the company and guaranteeing its lasting duration. What this research seeks is to propose a model that permits family-owned companies to identify the factors that generate agency conflict and through the adoption of best governance practices, achieve a healthier, more reliable business environment, and greater development for them. I.3. Research objectives The main objective of this thesis is to propose a theoretical model that allows us to identify the principal inducers of agency conflict in family businesses. Aware that in such companies, two distinct systems - the family and the business – do not necessarily interact harmoniously, the purpose of this study is to exploit each of the strengths and suggest structures and processes to neutralize the weaknesses created by the interface. This really matters when distinct family members play the roles of owners and managers, or when a nonfamliy manager plays an important role running the family business' destiny. In both cases, the agency conflicts are potentially relevant. Also, it is the intention to reach a better theoretical and empirical understanding as to how family-owned companies are governed, considering their governance structure, the existing corporate cohesion and alignment, the professionalization of management, the succession, the relationships between family and the company and vice versa, and the appearance of agency conflicts. This overall research goal is approached by addressing several more specific research questions, which are presented in Table 1.1. 4 TABLE 1.1: Specific research questions Number Questions 1 Who are the main actors in conflict in Brazilian family-owned companies? 2 Is the type of capital structuring of the company related to the level of conflict? 3 Does the size of the company influence the level of conflict? 4 Do companies with economic-financial performance better than the average of the segment to which they belong have a lower level of conflict? Does inferior performance imply a rise in it? 5 Do companies with nonfamily CEOs have a higher level of conflict? 6 Are the conflicts very different in a family-owned firm managed by the first generation owner-manager from those in a family-owned firm controlled by a second, third and subsequent generations and an extended family? 7 What about the tools and practices of management control: are they found to be in evidence in the Brazilian family business? 8 What is the real relevance of transparency and disclosure in family business and how can it be achieved? 9 Are the criteria for succession negotiated and made explicit in the Brazilian family business? And also, is there convergence on the natural process and the timing of its implementation? Is competence the most considered factor for hiring and promotion, regardless of the ties of consanguinity? 10 Is the Brazilian family-owned company basically managed by family members? Are they considered competent acting as managers? 11 Has the Board of Directors been a feature of governance adopted by Brazilian family businesses? Is its presence a factor in mitigating agency conflicts? What are its level of independence and efficiency? Source: Prepared by the author I.4. Contribution of the research This dissertation is based on the study of agency conflict in Brazilian familyowned companies addressing the relationship between the actors involved in the three key elements, viz property, family and management. In a family-owned 5 company, one or more members of a family exercise most of the administrative control because they are owners of most of the capital, there being a close relationship between property and control. Thus, it is essential to understand how family-owned companies are governed, focusing on various elements related to ownership, control and kinship and how they affect agency conflict. Family businesses provide a fertile research context for corporate governance studies due to their unique features and the fact that governance research in the area has been little explored. In fact, the volume of studies on family-owned companies related to this subject is still very small compared to the research into the governance of widely held firms, which is abundant. Previous research on family business has mostly addressed questions regarding how to improve family relationships and a large part of past studies has dealt with management succession. Daily and Dollinger (1992) have suggested some reasons for the lack of scientific investigation of corporate governance in family businesses. First, they believe that researchers base themselves on the idea initially presented by Berle and Means (1932) that the control of businesses rests basically in the hands of professional managers, not families. Second, it is difficult to study both family and business systems simultaneously because each belongs to a different scientific research branch. And finally, there is a common belief that work and family exist as distinct, independent systems. This research proceeds in the opposite direction to this belief, seeking precisely to understand more profoundly and systemically the connection between these elements and the balanced mechanisms to keep alignment among all the stakeholders. This dissertation intends to make three contributions to research on family business and corporate governance: 1st. First, to contribute to the small but emerging body of family firm research. 2nd. Second, to provide a new framework for the agency conflict theory, relating the elements of property/family/management to the reduction of tension caused by the diversity of interests. 6 3rd. Demonstrate how the costs of implementing family governance mechanisms are compensated by a reduction in agency costs, making it easier for family-owned companies to adopt them voluntarily. I.5. Structure of the thesis The diagram below shows the steps for the development of this study. Figure 1.1: Scheme representing the development of the structure of the thesis Source: Prepared by the author 7 I.6. Format of this document Chapter I provides an overview of the importance of family business and illustrates the gaps in the current literature concerning agency conflict in family businesses and how the objectives of this research will help to fill them. Chapter II includes a discussion of the literature regarding family-owned companies exploring their main attributes and peculiarities. Also there is a review of the literature related to corporate governance issues and the major implications of agency theory in family business. Chapter III presents the qualitative methodology used, combining grounded theory and the construction of case studies. Both methods and their application are presented. Chapter IV lists the detailed description of the four case studies conducted, highlighting the tensions via citations of the several interviews carried out. Chapter V presents a theoretical model built from observations of the four different case studies. Based on the findings of each case study, several hypotheses were proposed. Also, each construct from the grounded theory process was confronted with the literature in accordance with what other authors have proposed regarding its relevance, and mutual relationships with other constructs. Chapter VI presents the quantitative analysis, including the formation of the sample, the design and validation of the questionnaire, the survey application and the procedures for data analysis. Also, summarizes the hypotheses analysis, showing which are supported or not, and discusses the overall findings. Chapter VII offers the conclusions of the study, its limitations and suggestions for future research. It also discusses how this study makes original theoretical and practical contributions to knowledge. 8 CHAPTER II PRELIMINARY LITERATURE REVIEW II.1. Family-owned company II.1.1. Family business definitions As several studies indicate, the importance of family businesses has been continually documented by the academic literature since the early 90's (Lyman, 1991; Daily & Dollinger, 1992; Dunn, 1996; Fletcher, 2002; Rogoff & Heck, 2003; Ward, 2004; Zahra et. al., 2004; Chrisman et al., 2008; Milton, 2008; Zahra et al., 2008; Chua et al., 2009; Debicki et al., 2009; Steier et al., 2009). However, there is no consensus in the specialized literature regarding the limits of what constitutes a family business, presumably because there are a great number of definitions (Chrisman et al., 1999). This occurs essentially due to the fact that this issue has only recently become the focus of attention of researchers, as well as having been approached at different levels of analysis, such as individual, interpersonal / group, organizational, and societal (Low & MacMillan,1988; Basco & Rodríguez, 2009). Nevertheless, regardless of the diversity of conceptions the undeniable fact is the representativeness and economic importance of this type of business in major world economies, as pointed out by Wang et al (2007): In most economies, family business is estimated to represent over two-thirds of all enterprises and accounts for about half of the economic activity and private employment contributing to the Gross Domestic Product (GDP) (Wang et al, 2007, p. 173). To understand the characteristics that distinguish a family from a nonfamliy business, Mustakallio (2002) proposes distinct classification criteria: (i) ownership; (ii) family involvement in management and in the strategic process; (iii) generational transfer; (iv) family view of longevity and its intention to continue as a family business; (v) family goals; and (vi) interaction between the family and business. 9 i. Ownership A family business is defined by Barnes and Hershon (1994) as one whose control of property is in the hands of an individual (or individuals) of a single family. Donckels and Fröhlich (1991) recognize that a family business is that in which the family owns the most assets, while controlling its management. Aligned with this position, Litz (1995) defines a family firm as one in which both ownership and management are concentrated in the hands of one family unit. Neubauer and Lank (1998, p.21) describe ‘a family enterprise as a proprietorship, partnership, corporation or any form of business association where the voting control is in the hands of a given family’. For Lansberg et al. (1988) a firm is defined as family-owned when members of the same family have the legal control of its property. Taking as a starting point the family's influence on the property axis, Sciascia and Mazzola (2008) argue that family presence in ownership can become a benefit or a disadvantage. The authors assert that in this context of interaction, paradoxical conditions are created, which can only be understood by analysing the degree to which this level of family influence on the property can generate benefits to outweigh its risks. ii. Family involvement in management and in the strategic process For Gallo and Sveen (1991) what characterizes a family business is that its members have to take part in the management as well as having an important role as decision makers. In most first-generation businesses, with their low level of generational ownership dispersion, the founder alone has complete control of his/her firm (Lansberg, 1988; Harvey & Evans, 1994). Such domination of the control leads to greater relationship conflict due to the low level of family members’ participation in decision-making (Ibrahim et al., 2001). In contrast, in the presence of a high level of participative decision-making, conflicts should decrease due to the open communication and sense of involvement (Eddleston et al., 2008). For Bornholdt (2005), a family business is one in which the family bonds are crucial in the succession process and where the strategic positions of the company, 10 such as the Executive Board or Board of Directors are filled primarily by members of the family. Handler (1989a) adds that a family business is that in which most decisions related to the leadership succession process are influenced by family members present in the management or on the Board of Directors. Dyer’s research (1988) shows that most first generation family businesses have a “paternalistic” management culture, characterized by hierarchical relationships, top management control of power and authority, close supervision, and distrust of outsiders. On the other hand, descendent-controlled firms tend to utilize more professional forms of management (McConaughy & Phillips, 1999). Regarding strategy, first generation family businesses are a reflection of the aspirations and values of the founder-owner. Thus, the owner’s strategies are likely to influence the development path of the business, especially in its early stages (Basu, 2004). According to Davis (1983) even when the family is not part of the management of the company, if its members exercise a decisive influence on the business’ politics and direction, this is enough to characterize it as a family-owned company. In this sense, Chrisman et al. (1999) stress that it is not sufficient for family members to be owners and/or managers of the company. What makes family businesses unique is that the pattern of ownership, governance, management and succession concretely influences their objectives, strategy and structure and their way of formulating and implementing them. For them: The family business is a business governed and/or managed with the intention to shape and pursue the vision of the business held by a dominant coalition controlled by members of the same family or a small number of families in a manner that is potentially sustainable across generations of the family or families’ (Chrisman et al., 1999, p.25). These authors explain the dominant coalition as being the one that includes the main actors of an organization that fully control the organizational agenda. So, according to this definition and regardless of ownership control or participation of the family in management, it is sufficient that the family controls the dominant coalition for it to be considered a family business. 11 iii. Generational transfer To discuss what is a family business from the perspective of generational transfer, one must first understand what characterizes a first-generation family firm. For Yan and Sorenson (2006) it is one started or acquired by a family, which involves more than one family member, belonging only to the first and founding generation of the family. First-generation family businesses are highly dependent on the founders’ leadership, who are usually strongly involved in the management and who are the ones that bring the technical know-how (Lansberg, 1988). Also, founders establish the cultural patterns of family-owned firms according to their own personalities (Dyer, 1986). Chrisman et al. (2003), however, consider that ownership and management of the business by family members are not enough to define it as a family firm. Additionally, the intention to conduct the succession to the next generation must also be stated explicitly. Ward (1987) confirms that it is necessary to have the expectation of transferring control to the next generation, for a company to be characterized as a family business. The fact is that a growing number of studies have been focusing on the intergenerational aspects and succession as the essential characteristics that define a family business (Royer et al., 2008; Bigliardi & Dormio, 2009; Janjuha-Jivraj & Spence, 2009; Mitchell et al., 2009; Van der Merwe, 2009; Howorth et al., 2010). For Churchill and Hatten (1987), for a business to be considered a family business, the management transition, from an older to a younger member of the family, has to have happened or is still going to happen. However, Donnelley (1988) argues that a family business has to have lasted at least for two different generations under family control to be characterized as such. Chrisman et al. (1999) disagree with these latter authors about the theory that for a company be considered a family business, it has to have carried out one or two successions within the company. They think that the fact that parents are preparing their children to be professionals in a different area does not decharacterize the company as a family business. They support the idea that the definition of a family company may include those ‘family businesses owned by couples with young or no children, or those that pursue a 12 vision of the family business as a vehicle to educate the next generation for careers that exclude involvement in the business’ (Chrisman et al., 1999, p.25). iv. Family view on longevity and its intention to continue as a family business From this perspective, a business should be considered a family business when the members of the owning family ‘strive to achieve and/or maintain intraorganizational family-based relatedness’ (Litz, 1995, p. 103). Also for Chrisman et al. (2003) one of its peculiar features is the intention of maintain family control of the ruling coalition. Neubauer and Lank (1998), in turn, cite that, among the myriad of existing definitions about family business there has to be the intention that the family involvement shall continue to exist in the future. For Hoy and Verser (1994), despite all the difficulties that threaten the continuity of the family heading up the company, the owners of family businesses tend to have an avid aspiration for their businesses to exist and thrive over the long term. Therefore, it is often suggested that while first generation family-owned firms need to possess the special technical or business backgrounds necessary to start a business, subsequent generations need to be focused on maintaining and enhancing their business’ growth and success (McConaughy et al., 1998). Completing these several perspectives of family business, Donckels and Fröhlich’s (1991) conclude that the vision of longevity of the founders is that the business built by the family should continue under its control. And in order for this goal to be achieved, authors Fahed-Sreih and Djoundourian (2006) assert that the key to the longevity of a family business is conducting a planned and well-oriented succession process, allowing the change of control to occur in an orderly and nonconfrontational way. v. Family goals Donnelley (1988) identifies a company as a family business when there is mutual influence between its business policy (vision, values and goals) and the interests and goals of the family. Following this same line of thought, Distelberg and 13 Sorenson (2009) argue that to understand the family business it is necessary first of all to grasp the values and objectives that guide it. In the research of Miller and Le Breton-Miller (2007), the authors found that the goals of family businesses tend to rely on a long-term vision, in which the company is not perceived just as a financial tool for profit maximization, but also favours a win-win relationship with external stakeholders, especially with customers, suppliers and partners. In this sense, it is clear that It must take into account that family businesses have economic and non-economic goals (Olson et al., 2003), which makes it essential to understand from the very beginning, how each key individual of the family company defines success, and how much alignment exists around this understanding. Otherwise, distinctive conceptions of success or different pursued objectives, by these stakeholders, will certainly lead to conflicts (Astrachan & McMillan, 2003). vi. Interaction between the family and business Davis (1983) advocates that it is exactly the kind of interaction between family and business that characterizes the nature of a family business and defines its uniqueness. Chrisman et al. (2003) share this idea saying that what distinguishes a family business from other types of business is the involvement and interaction of the family with the business which results in unique, inseparable, and synergistic skills and abilities of the company. Recent studies, such as that of Nordqvist (2012), present the argument that members of the family business play a key role in developing the strategy of family businesses. Thus, the family interaction would be responsible for fostering a dynamic process that results in the strategic guidelines rooted in family values and the sociopsychological dimensions of the owners (Nordqvist, 2012). Furthermore, for this author, the overlapping of values and the distinctive capabilities of family businesses would allow the development of unique competitive advantages. In fact, the interaction between family and business, besides characterizing the family business, can result in positive aspects for the enterprise. However, it is 14 worthwhile to attend to the risks and difficulties that are inherent in this business model, as pointed out by Martínez et al. (2007, p. 85): Among these advantages or strengths we can identify a closer monitoring, an essentially long run perspective, higher unity of goals among shareholders, quicker decision making, stronger culture that come from family values, and so forth. On the other side, some of those weaknesses or disadvantages are related to the overlap of family and business issues, lack of succession planning, nepotism, resistance to change and to professionalization, agency issues, and others. Adopted definition For the purpose of this study and given these various definitions of family business, the one proposed by IBGC - Brazilian Institute of Corporate Governance has been chosen, which is that: The family business is one whose equity control (most of the percentage of voting shares) is in the possession of an individual family or group of families. Summarizing, companies can be classified primarily by family ownership and not by family management (2009). II.1.2. Family business characteristics To present the main characteristics which distinguish a family company, the criteria of Hollander and Elman (1988) which identify four distinct approaches were chosen: first, the rational; second, the one which focuses on the founder; third, the one which emphasizes the phases and stages of growth and the last, the systemic approach. i. Rational approach For the authors who share this thinking, one of the main distinguishing features of family businesses is the fact that feelings rather than logic control the decision-making process. Danes et al. (1999) define the company as being the rational arena and its family members as being the emotional. The family constitutes the core of the business and family members are tied by strong emotional bonds. Due to the fact that family and business are so intertwined by strong ties (Davis, 15 1983), it is virtually inevitable that emotions pass beyond the family boundaries and reach the business (Daily & Dollinger, 1992). The discussion about the kind of rationality prevailing in the family business has produced many studies that tried to attain a deeper understanding of the emotional ties that involve family members and business, as those of Zellweger and Astrachan (2008) who tried to examine how much the family owners valued subjectively their business. The work of Astrachan and Jaskiewicz (2008) that measured the costs and returns in the emotional context of family businesses can also be cited in this regard. ii. Focus on the founder Hollander and Elman (1988) affirm that much of the literature on family businesses has been concentrated on the figure of the founder as entrepreneur and main influence on the company’s destiny. The authors point to the conflicts that naturally arise in the growth process of the family business and the habitual inflexible nature of the founder. Therefore, they consider that the founder's personality and style can help to predict obstacles in the transition process that the company may have to surmount. Normally the founder brings tendencies, values and rules to his/her management, which actually are the results of his/her experiences and personal life history. Thus, the style of the founder does influence the operational culture of the company, affecting its development and the way it reacts to changes. The founder usually assumes multiple roles while the business grows, and involves him/herself personally, intervening at all organizational levels (Hollander & Elman, 1988). The difficulties begin to emerge when the company grows and it needs a more professional management, requiring more planning and delegation. New hiring practices, evaluation and promotion of personnel and management development must be implemented to allow the company to outlive its founder (Ward, 1987). In many cases the resistance of founders to hiring professional executives, or even to passing the business over to their successors stems from the concern to protect their legacy (Janjuha-Jivraj & Spence, 2009). Thus, even if unconsciously, it 16 is common for them to adopt a manner of speech that tends to underestimate the ability of their successors. To Janjuha-Jivraj and Spence (2009) this would be a way that founders show themselves to be crucial to the survival of the business, and putting off their retirement as long as possible. iii. Phases and stages of growth This is another important area of research that seeks to identify alternating periods of stability and transition that occur in response to changes and needs of the company, of the family and of the employees, members or not of the family. One of the more traditional studies in this field was done by Hershon (1975, cited in Hollander & Elman, 1988) which identifies three management standards that correspond to three different generations of the founding family. The first area is characterized by paternal and entrepreneurial management - typically a company's first generation. The second pattern occurs in the second generation companies characterized by a brotherlike management. And finally, the third pattern is one where collective management or family network prevails, run by brothers, cousins and other members of the third generation of the founding family. iv. Systemic approach Distelberg and Sorenson (2009) show that from the beginning of the 1980s, the area of family business studies has come to recognize the importance of theories based on a systemic view for understanding the complex universe of these companies. The school of thought developed within this perspective have contributed significantly to broaden the perception of family businesses as highly complex systems, developing a deeper understanding of the interaction between the family and business. Tagiuri and Davis (1996) proposed a referential framework to represent such interactions based on a three-circle model that represents three independent but overlapping subsystems: business, ownership, and family. An individual who is part of the family company system can position himself in each one of the seven 17 subdivisions that are formed by the overlapping circles of the subsystems, as presented in Figure 2.1. Figure 2.1: Three-circle-model Source: Tagiuri and Davis, 1996, p. 200. In this model, these ‘three subsystems are taken into account to explain how individuals in family business make decisions or develop strategies that fulfill the goals of each subsystem and the whole family business’ (Parker, 2004, p. 56). Along this line of reasoning, Davis (1983) explains the idea of a joint system, which operates according to the practices built up from the needs on each side, family and business, to meet the needs of the whole. In this sense, Basco and Rodríguez (2009) found empirically that family businesses that consider the family and the business as parts of a whole system have better results when compared to those that seek to limit the governance only to the business area. This possibility of linkage between business and family was crafted by Beckhard and Dyer (1983) who propose mechanisms of articulation between the business and family, for example, a board of directors. Davis and Stern (1988) address the need for creating processes and social structures that can deal with family problems in a proper context, separating them from the business issues. Mustakallio (2002) points out, that such structures do not have a defined format, since they are not required by law, but that the forum seen to be more common in family-owned firms is the family council, which 18 consists of a group of family members that periodically meets to discuss the family relationship with the business, among other things. Hollander and Elman (1988) state a summary that some authors see family issues as elements external to the business, while others understand them as interlinked systems: two subunits, family and business, which together comprise the family business system. For the authors, the fact is that family processes and business processes are in a relationship of interdependency and interactivity. Therefore, to Olson et al. (2003) the key to the success of family businesses is not individually working family and business systems, but understanding this interdependence and effectively managing their superposition avoiding potential conflicts. Typology of family firms What can be learnt based on a more detailed analysis of the so-called family firms is that they do not form such a homogeneous group. For one to carry out any study in family owned firms, Gandra (1999) recommends as necessary and useful a division of them into two groups: the first corresponding to small and medium sized privately held companies with centralized control, and the second consisting of larger family companies, public companies and with decentralized control. II.1.3. Family business conflicts As is evident, the intense interaction between family and business, social institutions that are driven by different objectives, ends up making a family business a fertile ground for the emergence of conflicts. While the company's purpose is to produce products and services for its customers generating profit for their shareholders, the family mission is to care and provide security for its members (Mustakallio, 2002). Chrisman et al. (1997) and Davis (1983) point to the risk that by adopting the prioritization of non-monetary goals such as independence, prestige and job creation for family, the family company might reduce its efficiency and performance. In this sense, Sorenson (1999) points out the complicating factor of the presence of family members in the company when they start to be more concerned 19 with their own satisfaction and their level of participation in the company than with the actual business performance. Donnelley (1988) reinforces this idea, heeding to the fact that the company's needs can be neglected at the expense of family needs. And the trend is that in seeking to maintain good family relationships, family businesses end up focusing on issues important to immediate family and even those related to close relatives. This is particularly true when parents are involved more directly with the business, whether as owners, managers, employees or non-participating shareholders (Sorenson, 1999). Family businesses are then to be hostages of psychodynamic factors such as: sibling rivalry; lack of harmony between the couple; parent-child conflicts; and symptoms such as nepotism, role ambiguity, dispersion of ownership among family members; low competence in management and divergences between family members and nonfamliy employees. (Hollander & Elman, 1988; Dyer, 1994; Schulze et al., 2001, 2003b; Martínez et al. 2007). Danes et al. (1999) cluster the family business conflicts in five distinct blocks: (i) justice conflict (concerns problems of compensation and quality of treatment along with allocation of resources); (ii) role conflict (confusion among roles when family members work together or when the family business employs others who are not part of the family); (iii) work/family conflict (those that surface at the intersection of the family and the business subsystems); (iv) identity conflict (involves family members’ need to differentiate themselves from family expectations and act as independent, autonomous people); (v) succession conflict (the effect of family business succession on harmony in the family business). Jehn (1997) identifies three different natures of conflicts in family-owned firms: task, process and relationship conflicts. Task conflict is about the discussion of goals and strategies. For the author, moderate levels of task conflict have been shown to be beneficial to the performance of top-management teams. Firms with very high levels of task conflict tend to have problems completing tasks and reaching goals, while firms with very low levels of task conflict often become stagnant and lack the development of new strategies. Process conflict is characterized by disagreement on how work should be accomplished and how members should be utilized and how much responsibility individuals receive. Family-owned firms with low levels of process conflict may have problems adequately adjusting family members’ responsibilities 20 and firm resources, while those with high levels of process conflict may suffer from role ambiguity, rivalry, and uncertainty (Jehn, 1997; Jehn and Mannix, 2001). Relationship conflict is the perception of personal hostility and incompatibility which includes affective components like annoyance, disappointment, and anger at others (Jehn and Mannix, 2001). It results in negative feelings leading to suspicion and resentment at the same time goodwill and mutual understanding decrease among individuals, interfering ultimately with work efforts (Jehn, 1997). On the other hand, Harvey and Evans (1994) categorize the conflicts in familyowned firms from the distinct fields they can originated in: internal to the organization, external to the family or families involved in the organization, and external to stakeholders (bankers, investors, suppliers, distributors, and members of the board of directors). Based on the three-circle model of Tagiuri and Davis (1996) these authors proposed three levels of conflict that can occur in the family business, which are illustrated in Figure 2.2. Figure 2.2: Three levels of individual and collective conflict in the entrepreneurial environment Source: Harvey and Evans, 1994, p. 335 21 Level 1 conflict occurs when there is no interaction among the three entities and the conflict does not spill over to the other constituents. Level 2 conflict involves those occurring in two of the entities as they overlap, creating a more complex form of conflict; the conflict sources may be different, but when they are combined, the conflict is intense and more difficult to address. Level 3 conflict encompasses all of the interested parties in the family business, that is, the business organization, family, and external stakeholders. In summary, apart from the adopted criterion to categorize the conflicts in family-owned firms, there is a point of view common to all the researchers: how critical are the challenges that exist in managing conflicts for the firm’s survival (Dyer, 1986; Sorenson, 1999; Ward, 1987). It is noteworthy that the issue of conflict in family businesses still represents a major subject in the production of recent academic literature in this area, either directly or indirectly linked to other issues. Table 2.1 gives a few examples. In this sense, it is one of the goals of this dissertation, to understand how corporate governance can be a helpful tool in managing such conflicts. Table 2.1: Conflict in Family Business Studies 22 Source: Prepared by the author II.2. Corporate Governance and Agency Theory II.2.1. The real meaning of Corporate Governance If the 19th was considered the century of the entrepreneur, witnessing the emergence of the modern corporation, and the 20 th was the century of management, the 21st century promises to be, according to Tricker (2000), the century of corporate governance, to the extent that the theme has become one of the key issues in the management and regulation of companies. Corporate governance began to gain importance, especially in countries with more advanced capital markets like the U.S. and England, from the 1980s, basically because of three sets of problems: failures in relations between shareholders and corporations; the existence of boards of directors that were only collegiate pro forma 23 and ineffective in monitoring the interests of owners; and performance of managers, whose interests often conflicted with those of shareholders, whether controlling or minority (Andrade & Rossetti, 2011). These problems eventually create opportunities for the appearance of abuse of power, fraud and strategic mistakes that were made in companies of worldwide recognition. In response to pressure from influential groups that called for changes in how public companies were being managed, several codes of best practices in corporate governance came out around the world. In the United Kingdom, the key document was the report by a committee chaired by Sir Adrian Cadbury, who emphasized the importance of the presence the non-executive directors, audit committees, disclosure of director’s compensation, disclosure of accounting statements, internal control systems, and the role of the auditor (Bartholomeusz & Tanewski, 2006). According to Tricker (2000), the Cadbury Report, published in 1992, was notable for its influence on the thought of corporate governance around the world. Similar initiatives multiplied in the following years until, in 1999, the Organisation for Economic Co-operation and Development (OECD), which brings together the thirty most developed countries of the industrialized world, published its influential ‘Principles of Corporate Governance’ that became an inspiration for existing codes of corporate governance in several countries (Andrade & Rossetti, 2006). In the US, corporate governance has become increasingly important after the adoption of the Sarbanes-Oxley Act of 2002, a legislative response to the ‘dot-com’ and ‘telco’ speculative bubbles at the end of the 1990s: ‘Provoked by the collapse of Enron (which used complex off-balance sheet vehicles to hide extraordinary losses leading to a write-down in shareholder funds of US$1.2 billion) and Worldcom (which used accounting improprieties to inflate reported earnings and cash flows by US$3.9 billion) the Securities and Exchange Commission (SEC) response, while similar in theme to earlier reforms, differs fundamentally in that the requirements are mandatory’ (Bartholomeusz & Tanewski, 2006, p. 250). Monks and Minow (2004, p.329) refer to this new regulation as ‘the most significant piece of federal legislation concerning public corporations since the post1929 stock market crash legislation creating the SEC’. The Sarbanes-Oxley Act imposed rules for the purpose of increasing the independence of directors and auditors, in order to align the behaviour of executives with shareholders' interests, by 24 ensuring greater transparency and accountability and reviewing the structure and the effective participation of the boards of directors (Bhagat et al., 2008; Bebchuk et al., 2009; Linck et al., 2009). In consequence, the corporate governance literature has been largely concentrated on the relationship between shareholders, board members and top management, characterizing an ‘outsider’ view, where owners depend on external elements to guarantee that managers are acting in their best interest (Monks & Minow, 2004). Shleifer and Vishny (1997) point out as an external element the legal system that defines property rights and sets boundaries within which the companies must operate. Kaplan (1997, cited in Mustakallio, 2002, p.31) indicates also the efficiency of capital markets as an important issue that impacts the way corporate governance is shaped in different countries. In this way, to promote strong capital markets ensuring access to global capital, countries, principally developing countries, need to align their companies and laws with international standards (Tricker, 2000). For Mallin (2004), however, corporate governance is related as much to external, as to internal aspects, such as internal control. From her perspective, corporate governance becomes an essential instrument for reaching corporate objectives. Thus, both from the external as the internal point of view, corporate governance can be defined as a set of arrangements, procedures, regulations and laws that allows investors (shareholders) to monitor the activities of the executives in companies where they have resources applied. It can also be defined as a set of values, principles and rules that regulate the power and control system, as well as the company management mechanisms, ensuring that they meet the needs of a given set of private and social interests. For Andrade and Rossetti (2006) the essential factor that compelled the development of good governance was the dispersion of company ownership and the consequent progressive distancing between the shareholders and executive management. According to the authors: ‘The multiple partitioning of the property of large companies was due both to succession processes which led to the repeated division, with each generation, of the accumulated patrimony of the founders among dozens of heirs, as also to the expansion project financing of the business in the capital market … the publicly traded company led to the appearance of the most 25 varied forms of opportunism within them, from the mildly offensive nepotism to the most varied and aggressive forms of misappropriation to the detriment of the shareholders’ interests and rights’. (Andrade & Rossetti, 2006, p.1). Tricker (2000) confirmed this idea stating that governance appears whenever the ownership of a company is separated from its management: it concerns therefore, essentially, the exercise of power. In this way, the corporation has become defined, in part, by this separation between ownership and control and also by a control that is being exercised through a small participation in ownership. Consequently, ownership without control and control without ownership seem to be the logical result of the development of the corporate world (Means, 1931). For Means (1931), the essential characteristic of ownership seems to be the possession of interests in a determined company while the essential characteristic of control consists in having power over the company. With the development of the modern corporation, ownership interest and power have become attributes of distinct groups. A group of individuals (shareholders) has ownership interests while a second group has power. The power of the second group becomes effective through the management of the corporation and, in particular, through the Board of Directors. However, for many authors, the interests of shareholders can only be effectively protected if Board of Directors has the necessary autonomy with the inclusion of independent members (Dalton et al., 2008; Fogel & Geier, 2007; Gordon, 2007). Thus, the group that has the power to select the members of the Board and/or give the management directives is the group that effectively has control. Means (1931) also identified five types of control: (i) control through almost complete ownership, that is, where only one individual or a small group of associates is the owner of almost all the shares; (ii) majority control, in which an individual or group retains the majority ownership of the shares; (iii) control through legal device without majority ownership, that is, by the creation of holding companies, that allow an individual or group to have control, although holding only a small fraction of the company; (iv) minority control occurs when an individual or group possesses a minority holding, but can maintain control by attracting other minority holdings; (v) management control, which exists when the ownership is so disperse that no individual or group can dominate the company. While in the first case, separation between ownership and 26 control is purely incipient, in the last case, it is complete. The shareholders have almost no control over the company while those that detain effective control have merely a tiny proportion of the ownership. Thus, the degree of dispersion of ownership varies from closely held corporations, as family-owned companies normally are, where owners are often key managers, to very widely held corporations where individual owners have only a small fraction of the common stock. In the last case, a logical consequence of ownership separation is that the shareholders delegate the executives the right to determine the company’s overall direction and the power to control its everyday decisions. The issue that arises is whether the managers would appropriate resources for their own benefit, at shareholders’ expense, or run the company properly in the shareholders’ best interest. To Berle and Means (1932) the managers of such firms still pursue their own interests and neglect shareholders. To this divergence of interests between those who have the ownership to those who execute the control, the name agency conflict is given. II.2.2. Agency Theory within the Corporate Governance studies The concept of agency theory has a long history in literature and appears as an object of study in several disciplines, e.g., law, political science and sociology, but it is in economics and finance where its main applications are founded in the areas of welfare economics, internal firm organization, theory of the firm, and contractual relationships (Mitnick, 2006). From the economic perspective, the study of agency theory started with the work of Berhold (1971), Ross (1973) and especially Jensen and Meckling (1976). Ross (1973) proposed the concept of agency conflict by means of a mathematical formulation which indicated the existence of divergent interests between the doer, called 'agent', and the owner, identified as 'principal', the latter being the subject who is affected by the actions of the agent. Given the existence of such misalignment of interests, Berhold (1971) justifies the need for the principal to motivate the agents through monetary incentives, in the 27 sense of making decisions in accordance with his/her guidelines. For Jensen (1994) such an argument makes sense since even a self-interested person will always have an incentive to reduce losses produced by the conflicts of interest, if at the end, he/she can share the rewards. These incentives, in order to engage the agent, are identified by Jensen and Meckling (1976) as generators of agency costs. These authors coined this term not only in relation to the cost of incentives, but also to those incurred in policing and monitoring the agents, such as auditing, formal systems of control and budget constraints. Chrisman et al. (2004) confirm the cost of incentives and monitoring of managers, as agency costs and complement the list with cost sanctions, forging links and developing management processes for the alignment of interests. Monitoring and alignment costs are necessary because principals tend to be afraid that their agents act opportunistically, shirking, holding-up, cheating, stealing, self-dealing, adopting a free riding attitude, corrupting themselves, as well as seizing perquisites for their own use. (Perrow, 1986; Shapiro, 2005; Schulze et al. 2003a). One way to avoid the existence of opportunistic behaviour of the agent would be conceiving contracts that would impose penalties when deviations occur. For Jensen and Meckling (1976) the degree of deviation depends, among other things, on the rules which were agreed and the sophistication of the contracts drawn up. If the information was perfect and people had unlimited mental capacity, the principal and his/her agents could produce perfect contracts that anticipated all the actions and reactions of the agents (Williamson, 1975). It is known, however, that it is impossible to produce complete contracts because of the large number of possible contingencies in the relationship between agent and principal; the multiplicity of possible responses to them; the fact that you cannot predict how and when these contingencies will occur; or because, even if it were possible to determine them, it would be extremely costly to specify all of them in a contract (Klein, 1983). Fama and Jensen (1983b) call attention to the exact composition and enforcement costs of the contracts, saying that they cannot exceed the benefits in reducing agency costs. Jensen and Meckling (1976) state, as a summary, that the agency relationships include, therefore, contracts, incentives, monitoring and bonding 28 instruments and other forms of social control adopted to minimize the agency costs, which are constituent elements of the contract. Young et al. (2008) emphasize the existence of various governance mechanisms aimed to minimize the agency conflict between managers and shareholders. The authors also argue that these mechanisms can be internal, such as boards of directors, ownership concentration, compensation packages to executives, or external, such as market competition, labour laws, codes of governance and the threat of hostile takeover. According to Bathala and Rao (1995), agency conflicts between principals and agents tend to be more clearly stated in large publicly held corporations, where top management is responsible for suggesting and implementing the main policy initiatives, while equity holders, who are largely diffused, assume the main part of the risk associated with those decisions. As management does not suffer a substantial portion of the wealth effects of their decisions, agency problem arises. In their research Ang et al. (2000) find that agency costs are really higher in companies that are not 100 percent owned by managers, and these costs increase as the equity share of the owner-manager declines. This finding confirms the prediction of Jensen and Meckling (1976) that an important source of agency conflict is related to the tendency that when the manager's right to property decreases, this also reduces his/her incentive to undertake significant effort in innovative activities, such as seeking new profitable ventures. However, Daily et al. (2003) show that the tendency of contemporary companies is to be increasingly characterized by the small ownership holding of the executives that manage them. Jensen and Meckling (1976) thus raise the question that if companies with diffuse ownership are more susceptible to agency costs, why is it the prevalent model today and why are millions of people willing to invest in companies run by professionals so little interested in their welfare? Because it is a safe investment. In 1932, Berle and Means published a study on ownership and control, considered the first emergent work on corporate governance, called ‘The Modern Corporation and Private Property’. In this work, the authors called attention to the fact that the existence of a corporation creates a system that exposes investors to risk no greater than their investment in a firm. Unlike a partnership, where all partners 29 are liable for all kinds of legal action even after the company is bankrupted, investors have no such risk. The reason is quite understandable: no one would buy stocks in a large corporation if the risk of losses were unlimited. The basis to invest in the stock market is to know that whatever happens, the risk of loss is limited to the amount of the investment (Monks & Minow, 2004). This rule allows corporations to make themselves capable of raising funds for the expansion of their business and financing their internal operations, not only through bank loans, from financiers that charge interest for the use of their resources, but also by other forms of raising capital, such as the sale of shares and bonuses for individual investors (Blair, 1995). In negotiating their shares in the public market, corporations promote the dispersion of ownership, which brings in its train the important consequence of the distancing of shareholders from managers (Tricker, 2000). The increasing number of companies around the world that are becoming public and the consequent risks of agency conflict arising from the separation of ownership has led the study of corporate governance in public companies to dominate most of the researches in this field of knowledge. Thus, the effect of agency cost in other ownership structures, especially family-controlled, has been little explored. And the studies that have been undertaken show a lack of consensus regarding the possibility of agency conflict being or not mitigated in family businesses. The analysis of whether increased incentives from the controlling family reduces or not agency costs still deserves a systematic investigation (Bartholomeusz & Tanawski, 2006). It is worth considering the fact that in developing countries like Brazil, it appears that the main agency conflict occurs between majority and minority shareholders (La Porta et al., 1999, Young et al., 2008) This shift in relation to the traditional model of Jensen and Meckling (1976) is explained by differences between corporations from emerging countries compared to those from developed economies, such as high ownership concentration (especially in family businesses), little institutional protection for minority shareholders, and low effectiveness of governance mechanisms (La Porta et al. 1997, Young et al., 2008). 30 Taking into account that to Miller and Le Breton-Miller (2006), the agency conflicts in family-owned firms will differ in degree depending on the exact structure of governance prevailing, further study of agency theory in family businesses becomes essential. II.2.3. Agency theory applied to a family business One line of thought in the literature supports the idea that family businesses are free of agency conflicts, especially those in the first generation which are managed directly by their owner. Jensen and Meckling (1976) identify them as zero agency-cost firms. According to Anderson and Reeb (2003) the family’s large ownership stake provides a strong economic incentive to maximize firm value. For Blanco-Mazagatos et al. (2007, p. 201) ‘in this stage, the family is usually nuclear, with closed relationships that minimize information asymmetries between family members and lead them to maintain concentric objectives’. For Pollak (1985) the cooperative attitudes adopted by the family members will be influenced by the trust and altruism generated by strong kinship. Altruism, in its turn, promotes a governance system that is particularly efficient in the first-generation entrepreneurial firm (Lubatkin et al., 2005). For Pollak (1985), even those family businesses whose ownership is dispersed among other family members, governance costs are minimized by having alignment on growth and risk. This alignment of interests would be responsible for reducing the tendency to opportunistic behaviour, avoiding the need for formal control systems and incentive payments, and promoting communication, cooperation and consensus (Jensen & Meckling, 1976). Bartholomeusz and Tanewski (2006) highlight the fact that the welfare of the family being directly tied to the welfare of the company makes it an incentive to reduce agency costs. Fama and Jensen (1983b) share the same thought by suggesting that family-owned firms tend to reduce agency costs in the presence of healthy family relationships, affecting the company positively. Daily and Dollinger 31 (1992) see, even, the family business as the most efficient form of organization, precisely because of the absence of agency costs. Herrero (2011), in turn has no doubts about the existence of agency conflicts in family businesses, although there is still no consensus on whether such conflicts are more frequent in family or nonfamliy businesses. There does exist a group of authors arguing that agency conflicts within family businesses are more significant, among them, Schulze et al. argue that ‘the dispersion of ownership in family-held firms drives a wedge between the interests of those who lead a firm - and often own a controlling interest - and other family owners’ (2003b, p. 181). For Dyer (2006), family businesses face significant agency costs due to conflicts caused by very different types of family involvement in business, which lead them to defend their own interests. La Porta et al. (1999) agree that in this kind of organization, the interests of the controlling family are often placed above the interests of shareholders and other stakeholders of the company. To Morck et al. (1988) and Young et al. (2008) the majority controllers tend to expropriate wealth from minority shareholders through excessive compensation, related party transactions and special distribution of dividends. In addition, family businesses would not be free to face the same reality as public companies, with managers focused on short-term results, showing aversion to risk and innovation (Davis & Harveston, 1998). For Schulze et al (2001) the fact that family businesses are free of the control imposed by the market, increases the risk of agency based on self-control. In this sense, Morck and Yeung (2004) show that the problems of governance in family businesses are worse than a public company. This is because, when such companies are poorly managed, the value of their share decline, inducing shareholders to open legal proceedings against the managers, and leading them to consider hostile takeovers and disputes in the general assembly. Moe (1984) adds that shareholders can withdraw their capital invested in companies whose managers fail to deliver the expected results. Another possibility, to minimize problems of shirking, would be to delegate the functions of control and monitoring to boards that have information and resources to promote the realignment of managers, or, as a last resort, dismiss them. Some studies indicate, in fact, that there is a greater tendency for shirking in family businesses since in some cases managers are chosen because 32 of family ties rather than their competence, resulting in poor performance and the creation of barriers to prevent dismissal (Herrero, 2011). In family businesses, specifically, replacement mechanisms of managers are contaminated by other factors beyond their ability, or lack of it, for their roles. When managers cannot be easily dismissed, Morck et al. (1988) explain that the company is experiencing a conflict of agency due to entrenchment. This type of situation is particularly found in family businesses where a family member can remain in an executive position, although not being sufficiently competent or qualified (Sciascia & Mazzola, 2008; Herrero, 2011). These authors demonstrated empirically that entrenchment reduces the value of the company. Hendry (2002) identified this phenomenon as 'honest incompetence' when the family agent does not act motivated by opportunism, but even so the conflict arises due to his/her inability to deliver the results that the company needs for a successful trajectory. Another aspect defended by Howorth et al. (2004) is that agency theory may provide only a partial explanation of the dynamics of family-owned firms. For the authors, when dealing with issues related to ownership and management, agency theory ends up focusing only on dimensions of a performance-based system whose primary objective is to maximize shareholder value. Also for Chrisman et al. (2004), imagining that the owners of family businesses consider only economic objectives of a company is a partial and limited view. They assert that many actions considered to generate agency costs in nonfamliy businesses may not correspond to the reality in family businesses that do not have financial goals as the main target. The controllers can see the family business as a place to shelter the family members, to generate family cohesion or even to act towards philanthropic or environmental preservation ends (Chrisman et al., 2005), even if it means a financial performance at a nonmaximized level (Chrisman et al., 2004). So if a family decides to employ its less skilled members, even with the future consequences of a loss of company value, this would be considered an agency conflict in nonfamliy businesses, but not in a family business. Thus, for a correct analysis of agency conflicts in family businesses, it is necessary to understand the interests of its owners, their decisions and actions, and 33 identify the activities, incentives, policies and structures that they introduce to prevent their interests being contested. (Chrisman et al., 2005). In this sense Young et al. (2008) emphasize that family relationships make the analysis of agency conflicts more complex in those organizations that employ family members as managers, since the relationship between the principal (family owners) and agents (familymember managers) will be based on emotions, feelings and informal connections, resulting in a less intense monitoring of the latter. II.2.4. Criticisms of agency theory Agency theory supplies a strong framework for the analysis of organizations and has been used in the majority of empirical studies. However, its use is accompanied by dissensions and it has been criticized by several authors who disagree with the stereotype of the psychologically egoistic agent, who only acts in his/her own interest, who is capable of lying, cheating, stealing and shirking (Shankman, 1999). Even the idea of a purely mercenary relationship, where the incentives would be sufficient to align the agents with the objectives of the principals, is contested by Donaldson and Davis (1991), who maintain that not all agents are motivated solely by financial factors, but those that seek the intrinsic satisfaction in doing a challenging job, exercising their authority and responsibility and, thus, gaining recognition from their superiors, peers and subordinates, also have to be taken into consideration. This vision, of a managerial motivation in opposition to the agency theory was identified as stewardship. Stewardship theory claims that organizational managers' main motivations are to serve the organization's best interests and mission, in opposition to the self-serving and opportunistic stimulus proposed by agency theory (Davis et al., 1997). According to it, the manager seeks essentially to do a good job and be a good steward of the corporate assets (Donaldson & Davis, 1991). For Davis et al. (1997) the reason that the steward adopts a pro-organizational and collectivistic attitude, is related to the fact that ‘by so doing, the steward's utility functions are maximized’ (1997, p.25). In the context of family businesses several recent studies 34 argue that stewardship theory offers a different and complementary understanding of the purpose of these organizations showing that the family owners and managers are not driven solely by economic interests but also act altruistically to the benefit of stakeholders, guided by their family values (Miller & Le Breton-Miller, 2006;. Prencipe et al., 2008; Granata & Chirico, 2010; Howorth et al., 2010; Scholes et al., 2010). For Pieper et al. (2008) two factors encourage stewardship behaviour: intrinsic motivation and organizational identification and commitment. Managers who are selfactualizing base their conduct on intangible needs or have high degrees of selfleadership. Those with high levels of organizational or value commitment believe in and accept the goals of the organization with, consequently, less goal conflict with that organization (Mayer & Schoorman 1992). Meyer and Allen (1991) and Riketta (2002) identified a correlation of commitment with satisfaction, intensity and duration of the work undertaken, as well as with the motivation to contribute with an additional effort to the interests of the collectivity. Consistent with this view, it is expected that within family-owned companies, the family member’s commitment to their firm leads to a stewardship behaviour supporting the organization’s mission and goals (Miller & Le Breton-Miller, 2006). But this not always the case. Corbetta and Salvato (2004) propose when each framework in the study of organizational behaviour, agency or steward, is more appropriate than the other: When financial goals prevail in a family, family members’ motivation to operate in the family firm will be based on lower order needs and extrinsic factors, thus favouring the emergence of agency relationships. On the contrary, when non-financial goals prevail, this will foster motivation based on higher-order needs and intrinsic factors, thus favouring steward-principal relationships (2004, p. 357). For these authors the values of the owning family determine the structure of the model of man that will dominate in its organization from the second generation ahead. When the model favoured is that of the self-serving, economically rational model of man, the agency relationships tend to prevail, especially among brothers and cousins, and those family members who act as employees/agents will tend to use all their discretionary power in their own interest, and to the disadvantage of the 35 other shareholders/principals. On the contrary, when the model of man chosen by the owning family is the self-actualizing man, steward relationships may prevail. Miller and Le Breton-Miller (2006, p. 77) corroborate this view, by stating that Family management [...] can reduce agency costs and increase attitudes of stewardship, thereby extending investment time horizons and building firm capabilities. But left unchecked, family management can be dangerous, promoting leadership irresponsibility, expropriation from minority shareholders, hubris, and excessive risk taking—in short, the antithesis of good stewardship. Another source of criticism of agency theory is stakeholder theory. This disputes the emphasis on the maximization of shareholder wealth, based on the management of the principal/agent relationship, maintaining that the company should assume multiple objectives, which take into account the needs of all stakeholders (Donaldson & Preston, 1995; Ayuso et al., 2007; Laplume et al., 2008; Zellweger and Nason, 2008; García-Castro et al., 2011). It should be pointed out that stakeholders are all ‘persons or groups that have or claim to have ownership, rights or interests in a corporation and in its activities past, present and future’ (Clarkson, 1995, p.5). Included in this definition, therefore, are shareholders, their representatives on the boards, employees, suppliers, customers, government and other groups whose relationship to the corporation was primarily non-economic. Freeman (1984), considered the first author to discuss stakeholders’ theory explicitly, proposes a normative model in which the managers would have fiduciary duties not only towards the stockholders, but also the other stakeholders. McLaren (2004), however, recognises the difficulty of implementing such a system because as a rule companies give priority to the search for shareholder value, concerning themselves less with the generation of value for other stakeholders. The author maintains that the interests of the shareholders should be balanced against the interests of the other stakeholders because both are equally important to long term company sustainability and also because people have sets of interests that are not divisible. The adoption of this model represents the recognition that the inclusion of the perspective of publics other than the shareholders is relevant to the long term survival of the organization (Andrade & Rosseti, 2011). The legitimization of claims 36 arising from viewpoints other than the merely economic, and the review of ethical and moral principles that regulate business relationships constitute today relevant matters on the agenda of the post-industrial company. For Quinn and Jones (1995), such principles are regulators of market efficiency, and are therefore more important than any agency relation. The authors stress the following moral rules that should be observed: (i) respect for agreements; (ii) prevention of fraud; (iii) respect for autonomy and (iv) care not to cause damages to others. The first principle, of respect for agreements, is the central element of the principal-agent relationship, but at the same time it reflects a morality necessary for the working of markets. For Shankman: Without binding agreements, agency relations and therefore agency theory cannot exist, since agency theory takes as its focus the contracts between principals and agents. Without binding contracts, individuals are unable to engage in the kinds of activities that depend on principal-agent relations, namely the creation of wealth through the production of goods and services. That is, business enterprise itself cannot function without the implicit assumption that individuals will honour the agreements that they create freely (1999, p. 328). The second rule that of avoiding fraud is likewise necessary for the good working of markets. As Bowie and Duska (1990, cited in Shankman, 1999, p. 328) indicate a business community in which nobody trusts anybody cannot survive. The last two principles (avoiding harm and respecting others’ autonomy) are necessary for liberty to be established which in the last result is an essential condition for individuals to engage in the fulfilment of contracts (Quinn & Jones, 1995). For Quinn and Jones (1995), therefore, agent’s responsibility is far wider than merely guaranteeing shareholders their economic results, as agency theory presupposes. The pressures of moral behaviour in relation to the markets and to the other stakeholders are equally important. Shankman (1999) concludes that agency theory is subsumed by stakeholder theory, and that due to its restricted perspective, managers should run a company in accordance with the stakeholder model of the firm. The contrary position to stakeholder theory, of authors like Jensen (2000) should be mentioned, maintaining that the company should follow only an objective 37 function, which is the value seeking of the company. For this author, the multiple objectives of stakeholder theory are in reality only strategic, and value creation, this is really the principal object to be attained. In this same sense, Sternberg (1999) calls attention to the fact that managers can sustain themselves through multiple diffuse objectives in the organization to justify decisions that are not necessarily conducive to value maximization of stockholders, thus causing a serious diversion in the agentprincipal relationship. In spite of this array of critics, agency theory is a powerful theoretical framework in many research contexts. Thus, the option chosen in this dissertation was to consider agency theory, whenever it is applicable, with complementary theories, as recommended by Eisenhardt (1989a). 38 CHAPTER III QUALITATIVE RESEARCH METHODOLOGY III.1. Research methodology approach In the development of this thesis, I decided to avoid unproductive confrontation between qualitative and quantitative approaches, discussing which method is superior to the other. My understanding is that such a view is reductive and is based on an already overworked scientific discussion, only tending to impoverish it further. Clearly, each of both methodological strategies has advantages and limitations that are inherent. The qualitative methods, for example, are suitable for situations where one wishes to construct theories, allowing a greater depth in relation to research subjects and the reality that surrounds them. In turn, quantitative methods are pointed out as ideal for testing theories on the basis of statistically generalizable results. Yin (1994) indicates that questions like ‘how’ and ‘why’ have a more explanatory nature and should be treated by qualitative methods, while questions of ‘who’, ‘what’ and ‘where’ have better treatment as quantitative data. Zhang and Wildemuth (2009) point out that the essential difference between quantitative and qualitative methods is that the former are deductive, aiming to test hypotheses generated from theories or previous empirical research. In contrast, qualitative methods are basically inductive, ‘grounding the examination of topics and themes, as well as the inferences drawn from them, in the data’ (Zhang & Wildemuth 2009, p. 380). It is important to note that qualitative research is not necessarily opposed to quantitative; on the contrary, in certain situations, they can and should be complementary (Triviños, 1987; Goldenberg, 2002). Goldenberg (2002) maintains that the integration of quantitative and qualitative analysis is extremely important in allowing cross-checks and conclusions, through the technique of methodological triangulation, which aims to maximize the understanding of a phenomenon, in which the limits of one method are complemented by the other. Weber (1990) also pointed out that the best content-analytic studies use both qualitative and quantitative operations. 39 In line with this proposal, the present study was developed from an approach that combines qualitative and quantitative research. In its qualitative stage the use of non-participative direct observation was adopted at first in order to identify macro categories of corporate governance relevant in the study of family business in Brazil, which could then be refined through successive interviews within the process of building multiple cases, generating at the end a statement of theory. These case studies have allowed the identification of factors related to the tensions and conflicts involving the actors present in the family, ownership and management dimensions. In this sense, the qualitative phase was essential in providing the elements for the construction of the Family Business Conflict Agency Model, proposed by this research. III.2. Grounded theory As a qualitative methodology the grounded theory goes beyond merely counting words or extracting objective content from texts to examine meanings, themes and patterns that may be manifest or latent in a particular text. It allows researchers to understand social reality in a subjective but scientific manner (Zhang & Wildemuth, 2009, p.1). Among the existing qualitative methods, grounded theory was initially chosen as a methodology where theory emerges directly from real-world data collected during research. On the premise that a theory evolves through a process of research, Taber (2000) defines grounded theory as an inductive method of formulating theories through systematic and simultaneous processes of data collection and analysis, which are continuously interpolated. For Lings and Lundell (2005) the ultimate goal of grounded theory is to propose the construction of theories differing from other methods that claim only to verify the validity and applicability of certain theories to certain situations. The grounded theory grew in importance and recognition in the sixties by Barney Glaser and Andelm Strauss’ publication The Discovery of Grounded Theory, which sought to diminish the distance between theory and empirical research in social science. The authors intended in this way to offer an empirical basis for 40 theories based not on hypotheses, but on data, suggesting a logic and specificity for them and legitimizing qualitative research. In the following years, the originators of grounded theory continued to develop the method independently of each other, promoting a bifurcation in the original model. Warburton (2005) explains that Glaser emphasised a more ‘purist’ approach in which the construction of the theory was not influenced by the preconceived knowledge of the researcher, emerging directly from the data. But Strauss subsequently joined Corbin and both started to defend the legitimacy of the researcher associating other methodologies to grounded theory and applying their pre-existing perceptions and experiences whenever appropriate because what counts is to ensure the end result, which is to elaborate or extend existing theory (Strauss & Corbin, 1990). Therefore, their assumption was more ‘pragmatic’ defending a more ‘structured’ attitude to theory building, and prescribing the use of a set of analytical tools and guiding principles (Zhang & Wildemuth, 2009). To Zhang and Wildemuth (2009) grounded theory comprises a process designed to condense raw data into categories or themes based on valid inference and interpretation. This process combines a deductive attitude to data analysis with inductive method. Through an inductive approach, researchers carefully examine and carry out data comparison, leading to the emergence of themes and categories. Otherwise, the approach is deductive when concepts or variables are proposed from theory or from previous studies (Berg, 2001). Thus, following Charmaz’s (2000) understanding of how grounded theory should be developed, this study chose to combine the two approaches, resulting in an iterative deductive / inductive process leading to successive approximation theory generation. This in no way invalidates the conduct of this research in accordance with the principles of grounded theory, after all, as expressed by Langley (1999) ‘we should not have to be shy about mobilizing both inductive (data-driven) and deductive (theory-driven) approaches iteratively or simultaneously as inspiration guides us’ (1999, p. 708). Before each step of data collection shown in Figure 3.1, in addition to the categories that have emerged in the process of data analysis of each of the cases 41 (inductive domain), the literature review carried out during the development of the study has influenced me, allowing me to make new connections with what was perceived during the development of cases, adjusting questions and making inferences that influenced the setting the questions and which, therefore, helped to shape the final model (deductive domain). The literature review which is presented in section V.5, therefore, cannot be interpreted as if it had been conducted after completion of all four cases. Figure 3.1 - The grounded theory 'data dance' Endgame:  Saturated categories  Theory validation against literature  Statement of Theory Inductive domain Deductive domain Data analysis: More refined categories Data collection by interviews Case Study 4 Data collection by interviews Case Study 3 Data collection by interviews Case Study 2 Iterative ‘dance’ towards greater saturation of categories Data collection by interviews Case Study 1 Data analysis: More refined categories Data analysis: More refined categories Data analysis: Refined categories Data analysis: Preliminary categories Data collection by non-participative direct observation Kickoff:  Selection of research topic  Initial literature review Source: Adapted from Kelsey, 2003 In this recursive approach, each round of data collection is conditioned by what has been learned in the previous round of data analysis. This ‘data dance’ 42 proceeds iteratively so that theory is progressively refined, until a point of theory saturation is reached (Kelsey, 2003). An important point that should be stressed is that to conduct a case study, the development of the theoretical foundations represents the basis of support of the whole scope of the research, helping the researcher in the deepening and mastering of the theory. As Yin (1994, p. 28) recognizes in affirming that ‘theory development prior to the collection of any case study data is an essential step in doing case studies’. Such a statement, nevertheless, runs counter to the basic principal defended by grounded theory, eliminating the naive spirit that is expected of the researcher in his work. For Glaser (1998), whenever one opts to combine the use of both methodologies maximum care must be taken to ensure that basic principles of the development of a case study do not deform true emergence for theory creation. Thus, in combining the use of grounded theory and the case study, it should be made clear which of them is guiding the investigation. The fact that I have chosen as the basic theme of this dissertation, corporate governance in family-owned companies, which is for me a field little explored previously, has made the adoption of grounded theory more genuine, without significant contamination. Therefore, in this research the grounded theory was used as the prevalent methodology. III.2.1. Initial data collection According to Warburton (2005), the two conventionally acceptable sources for data in grounded theory studies are first person observations and face-to-face interviews. For the initial data collection, the non-participative direct observation tool was chosen. Non-participative direct observation Observation is one of the methods of data collection much used in the investigations of social sciences. In the method of non-participative observation the researcher observes the facts occurring in the group studied without manifesting 43 himself. These facts are observed directly by the investigator without intermediaries, reducing subjectivity. The researcher places himself in a situation that allows him to see the manifestations of the phenomenon to be studied without becoming involved, filling the role of spectator. This type of methodology is flexible because it permits the observer to register aspects perceived relative to facts and behavioural patterns as they occur and permits change of focus of observation should another become more interesting and suitable. Gil (1999) mentions specifically that among other advantages, observation permits obtaining the elements that define research problems. Unit of analysis In view of the scarcity of the literature on corporate governance in familyowned companies in Brazil and due to the easy access to the companies participating at family business programmes delivered by Fundação Dom Cabral 1, the researcher has opted initially to convey a non-participative direct observation in the Programme for Shareholder Development - PSD, group MG-7, which started its activities in October 2006. The PSD includes family-owned companies that seek to develop present and future shareholders, providing them with the capacity to comprehend and exercise their responsibilities and rights related to management, to preserve and maximize the family patrimony. In the same way, they are encouraged to reflect on the succession processes, to ensure the longevity of the company. Each group develops itself in sessions of a day and a half per month, for one and a half years, delivering structured activities such as lectures, testimonies, formal classes, and in house project monitorships. Throughout the component activities of this programme, the researcher observed behavioural patterns and testimony in the debates, panels and exchanges between the participants within the programme. Ten companies made up the PSD group MG-7, all of them family-run, with the following profile (base Dec/2006): 1 Fundação Dom Cabral (FDC) is a Brazilian business school founded in 1976, located in Minas Gerais State, Brazil. According to the 2011 Executive Education Ranking, published by the English newspaper the Financial Times, FDC reached the 5th position among the most important business schools in the world. 44 Company Minas Medical School Kent and Brower Ltda. Honey Designing Better Achievement Annual revenue US$6.5 Mi Economic sector Education Time of existence 35 years US$26.7 Mi Pharmaceutical and cosmetics retail Furniture manufacture Technical Engineering Services Petrol distribution Dairy produce 29 years 2 19 years 1 11 years 1 23 years 2 45 years 1 Telecom & Agribusiness Prefabrication of building components Fertilizer production Drugstore chain 52 years 3 47 years 2 14 years 2 73 years 1 US$2.8 Mi US$14.0 Mi Go-gas US$160 Mi Milk & Candies S.A. Bental US$192 Mi Prize US$31.0 Mi Minas Agro Group Mainstream Drugstore US$14.0 Mi US$950 Mi US$4.8 Mi Generation in control st 1 nd st st nd st rd nd nd st Capital structure Privately held Limited liability Board of Directors None Limited liability Limited liability None Privately held Privately held Privately held Privately held None None None Present Present Privately held Limited liability None None None  Average annual revenue: US$ 140 Mi. (with large dispersion, ranging from a minimum of US$ 2.8 Mi. /year to a maximum of US$ 950 Mi. /year).  Economic sector: 30% industry, 30% services, 20% trade and 20% of agribusiness.  Average time of existence: 34.8 years, the oldest 73 and youngest 11.  Generation in control: 50% in the first generation, 40% in the second generation, 10% in the third generation.  Capital Structure: all observed firms were privately held.  Only two companies had a formal Board of Directors. III.2.2. Application of the method This stage lasted about a year, from October 2006 to September 2007. Despite the long-term investment, this was at the end vital time for the development of this research, since it was my first extensive contact with the universe of family 45 businesses. In my academic and professional history, my previous experience included the state-owned enterprises and multinationals in Brazil. Sixty hours were invested to monitor the initial modules of the program, which addressed the following issues: i. The vision and the dynamics of family business ii. Building relationships of trust iii. Pulverization of the patrimony and the new Civil Code of Brazil iv. Corporate asset and liability of owners and managers v. Organizational solutions: Shareholders Agreement, Holdings and Family Office vi. The basis of success and causes of failure in family business vii. Family myths and family and business relationships viii. Succession in ownership and in management. The initial intention of this phase was to establish the most relevant constructs on family business matter. I concluded, however, that the group’s dynamic was not supplying the necessary data that could contribute to this end. One of the central points that became increasingly clear was that the attitude of the group was only reactive to the points brought up for discussion by the professors, and was therefore contributing little to the emergence of genuine content that could actually characterize this stage as a real grounded theory process. From this assumption, I decided not to invest a further six months observing the other sessions and to start working directly on multiple-case studies, and from them, raise the hypotheses to be tested at the quantitative research stage. This observation stage, however, was positive for getting to know the group better, identifying the companies that best deserved to be the object of case study construction, and for establishing contacts that would facilitate the subsequent planning of the interviews. It was also important for the identification of common points of concern that should be taken into account in the preparation of the interview guide for the subsequent interviews in the case studies phase. 46 III.2.3. Results of non-participative observation Table 3.1 presents the main dilemmas that were identified during the application of the non-participative observation method in the PSD group MG-7. Table 3.1 - Results of non-participative observation Observed dilemmas Description Longevity Central element of concern of the firms participating in the programme. This is quite understandable, as the PSD´s proposal is exactly the training of shareholders of family-owned companies with the view to making the company’s life a long one. In the programme the role of family in management, the attributions of shareholders and the process of family succession were discussed. Principles and values According to participants' opinion, these elements act as ‘glue’ keeping the owning family bound to the company at the same time as they promote the continuation of the founders’ purposes. Succession It has emerged as an element of concern, albeit in a veiled way, especially when in the presence of the founder. Around this theme, special reference to the discussion about the importance of the level of competence required for key positions, and the tendency of parents to overestimate the abilities of their children were made. Presence of family members in management This was also a subject of discussion. Different types of family businesses were compared from the working family company, which admits every family member who wishes to work in it, regardless of their training, to those that are classified as family-run businesses, where the rules are very clear for admission of family member. Companies with family governance, where family members are incorporated only in the governance bodies were also considered. Trust The establishment of relations of trust also emerged as a major theme, whether between family members, or in relation to nonfamily managers. In this regard, two questions have attracted attention. The first was for a possible confrontation between confidence and competence; after all, what is more important: to have a reliable professional or one who will be able to deliver the expected results to the family? This apparent dichotomy, as if those attributes could not coexist, or even more, as if competence was not one of the main attributes for building trust in the business environment, creates the possibility of justifying the preference for family professionals because he/she naturally is the first choice in terms of reliability. Mechanisms of control Another issue that deserves mention was the discussion regarding to what extent the presence of trust reduces the need for more robust mechanisms of surveillance. Separation of family and company’s assets An awareness of the need to separate family assets from those of the company appeared as an element of concern regarding the professionalization of the company. Board of Directors The relevance and benefits of the adoption of a Board of Directors, as well as the presence of independent directors, has emerged as a topic of discussion. Source: Prepared by the author 47 Finally, another point of great interest was to understand how the IPO would be beneficial for family businesses in Brazil. To understand why this issue has entered in the agenda of meetings of the PSD, it has to be considered the historical capital market in Brazil in 2007. The volume of funds raised by Brazilian companies in initial public offerings that year amounted to $ 60.5 billion, a jump of 380% over the previous year, according to data from Thomson Financial Consulting, representing a total of 105 operations. On the same basis of comparison, the sector operations worldwide have expanded only 17%. The global financial crisis of 2008, however, strongly affected investor confidence, slowing the rush of companies to assess the possibility of seeking financing on the capital market and leading the researcher to no longer prioritize the IPO variable as part of the research. Figure 3.2 presents this evidence. Figure 3.2: Evolution of the Brazilian capital market Source: ANBIMA and CVM (Brazilian SEC), 2011 48 III.3. Case studies As defined by Yin (1994) case studies are empirical studies of particular instances of a phenomenon that are usually based on a variety of data sources. We may build theories from case studies using one or more cases to create theoretical constructs from empirical evidence (Eisenhardt, 1989b). In constructing the theoretical model of agency conflicts in family businesses, I decided to use multiple case studies to build grounded theory, inasmuch as ‘theory building from case studies is likely to produce novel theory, juxtaposing contradictory and paradoxical evidence’ (Eisenhardt, 1989b, p.546). It is known that the case study is a method that allows investigating a particular problem in depth, identifying specific cases that stand out for their wealth of information, making it possible to shed light on a phenomenon more generally, from this specific example. Founded in US in the early 20th century, the case study was used primarily in the administration area, but it is currently adopted in the investigation of phenomena from various fields of knowledge (Gil, 1999). In the case study, I seek numerous and detailed information about the object of study in order to grasp the situation in its entirety. Although it is not possible to draw generalizations from a single case, since we cannot say that the results will be repeated in all the examples of the phenomenon under investigation, the detailed study of a particular case results in comprehensive knowledge of a specific reality, and the results achieved can enable hypotheses to be formulated for guiding further research (Triviños, 1987). It was so in the present study, seeking to understand the main tensions in relations between business and family, allowing the construction of a theoretical model of the prevailing conflicts in family businesses. Furthermore, the adoption of the multi case study method, as recommended by Yin (2005), allowed a deeper analysis of the phenomenon, encompassing the logic of planning, data collection techniques and approaches specific to their analysis. In addition, one must consider that multi cases have the advantage of showing results, generally considered more compelling and robust compared to single case studies (Yin, 2005). According to Eisenhardt (1991) multiple cases enable comparisons to be made to clarify whether a finding is valid only for a single 49 case or whether it is replicated in several cases. The line of reasoning used was the replication of results. After each significant discovery was announced, from a single experiment, the attempt was made to replicate it, refining the categories initially identified, through the second, third and finally the fourth case. After completion of the analysis of each of the four cases, the final step was initiated for the construction of the questionnaire to be used in the quantitative stage. In addition to the evidence and the results that unfold from the analysis of the cases, the theoretical basis was used as a counterpoint to the empirical results achieved, allowing the identification of gaps or strengthening the theories adopted. It followed, therefore, the logic provided in grounded theory, that the literature must only be researched in the final stage of the analysis, when the theory discovered might be confirmed. Thus, the contrast between the adopted theoretical and empirical evidence was used to structure and justify the elements present in the Family Business Agency Conflict Model. III.3.1. Interviews The method of data collection utilized for categories’ refinement was semi structured interviews. The main task in interviewing is to understand the meaning of what the interviewees say (Kvale, 1996). Thus, this is a technique of data collection very suitable for obtaining information on people’s sentiments, desires and intentions. It can be defined as a face to face dialogue between the investigator, who collects data, and the person investigated, who supplies information. According to Gil (1999), it provides a good opportunity to obtain information of greater accuracy and depth, and the possibility of clarifying doubts. The interview can be either formal or informal. The formal interview follows a concerted plan of action and is used when one wants in-depth information that can be found in private and with respondents recruited from predetermined sites (Chenitz & Swanson, 1986). The formal interviews can also be structured or unstructured. The structured interview is based on the premise that all responses should be comparable with the same set of questions capturing the differences between individuals. In the case of unstructured interviews, the interviewer doesn’t always try to get the same 50 kind of response using the same type of questions: the purpose is to get the information in the respondents' own words, get a description of situations and clarify details. Also to be mentioned are interviews where the two types are combined at different times, or ‘hybrid interviews’, called semi-structured, which are moderately standard, but allow modification of the sequence and number of questions, and even the words used in each question. Important also to note, that the success of this method depends fundamentally on the level of the personal relationship established between interviewer and interviewee. The initial contact established during the period of accompanying the PSD - MG VII group has facilitated data collection via interviews. Even if not all the players necessary for data collection took part in the PSD’s activities, those who did constituted important channels of entry for all necessary interviews, including those that did not. All interviews were conducted personally by me, and in addition to the notes made at the time were also recorded. Thirty-four hours and thirty-nine minutes of recordings related to the interviews were stored, which after transcription, generated a vast amount of material for analysis. III.3.2. Interview guide A script to assist in conducting the interviews was developed as a data collection instrument. The main elements raised in the preliminary stage of nonparticipative observation and supported by the initial literature review guide the formatting of the questions to be investigated in the interviews, as can be seen in the Table 3.2. Table 3.2: Origin of interview guide and its main questions Issues Agency conflict Source of issues Berhold, 1971; Berle & Means, 1932; Jensen & Meckling. 1976; Mitnick, 2006; Ross, 1973. Questions comprising the interview guide Examples of conflict situations due to:  Lack of impartiality  Loss of confidence  Interests of the controlling family not being pursued  Lack of competence  Family rivalry  A quarrelsome succession process. 51 Competence Dilemma emerged during non-participative direct observation.  Corporate Dilemma emerged  cohesion and during non-participative  alignment direct observation.   Family   Mustakallio, 2002 Council   Family goals Astrachan & McMillan,  2003; Chrisman et al.,  1997; Chirsman et al., 2005; Davis, 1983;  Olson et al. 2003.  IPO choice Dilemma emerged  during non-participative direct observation.  To what extent the criteria of competence are taken into account in hiring and promoting managers, regardless of kinship. Family vision about the business. Alignment of values between shareholders Sharing of the vision and strategic alignment Family unity and its impact on the alignment of the decision making process. Existence of Family Council. Family Council contribution in conflict management. Family Council role in the planning of succession and in preparation of new generations. Family Council as a tool of strengthening values. Family definition of success. Main financial goal outcomes considered by owners. Dividend payment polices. Family role in goal achievement Projected growth and need for raising funds from third parties. Interest in going public: reasons to compel and to prevent it. Mustakallio, 2002; Neubauer & Lank, 1998. Longevity Dilemma emerged  during non-participative direct observation.   Mechanisms of control. Dilemma emerged  during non-participative direct observation.  Jensen&Meckling, 1976.   52 The way the owning family considers longevity: keeping its control or, independently of this, preserving the existence of the company following its mission. Possible trends regarding merges and acquisitions. Risk profile of the family ownership group. The decision making process within the company. Mechanisms of planning and control adopted and its contribution to the business evolution. Existence of a system of accountability which permits delegation, reward and attribution of responsibility for success and failure in the attainment of strategic targets. Analysis of results compared to the defined strategies. Presence of Dilemma emerged during  family non-participative direct members in observation. management  Chrisman et al., 1999.    Principles and Dilemma emerged during  values non-participative direct observation. History of the business and the involvement of family members in the company. Policies concerning employment of family members. Level of autonomy of nonfamliy managers. Competitive advantages and limitations with the presence of the family in the business. Source of income for the family members. Family values and cultural characteristics that influence the way the business is run. Basu, 2004; Donnelley, 1988; Hollander & Elman, 1988. Role of Board Dilemma emerged during  of Directors non-participative direct  (BD) observation.  Beckhard & Dyer, 1983; Moe, 1984. Succession Dilemma emerged during  non-participative direct observation.   Danes et al., 1999; Davis et al., 2000.  Trust Dilemma emerged during  non-participative direct  observation.  Bowie & Duska, 1990; Pollak, 1985. BD existence and its real role. BD details about its size, composition and profile of members. CEO’s selection and performance appraisal. Description of previous succession processes. Existence of rules of succession. Competence and merit as principles for succession. Process of legitimation of the successors, if any. Situations of breach of confidence. Presence of opportunistic behaviour. Lack of the necessary competence to take the organization within range of business results. Source: Research data An interview guide ensures that the same general areas of information are collected from each interviewee, but still allows a degree of freedom and adaptability in getting the information, which is essential in a grounded theory process. 53 III.3.3. Unit of analysis The body of qualitative research consisted of 26 individuals belonging to the four family businesses studied. These were chosen from among the twelve companies participating in the PSD group - MG7. As selection criteria, companies with different volumes of revenues, companies with different generations heading management, and companies with somewhat peculiar succession histories were chosen. In order to comply with the commitment to confidentiality assumed with those interviewed, all details such as the names of the company, of the founders and the family members were substituted by fictitious names, and data that could help to identify the organization were suppressed. Respondents were selected intentionally, rather than at random, which confers more objectivity on the data collection, because, as stated by Thiollent (1988), the delimitation of respondents represents an intentional way to access relevant information on the subject studied, from one group of informants. The choice fell on the following actors, when present in each of the companies: i. The owner(s) acting as CEO and/or Chairman of the Board of Directors. ii. The CEO/Chairman of the principal business whether member of the family or not. iii. At least one more member of the owning family (if any) that is active in the company business (as member of the board and/or employee). iv. At least one member of the owning family (if any) which has no participation either as member of the board or company employee. v. At least one active member of the owning family (as member of the board and/or employee) that represents a generation distinct from those previously represented (as a means of guaranteeing a multi-generation perspective). vi. At least one nonfamliy member in senior management of the principal business. vii. If there exists a significant nonfamliy partner who should be interviewed. 54 Regarding the definition of the number of respondents, the level of access and availability of the research subjects together with the criterion of saturation presented by Alberti (1990) were considered. This author states that at any given time, the interviews end up becoming repetitive, both regarding their content as also the manner of construction of the narrative, that the time has come to terminate the data collection. III.3.4. Secondary data gathering As a complement to the interviews, the basis of primary research data was supplemented by other documents requested such as statutes, minutes of meetings, internal reports, manuals and codes of ethics and conduct. Secondary data were also collected related to the organizations studied, including: newspaper clippings, information posted on company websites, articles, dissertations and other scholarly publications related to the organizations in question. The combination of primary and secondary data was important because it helped me to progressively broaden my understanding of the critical issues affecting the companies and incorporate new questions throughout the interviews which had not emerged naturally from the previous interviews, but which could then be explored in the following interviews based on what had been gleaned from the secondary data. Thus, I sought to establish data triangulation from the use of multiple sources of evidence, which, according to Yin (2005), enabled me to encounter a range of issues of a historical and behavioural nature and relative to attitudes, which have the advantage of consolidating converging school of thought. The objective here was to collect information that allows the researcher to: i. Construct the general profile of the owning family and of the company. ii. Map the principal strategic and business movements. iii. Describe the principal demographic and contextual data of the company environment. 55 iv. Register the principal results of the companies through financial reports, balances and others. v. Understand the governance model, the family structure and their interconnections. vi. Effect triangulations to corroborate information collected during interviews and observations. 56 CHAPTER IV INTRODUCTION OF CASE STUDIES IV.1. The Milk & Candies Case IV.1.1. The company Milk & Candies S.A. is a company with activities in the dairy foods sector with the production of powdered milk, cream, condensed milk, chocolate preparations, milk-based sweet, butter and milk caramel. It is located in the state of Minas Gerais, and maintains branches or distributors in all the main commercial centres in Brazil. The company was created from the merger of two separate companies: Candies Brasil founded in 1935, in the interior of the state of São Paulo and White Milk, founded in 1948, in the State of Minas Gerais. In 1955, the then Candies Brasil initiated the manufacture of milk caramels, a product that became its trade mark. In its turn, in 1958, White Milk, which up to then had produced only butter, initiated the production of powdered milk. The acquisition of White Milk by the present controllers – Edgar Mills and Friedrich Haag – occurred in 1961. In 1963, the partners also became owners of Candies Brazil. Up to 1969, however, the two companies continued to operate totally independently. In this year, the Minas Gerais Integrated Institute of Industrial Development (INDI) was created, which offered various tax incentives to the controllers to transfer the White Milk factory from São Paulo to Minas Gerais. As the two factories were small, the controllers opted for merger, as they would gain greater economy of scale, due to the grouping of all the production in only one industrial unit. In this context, the current company name was chosen: Milk & Candies S.A. It is valid to emphasize that the tax incentive contract signed with INDI was not honoured by the Government of the State of Minas Gerais, a fact which brought great difficulties in its train for the new recently-incorporated company. It had repercussions during the whole of the 1970s, forcing the shareholders to make big sacrifices to pay off the liabilities acquired with the investments made in the structuring of the new 57 plant. In spite of all the difficulties, in 1973 they managed to complete the expansion of the powdered milk manufacture, going from a volume of 20 thousand to the processing of 120 thousand litres /day. In June 1980, a joint venture was made with Kraft Foods, the largest NorthAmerican company in the food sector at the time, which took over 40% of the capital. In 1982, Kraft exercised the option of increasing its participation, becoming the owner of 60% of the share control of Milk & Candies. At this moment, Haag and Mills left the company direction, remaining only as members of the Board of Directors which had been established. In addition, at this time, Friedrich Haag and his family decided to return to Germany. In 1985, due to restructuring in its operations in Brazil, Kraft decided to relieve itself of the business, leaving it to Haag and Mills to resume the share control. These, in their turn, confirmed the executives that were in the command of the company, remaining themselves outside the daily routine of management. In 1990, the Collor Plan2 was decreed by the Brazilian Federal Government which involved the confiscation of company investments in the country, bringing new difficulties for Milk & Candies. The company, which had large stocks to meet demand during the off season, ended by entering into an accommodation with creditors. One of the consequences of this creditors' agreement was the acceptance of the entry of a new partner, John Harris, in order to leverage the necessary capital to recover the firm’s financial health. The amount made available by this current minority partner, was first contracted as a loan. Subsequently, he was invited to join the company due to the high interest he was charging and the incapacity to pay off the debt at that moment. Thus, from this moment on Milk & Candies’ capital structure has followed this distribution: 2 The Collor Plan was one of the three plans to stabilize inflation carried out during President Collor's two years in power. The first approach, adopted in March 1990, was freezing government liability (such as internal debt) and restricting money flow in order to halt inertial inflation. 58 Table 4.1: The Milk & Candies’ capital structure Shareholders % Capital Edgar Mills 37.5 % Friedrich Haag 37.5 % John Harris 25 % Source: Research data In 1993, Edgar Mills supported by his partner Friedrich Haag, decided to return to the company direction, bringing with him his sons, to be part of the executive body of Milk & Candies. The family management was sufficiently effective to turn the situation around and initiated, from 1994, a period of great dynamism, bringing vigorous expansion of the candies line with the acquisition of high production capacity equipment. In 2000, the project was concluded that represented the largest investment in one stage in its history (R$ 32 million), utilizing only its own resources. This project envisaged an expansion of the installed capacity in dairy product processing, jumping from 300,000 to 800,000 litres of milk/day. In 2005, the company started its condensed milk, cream and milk-based drink plant, expanding processing capacity to 1,100,000 litres /day of milk. Since the return of the Mills family to the management, Milk & Candies has grown at an average rate of 18.5% per year, leading their controllers to estimate that even before 2018, it will be the largest dairy products family company in Brazil. Table 4.2 gives the evolution of some performance indicators of Milk & Candies from 2004 to 20073: Table 4.2: Milk & Candies’ Business Performance Evolution of the figures 2004 2005 2006 2007 Number of employees 1,110 1,226 1,223 1,443 R$ 12,112 R$12,845 R$ -1,689 R$23,468 Sales (in millions) R$ 286 R$ 347 R$ 346 R$457 Net worth (in millions) R$27.4 R$39.0 R$ 37.3 R$54.9 Net profit (in thousands) Source: Research data 3 This case study was undertaken in 2008, and therefore contains data only up to the year 2007. 59 IV.1.2. Profile of the interviewees Table 4.3: Milk & Candies’ profile of the interviewees IV.1.3. Members of the families In the following charts, the members of the Mills and Haag families are presented, those that were interviewed for this case study being given in green. Figure 4.1: Haag Family Tree Chairman Board member 60 Figure 4.2: Mills Family Tree CEO Adm/Fin. Director Edgar Mills Jr. Sales manager Source: Research Data IV.1.4. Top Management Team Figure 4.3: Milk & Candies’ executive team IV.1.5. From the present harmony to the uncertain future – the succession issues In spite of various difficulties, including the creditors' agreement phase, the work and commitment of the two main partners of Milk & Candies were responsible for the company clearing up its debts and continuing growth. In this sense, in addition to the entrepreneurial profile of both, the fact that their views are in good alignment 61 and the relationship of trust constructed between them are pointed out, several times, as essential factors in the success of the business. (CI01) I think that the great trump card in the success of this company is that the two [Edgar and Friedrich] never left a subject pending. When they got out of step about a matter, which they couldn´t resolve immediately, they used to say: tomorrow we´ll return to this subject’. Sometimes one gave in, at other times it was the other, or perhaps the two found a happy medium between the two positions. (I02). (CI02) The two families have an almost absolutely identical idea in relation to the company and its development. [...] We are in such perfect harmony that it couldn’t be better. [...] I know that it is almost unbelievable, but Edgar and I have never had a disagreement. (I03). (CI03) It is very good seeing them always in agreement. It’s a positive thing. The day will never dawn when Edgar gives an order and tomorrow Friedrich countermands it (I05). (CI04) They work together, with cohesion. I think the respect that exists between them is very important. I think that my father and his brothers succeed in achieving this also, and it is something that I shall try to do and, I believe all my cousins also. (I06). (CI05) The integration between them is important even for keeping the staff pleased. We can see that they think together, from now to 20, 30 years ahead. They are almost 80 years of age and still think about what they are going to do in 10 years’ time. They create a common vision of the future. (I04). This alignment is perceived and appreciated even by the minority shareholder, who added: (CI06) They have been more than 50 years together. I have been a partner for 10 years and during this period I have always seen how the relationship between them is very good. Excellent indeed. (I07). In addition to valuing such alignment, John Harris has shown that he trusts the leadership of the Mills family, the reason why he systematically supports the main strategic decisions. Actually, he recognized that more than trust, the principal attribute that makes him agree with the other partners is the competence of the managing family. (CI07) They are very competent. It is not just a question of trust. It is that they are really competent, within the operational area as well as the financial, the responsibility of [Alexander] and [Edmond] respectively. What makes the difference is their quality. (I07). 62 In fact, both the 2nd and 3rd generations of the Mills family show that they deserve the recognition that they have from the other partners, as regards their competence, as well as from the staff and other family members. (CI08) In management, [Edmond] can already substitute [Edgar] in the function of CEO. And will do it very well. The other son is a well-centred person, the one who is directing the plant. And there is the first son who is in the company, but in a more subdued role. A grandson should also be mentioned who was nominated general sales manager for Brazil, not because he is a family, but because of his capacity, with my full agreement. He is 27 and does a very good job of work. Therefore, the sequence of the company management will certainly remain in the hands of people from the Mills family. (I03). (CI09) I joke with him [Edgar] that his children are better than he is. Because we had this concern about what would happen when he [Edgar] died. But I don´t worry about this anymore. [Edgar] has brought up his children based on professionalism and not on paternalism. (I04). Contrasting with this recognized competence of the Mills family in management, Milk & Candies has already gone through conflict situations arising out of the limited capacity of a nonfamliy CEO, Ernest Lopes. This happened precisely during the period immediately preceding the Mills’ return to the company and postcreditors' agreement. There was the need for a strong restructuring process and Ernest did not come up to the demands that the company’s new moment required. Edgar Mills described his profile as being (CI10) [...] a very good person, a man of great good character, but very conservative (I01). For Edmond, there was a clear divergence of viewpoints between the controlling families and this CEO, on the investment profile and the risks that should be accepted to stimulate the growth of Milk & Candies. In his opinion, Ernest made a grave mistake in (CI11) [...] not looking at the market as it should be looked at, what direction it was taking and where we should position ourselves to increase our market share. The company went through a period without any investment in its industrial park, which certainly harmed it (I02). But in addition to this episode, the company also faced the typical agency conflict due to bad faith in the management of company resources. According to Edmond, the fact of the company Head Office previously being in Rio de Janeiro, far 63 from its industrial park, made greater control over company operations difficult, reinforcing the risk of opportunistic behavioural patterns. (C12) In April ‘97, [...] when we were doing an audit, we discovered fraud. It was a time of 40% inflation a month, which created a favourable atmosphere for deviousness. They had taken part of the company invoicing and deposited it in their own bank account, keeping the result of the financial applications and at the end of the month, returning only the value of the principal to the company [...] even directors were involved: and everyone was dismissed. They had set up a real racket there. So, we made the changes necessary to give transparency to our financial operations. (I02). If on the one hand, the harmony between the shareholders and the competence of the family managers seems to have removed any source of agency conflict in the short term, the succession process still remains a pressing source of tension. Apparently, it is tacitly agreed that in the absence of Edgar, the management will be assumed by his son Edmond. Although this decision appears in some minutes and documents required by creditor banks, the subject has never been formally discussed between the person succeeded and his potential successors. (CI13) It is not that this is a taboo subject. But we never sat down to talk about it. My brother [Alexander] himself has already said: ‘[Edmond], I don´t have the slightest desire to deal with the business. It ends up with a CEO now and again having to curry favour with the politicians, to participate in social events. I am very well here in the factory, I like what I do. I don´t want to leave here. As far as I am concerned, you can definitely count me out’. He says to me: ‘the person my father is preparing is you’. OK, but dad [Edgar] doesn´t encourage discussion of this subject. Because he doesn´t say clearly when he is going to leave. (I02). Even without the existence of formal planning to guide the passage of management from the 1st to the 2nd generation, Edmond states that he feels up to taking over the challenge of running the family business. Supported by his long experience within the organization and in the establishment of good relations with the principal stakeholders involved in the business, he sees the transition as a natural process and actually already occurring. Although the reluctance of the person about to be succeeded in defining the date of his departure, or even the excessive respect shown by everyone in even mentioning the subject, permits everyone to have their own opinion as to when this process will actually come to pass. (CI14) So the natural order of things is that I will take over the position of CEO. And I even believe that it will be next year. My father said years ago that he would go on until he was 80. Now that the 80 is arriving, he might 64 even change his mind. But even he, now and again says: I am really very tired’. (I02). (CI15) I personally promised my children that I would retire when I was 85, but I am already regretting it, because it is arriving. It is going very quickly. And I think that naturally I will be succeeded by [Edmond], as my other son really likes the production side. (I01). (CI16) My grandfather’s life is the company. Whenever I went to the factory, ever since I was small, he showed it to me as if he was pointing out a baby in the nursery [...]. Sincerely, I don't see him today leaving the business alive. Only if his body doesn´t let him [...]. And I told him to stay 15 days or so at the farm, to rest a little. Then he starts to say that we want to get him out of there and said: I only leave in a horizontal position’ (laughter). And I think that this will be precisely what will happen. His life is in here. (I05). As, therefore, the most probable outcome for the succession has never been discussed openly, the possibility cannot be completely ruled out that after Edgar’s death, conflicts or disputes will break out in the controlling family. Nor can the possibility be entirely disregarded of the existence, even if veiled, of divergent interests that have not yet surfaced out of respect for the presence of the founder, perceived as the moral reference of the family and the company. Another point of apparent disagreement, relates to the insertion of the 3 rd generation of the Mills family in the business. Such concern is justified by the increase in complexity of the corporate issues due to the multiplication of the number of shareholders and potential successors when one considers the members of the coming generations. (CI17) Unfortunately my grandfather is not eternal. Therefore, I hope that his successor, who will probably be my father, or one of my uncles, has the vision that he has. Now, when the turn of the grandchildren arrives, I do not know how it will be. Because now it is easier to deal with the succession, as there are only three options. But, when we get to the next generation, with ten grandchildren, it will not be so easy. The people are still young and not all of them are here within the company. There will have to be preparation, an overseeing of everyone (I06). In order to resolve this problem, the creation of formal rules to regulate the entry of family members in the company is stressed as indispensable. The prevailing idea is that capacitation and competence are primordial factors for the insertion of family members in the business. And that, accordingly, the adoption of objective criteria will help to prevent family ties speaking louder in the contracting processes of new managers. Edmond highlights among them: three years external experience in 65 organizations that do not belong to the family; a year abroad, working in one’s area and; a post-graduate or MBA course in areas related to the occupation that one wants in the family company. But such rules still seem to be more in the field of intentions than of effective application, in accordance with what can be gathered from the quotes below in connection with this matter. (CI18) This is a question that I am very concerned about. I see my cousins entering the company and I think that they should be tested a little more. But they are entering. I am third generation and how can I question another member of the same generation that is also entering? I have to stay silent, but I am a little worried about it. (I05). (CI19) If the family member is a good professional, why refuse him or her? It doesn´t make sense. But it also cannot become a rule, where everyone gets something for a relation, a son, a husband. (I06). (CI20) I have requested our HR Manager to put in place the conditions as regards the entry of family members into the company. I told him exactly what I wanted. To go in is one situation, to rise up is another [...]. We cannot allow this to become a ‘jobs for the boys’ situation for the family [Mills]. (I02). To pass a phase of one’s career working in another company that is not owned by the same family seen as an important factor in making it possible for the successors to accumulate prior experience and to start in the company with the capacity to aggregate new visions and ideas. In addition, professional maturity and the recognition achieved in the market are also perceived as fundamental for raising the self-confidence of the successor, helping him or her in their legitimization with the executive group and the other family members involved in the business. (CI21) I think that it is very important working outside! Because I did, I was promoted, I had a good career. This I think gives you a little more confidence. Whoever works in the family company, deep down, right deep down, always has that doubt: ‘if I were a normal employee, what my salary would be? Would I be here in this position?’ I am untroubled, because I came here earning the same salary that I was earning in the market. But before I really had this doubt: ‘Am I here only because I am a grandson?’ (I05). The fact is that, based on the discourses of the interviewees of the 3rd generation, the bond existing with the company is patent, as also is the desire to make their careers there. (CI22) My future is working in [Milk & Candies]. I never thought that my career would be here. I left, but my path brought me back. (I05). 66 (CI23) I know that [Milk & Candies] is priority in my life. You grow up hearing about the company, at the family lunches everyone is there speaking. I always had this curiosity. Ever since I was a child I visited the factory [...]. And I am very close to my grandfather. I had a lot to do with him when I was small. So I have many stories, I am very linked to all this. It is part of my life. And it is not easy to disconnect from a thing like this, from a bond like this. (I06). IV.1.6. Values, conduct and role comprehension The controllers of Milk & Candies demonstrate that they run the company based on solid values, which can help in the alignment of the family members, providing a common vision of the business, a greater convergence in the decisions and, also, attenuating the risk of disputes and family conflicts. (CI24) I think that I and my cousins grew up with the same values: the ethics, the same human understanding, the same value that people give each other, are things that we learned since we were small. (I06). A quite common posture to be found in family-owned companies is the difficulty of the separation between company assets and those of the owners, especially in those with more than one controlling family block, as is the case of Milk & Candies. The attitude of benefiting from company services for personal ends, in addition to creating costs for the company, can nourish quarrels with the other partners in the business. On Edgar’s part there exists great zeal in avoiding any patrimonial posture that could encroach on the other two family blocks that delegate to the Mills family the management of the company. This concern seems be a value that is transferred to the other generations. (CI25) It is striking how careful he [Edgar] is with this. He separates this very well. He knows that he is only one of the partners in the company. For example, he has a farm and if he takes a refrigerator there, he wants to charge for the use of the company truck. We can even argue: ‘But [Edgar], the truck’s got another delivery nearby, we´ll have to go past the front door’. But he will not accept it. He separates company assets from his own very well. (I04). (CI26) A thing that troubles me is when someone of the family takes a driver, I know not where. This irritates me profoundly. Because I don´t do this, ever. I think that the business has to maintain the family in a different manner. It is not generating expenses for the company, but through dividends. (I05). 67 It is valid to point out that these family values do not affect only the future generations of the family, but also influence the vision of the company staff and their conduct. Accordingly, it is possible to infer that the family character of the company can help in the dissemination of a code of conduct for all who work in it and to contribute to the establishment of a healthy work atmosphere. From the operational point of view, these aspects are translated into a low rate of rotation. (CI27) I feel good working here. I see that he [CEO] has ethics. [...] I think that to work in the family company, you find this ethical spirit. No one has the courage to propose anything wrong to him [Edgar]. Therefore, he is an example for the staff. His children also were brought up this way. This was the direction that he gave them [...] my pride in working here is because of the ethics. [Edgar] taught the staff here to reflect his attitudes. (I04). (CI28) I think that this family environment is a very good place to work in. So much so that staff rotation is very low. This helps the professional to grow, which can be seen by the fact that many staff that started with me in the lowest functions are today already in positions of leadership. For the family, honesty is everything. And whoever works in the company gets this message. (I05). In relation to the nonfamliy managers, the controllers seek to adopt salary policies identical to those practised for the family managers. Equality of treatment indicates that there exists a maturity in relation to the understanding of the roles of the controllers, who understand that salary and professional valorisation should extend to the whole executive body of the company, independently of the existence of family ties. It can be inferred that this posture can help in the convergence of viewpoints between controllers and managers, minimizing the risk of agency conflicts. (CI29) The directors have the same salaries [...]. I mean, all directors, family or nonfamliy, have the same salary and, they are evaluated in the same way by the delivery of their results. (I01). Another fundamental point, capable of contributing to the alignment between owners and managers, relates to the tradition of internal promotion, giving priority to the staff longer in the company in the assumption of management positions. Promoting on the basis of time in the company is, up to a certain point, a career plan, albeit informal, and generates Fidelity in the executive body by providing growth opportunities within the company. The reservation to the adoption of such a policy is that the filling of executive positions only by internal promotion brings with it the risk 68 of maintaining the same strategic vision, making the rise of other, perhaps more innovating, ways of thinking, more difficult. These might be aggregated if experienced professionals drawn from other companies were to be contracted in the market. On the other hand, the promotion of managers linked for decades to the company presents the advantage that it maintains and deepens the already incorporated links of trust, minimizing the occurrence of opportunistic behavioural patterns or actions of bad faith on the part of the managers. (CI30) He [CEO] has not sought directors in the market. He has sought internal promotion, finding the in-house potential that we have. [...] Internal valorisation is a company practice. (I04). (CI31) All our directors [...] are men that have been more than 25 years with us. [...] they are very committed and always much aligned. (I02). It is important to emphasize that the work of nonfamliy managers is positively evaluated by the family members who work in the company, who state that it is an advantage to maintain distance between them and possible family problems, which would affect directly the managers belonging to the family. In this sense, at present, of the seven directorships, only three are held by family managers. It is valid, however, to stress that this panorama can suffer alterations due to the interest in management positions shown by the 3rd generation, whose members are already in the majority concluding their higher education and looking for their space in the work market. (CI32) Our company is a family managed one, but this does not prevent executives who are not shareholders having the possibility of reaching an executive position. We have four directors that are not shareholders; we are not people who only think about their children and grandchildren. The company is for those who have the ability to exercise the function for which they are nominated. (I03). (CI33) Many people that are here today as directors, are not family members. They are extremely respected and fulfil an essential role. Nor does everyone from the family have the aptitude for the positions that the company offers. And perhaps the management of a person who is not from the family even avoids family conflicts. (I06). Respect for the hierarchy of the organization is stimulated by the family managers even in cases that involve individuals belonging to the family, placing the authority of nonfamliy managers above kinship ties. This standard of behaviour helps 69 to strengthen the clear separation between family questions and the organizational environment, in addition to avoiding the draining away of the authority accumulated by the nonfamliy managers. (CI34) To be in the company, to be of the family and have a hierarchical superior that is not from the family, I don´t have a problem with this. My boss is a nonfamliy director and I never go over his head. I always try my utmost to see that the decisions flow through him [...] And I get on very well with him. (I05). (CI35) So much so that my son when he came here to do an internship, we had a meeting, I, him and the [nonfamliy director] and I said: ‘look, I asked her [director] and she met my wishes with this internship. Now, you are here inside the company, it depends only and exclusively on you. If she arrives tomorrow and tells me that you are not interested in anything, it´s not going to work out. It´s over, my friend, I am not going to keep you here’. I cannot demand results from her with a team that I want and that she doesn´t. So this was clearly established in his head [...]. It is not because he is a son, a grandson that he will be favoured. I said to him: ‘you have come to work here; your boss is her over there, sort what is necessary out with her. Don´t involve me in your questions of work’. (I02). On the other hand, due to the double role concentrated in the figure of the family members, that is, of staff and also of future shareholders, it was possible to detect the existence of a differentiated treatment by some nonfamliy managers of company heirs. This fact does not cause great surprise, as the power relations rooted in this type of organization demand of nonfamliy executives the capacity to articulate politically with the potential successors who will substitute the present controllers as much in the field of property as of management. (CI36) Clearly they treat us differently in the company. It is a natural thing. The directors, even with me being in a position below them, they talk in a different way, only because tomorrow I shall be a company shareholder. (I05). (CI37) There is this... (laughter). I myself, if I could choose I would prefer to have working below me someone who is not the owner’s son. Even if they do know very well how to separate these things, not demanding privileges, we end up adopting a more reserved posture. It’s natural. (I04). IV.1.7. Family management, property and governance In agreeing to the entry of a new partner, John Harris, during the crisis period that the company was going through at the start of the 1990s, Mills and Haag, had to live with the a partner, that apparently did not share the same vision and values that 70 they had, which always brings with it the possibility of conflicts between majority and minority shareholders. This latent tension is evinced in the following quotation. (CI38) During the period 1985 to 1993, when we had the problem of the creditors' agreement, he was our customer (really he is a customer to this day), who loaned money to the company, at very high interest: a real moneylender. And when we saw that this was going to harm us a lot, we admitted him as a partner and today he has 25% of the voting capital, being part of our Board. (I01). Up to the present, Harris has behaved as passive partner, filling the role known as that of sleeping partner. Sleeping partner is one that invests money in a company, but does not involve himself in the daily routine. This type of partner can represent a useful source of short term funds, however, in the long term some conflicts can emerge as pressure builds up for the entry of members of his family block to take over managerial positions in the company. In addition, there can be pressure to increase the margin of annually distributed dividends on the result achieved by the company. This second point is especially relevant in the case of Milk & Candies, as, historically the controllers have used the company’s positive results for new investments to stimulate growth, always relegating the distribution of dividends to second place. Such problems, however, did not break out, but, as if anticipating this possible situation the majority controllers have already been ready to buy the shares belonging to the minority partner, but due to the continuous phase of growth that the company was passing through in recent years, this offer has been systematically declined. (CI39) He [John Harris] thinks that [Milk & Candies] is a guarantee of his future, of his wife and children. He is partner of a very certain business, very cohesive. Accordingly, his attitude is always one of not opposing anything, always following our opinion and that of [Haag], which are very much based on consensus. The effective decision is always from all of us. (I02). (CI40) [Mr. Haag] now and again says: ‘Ah! There is no dividend?’ But if there is an investment further down the line and the company will require cash, he is not greedy and agrees in cutting dividends. Now it is more difficult with [John Harris] because he is an investor. He entered the business, so he wants his return at least that guaranteed by law, 25%. (I02). (CI41) My children are working and for the time being I do not see the least need to think about them entering the company. If tomorrow or in the future it becomes necessary, then we can sit down and talk. But not now. (I07). 71 In relation to the corporate issues present in the Mills family, Edgar affirms that in the future he sees the creation of a family holding company, directing the members of the coming generations to fill positions on the board or simply exercising the role of company shareholders. This option would be a way of preventing positions in the company being disputed by future generations of the family, which, naturally are growing at a rate far superior to the company’s capacity to absorb them. In addition, in the absence of the current CEO, the risk is eliminated that family ties are utilized as a criterion for contracting executives, in detriment to the level of competence required for the function. Among the present initiatives which are in harmony with this objective, the participation of various Mills family members in a programme for the development of shareholders can be singled out, which seeks to prepare the coming generations to carry out this function satisfactorily. (CI42) I have a project for a family holding, because there are no ‘jobs for the boys’ here. If not, I would be prejudicing the company. What will they be? They will be partners of the holding company and will have shares of the firm. Thus, no-one will be automatically an employee. […] Today they are grandchildren, in a short while they will be great-grandchildren. And after that comes daughter-in-law, sons-in-law. (I01). Regarding the maintenance of the company under the control of the family, the existence of three family blocks has to be taken into consideration, each with its own conditions and interests. Obviously the greatest stake is that of the Mills family, if only because of the already mentioned desire of the third generation in making their careers within the company. In the case of the Haag family, Friedrich has two daughters and no grandchildren, which, in the future could signify a change of this interest as regards the maintenance of share participation. The fact of them living in Germany leads one to believe that succession in the Haag family will be limited to participation of the successors only on the Board of Directors. (CI43) On the part of the family [Haag] my daughters evidently will have more the function of board members, of watching as the company develops and taking care of their patrimony. (I03). (CI44) For me there is a very big question mark [...]. Today [Mr. Haag] is over 75, so if there is some problem with him, I do not know what the course of the daughters will be as regards the company [...]. What I perceive of him today is that he is concerned with integrating his daughters with the business. One year one comes, another year another comes to Brazil. Formerly they were not very linked, but now they seem more interested. One of them already is even a board member. (I05). 72 (CI45) Our intention is to maintain control in the present families. But one never knows what will happen in the country. I always keep my mind open to other possibilities, although not in the form of concrete plans. But the intention is to continue as a family company, as it is now. (I03). (CI46) We have already had offers of acquisition. And they were definitely not rejected. We have a business to negotiate. They make their due diligence and I make my valuation. After that, I multiply by five. If they accept it, we have a deal. (laughter). (I01). In order to regulate all the issues that surround the property of the company a shareholders' agreement has already been developed and signed between the partners and main heirs of the business. However, as the informal relationships and the ties of trust between the controllers of the 1 st generation prevail, who still account for the company shares, the agreement is perceived as an instrument intended to become effective only for the coming generations. (CI47) I see this shareholder agreement merely as formalization and which will be much more useful for the coming generations. Because, while [Mr. Edgar] and [Mr. Haag] are talking, the word will be much stronger than any formal document. So, when we drew up the shareholder agreement, those who were working here, we all signed it, with the intention of defending the company patrimony. (I02). In relation to the governance structure, Milk & Candies possesses a Board of Directors that functions only in a pro forma way. Because it is a closed capital corporation, at the moment with only three distinct shareholders – the two majority shareholders demonstrating a high degree of alignment – in the view of the controllers, a more structured and active board is unnecessary. The Milk & Candies board is at present composed of six members, who are:  Friedrich Haag (chairman);  Edmond Mills (Vice-president);  Gemma Haag (Member and daughter of Mr. Haag);  Alexander Mills (Member and brother of Edmond);  John Harris (Member and minority partner);  Evan Smith (Member and longest serving employee in the company). Haag gave voice explicitly to his belief that an active board is necessary only for larger companies and that the reigning informality is even an advantage because it confers greater agility on the decision taking. 73 (CI48) What occurs is that because of family questions I am here in Germany and I am leaving the executive part for [Edgar], for his colleagues and sons. But we are permanently in contact by telephone, and e-mail, and naturally personally. Depending on the problems I go to Brazil, or we meet in Paris to discuss them, mainly the most important matters of strategy. (I03). (CI49) We are extremely informal. At times, we decide things by phone and afterwards draw up the minutes. Now we have the possibility of having electronic meetings. We didn´t even have this before. (I01). (CI50) Speed of reply. I think that our great trump card is this. (I02). (CI51) [The decisions are] very fast. It is one of the advantages of working in [Milk & Candies]. Rapid and with quite a lot of autonomy. You do not have to continually be asking for a blessing. When the directors trust you they give you autonomy [...]. They look at the results. They do not pester you every day. It is a company that responds very rapidly. (I04). Despite the perception of the suitability of the governance model in force in Milk & Candies, where the two main shareholders are the real strategic decision takers, there is a concern demonstrated by the probable successor, where a formal board may be an important element for his legitimization in the absence of his father, and may help in possible disagreements with the other family managers. (CI52) I think that we will have to be more methodical and work accordingly. Because, before, the company was much smaller, making it easier to control the business. But from three years ago up to now, we went from four directors to seven. [...]. It is not so easy to manage the company as it was. We will have to improve the governance of our board, still more faced with [Edgar]’s possible departure of As he is the founder, there exists very great respect, but when I sit in his chair, what will it be like? My brother is another director and I will be the CEO. Will he respect my position? Everything will have to be negotiated so that there does not exist any doubt. (I02). In spite of his physical distancing, in opting to return to his country of origin, Haag insists in keeping close to the reality of the company, reinforcing the bonds existing between the two controlling families. (CI53) [Mr. Haag] is not involved in the daily routine. He lives in Germany, but whenever he can, he is here [...] And his relationship with the family is very strong. He always comes for birthdays, family parties. I think this is very important, as it keeps the ties tight. (I06). (CI54) I accompany everything: the invoicing, the financial data, the launching of new products. But I am also watching even the social part, the parties with staff. 74 IV.2. The Bental Case IV.2.1. The company profile Bental is a family controlled, closed capital corporation, with a professional management, active in different economic segments, and established in various Brazilian states. It takes this name in honour of its founder, as a result of joining BENedito with TALma. Enjoying good financial health, in 2007 it reached a value of over US$1 billion for net income from operations, EBITDA of more than US$ 200 million, and CAPEX of around US$120 million. IV.2.2. Profile of the interviewees Table 4.4: Bental’s profile of the interviewees Code I8 I9 I10 I11 I12 I13 I14 I15 Name Antônio Luiz Talma Pedro Talma Maria Talma Silveira Margareth Talma Silveira Júlio Talma Kleber Barbosa Emerson Dias Giuseppe Mancini Position Chairman CEO Chair of the Family Council Commercial Assessor Board member HR Director Controller Board member Reason to be interviewed Member of the owner family acting as Chairman (2nd Gener.) Member of the owner family acting as CEO (3rd Generation) Member of the owner family not active in the company (3rd Gen.) Date 7/5/2008 Duration 1h34 min 7/5/2008 1h32 min 6/3/2008 1h05 min Member of the owner family active in the company (3rd Gen.) 5/3/2008 1h50 min Member of the owner family not active in the company (4th Gen.) Non-family member in senior management position Non-family member in senior management position Former non-family CEO and actual board member 7/5/2008 1h25 min 5/3/2008 1h27 min 19/3/2008 1h13 min 20/9/2008 1h08 min Source: Research data 75 IV.2.3. Stages in the development of the company It is possible to analyse Bental’s evolution divided into six distinct phases: (i) foundation, growth and diversification; (ii) time of crisis and start of professionalization stage; (iii) building trust in a contentious environment; (iv) transition phase with other nonfamliy CEOs; (v) the return of the family to management control and the future generations; (vi) establishment of formal structures of Governance. In Figure 4.4 below the time line that details these different phases can be visualized. Figure 4.4: Bental time line Entry of 3rd non-family 3rd CEO generation CEO takes over nd Entry of 2 non-family CEO Entry of 1st non-family CEO Entry of 2nd generation into the business Diversification phase Initial operation of Board of Directors Year of foundation Growth phase Financial crisis Source: Prepared by the company. (i) Foundation, growth and diversification The history of the group studied goes back to the enterprising activities of its founder, Benedito Talma, born at the beginning of the last century, son of Portuguese peasants. In search of better opportunities, his father decided to follow in the footsteps of his brothers and move to Brazil in 1914, leaving behind his wife and four children. As the eldest son and the ‘man of the house’, Benedito had to start working 76 while still young, to help in the family upkeep, without, however, abandoning his studies. The family was united again at the end of the 1 st World War, when the father returned to Portugal to fetch them. The whole family moved to Brazil, in 1919, when Benedito was then 12 years old. The start of the Talma family in Brazil was beset with difficulties. Settled on a farmstead in the interior of the country, they gained their livelihood from the cultivation of the land. Benedito, who had four other Brazilian brothers, had several occupations in his youth: he was a mason’s mate, a smith, a mechanic, a lorry driver and became quite proficient in the handling and maintenance of machinery used in rice cultivation and processing. In 1933, he started his first business enterprise with his father and brothers, acquiring a machine for the processing of rice. From 1941 on, he changed his field of activity, convinced that the initial company would not be profitable, and rented a petrol station, succeeding in expanding this business. Subsequently, in 1944, he decided to merge his company with that of his brothers creating Irmãos Talma Cia. Ltda. which commenced its operations in the automobile / lorry resale field. Business prospered and, ten years later, Benedito led the constitution of a new company, Bental Services a public services concession which was to become the most outstanding business for him and for the group he had created. Benedito’s main reason for this new undertaking was to provide the region with an essential service that it lacked, with the quality and ready availability necessary to meet the population’s needs. In 1959 and 1964 respectively, Nilton, the eldest son, and Antonio Luiz, the youngest entered the business, coinciding with a period of rapid business expansion through the acquisition of other companies. However, as a result of the government policy of nationalization of services in the field where they operated, launched at the start of the 1970s, the company was deprived of possibilities for growth through geographical expansion. Even more important than this was the fact that, Benedito and his sons, afraid that the licence would possibly not be renewed, started to seek out new investment opportunities, which led to a great diversification of the group. 77 In this period, it is important to note that in addition to the business success achieved, the essential principles and values that to this day permeate the family and the organization were laid down: (CII01) The principles and values of an organization are born with its founder, and we greatly venerate the figure of our founder. Not because of the success that he had or the material gain that he achieved. But for the principles and values that he preached his whole life. (I09). It is commonly said that the company ends up being a mirror of the family, and reflects, accordingly, the values that are the basis of the family group. The Talma family has always cultivated firmly-rooted values such as work, thrift and simplicity, values which have been transferred to succeeding generations and have also influenced the organizational culture. (CII02) The business culture is a catechism. It is an act of faith. […] The example has to come from the controlling shareholder. And this has much to do with the manner of being, of the family DNA, so to speak [...] The fundamental thing for the transfer of these values is our leadership by example. If you don’t have the CEO practising them, and the directors and managers as well, it will not happen (I09). Although this initial phase of growth and development was a period of establishing such principles and values of the founder, which have started to permeate the whole organization and be a reference for future generations, this period also became marked by difficulties related to the interface between the family and company dimensions. Benedito Talma was known as a rigid man in his treatment of employees and family, but at the same time, a person with a ‘soft heart’. He did not see the company predominantly from the economic perspective, thinking only of profit. Because of his childhood with its great financial restrictions, the emigration of uncles and father in the search for work outside the country of their birth, and the fact of taking over the responsibility for the other siblings so early, Benedito started to have a vision that the company should be a means of livelihood and a guarantee of jobs for the whole family. It should be remembered that being a Portuguese family, the first generation of the Talma’s in Brazil adhered to those typical Portuguese characteristics of attachment to tradition and concern for strengthening family ties. Such characteristics formed a suitable environment for the forceful presence of 78 Benedito’s altruism, leading him to contract family members to work in the company, independently of their capacity, competence or level of commitment. (CII03) Dad came from Portugal to find work. He got it into his head that he should generate jobs for the family to work, so that they were not unemployed. And every time you tend to extremes there are abuses. So there were people from the family who drew their salaries without working, or producing very little. Together with these, there were others who worked themselves to the bone to cover them. [...] My father [the founder] took his brothers-in-law who didn´t have a profession, didn´t have anything, they were simple people, and he gave them a position as managers. […] They didn´t want to work, didn´t want to do anything. They harmed the organization. (I08) Another point that deserves mention was the absence of formal mechanisms of control and management in the organization. The entrepreneurial profile of the founder up to a point was able to impose the objectivity and dynamism necessary for the growth of the business. However, allied to the lack of a distinction between the family and professional roles, the absence of a system of structured management led the company into difficulties. (CII04) At the start, the company was a family thing. Everyone entered it, thought they were the owners of it, and it was a mess. Imagine it! There were no rules. (I15) (ii) Time of crisis and start of professionalization stage Between 1987 and 1988, Bental went through a severe financial crisis, because of a phase of accelerated diversification, added to the altruistic practices and also to the lack of appropriate management mechanisms, as reported by one of the nonfamliy executives that experienced it. (CII05) [...] we went through a great crisis. We had diversified the group a lot [...] and we invested much money and we did not have the scale to obtain competitive prices. Therefore we sank. Suddenly we were up to our necks in debt. At this time we had 64 separated companies. We had a big debt and 13,500 employees working in the group. And we did not have a management model where the holding company had control, had power, had command and had rules. So each company did it its own way, the way it liked it. The great truth is that we were lost with the management of the group as a result, mainly, of this absence of having things formalized. (I13). 79 Additionally, Nilton, the eldest son and Benedito’s natural successor, whose support was seen as essential, principally in financial issues, had died in the previous decade, at the age of 42. Antonio Luiz who was the youngest of the three brothers (the sister in the middle, Neuza, went to study in the capital, got married and never involved herself with the family business) opted for a career of engineering and on finishing his studies, entered the company as Operational Director. His relationship with his father was marked by respect, but also by significant divergences. Both had forthright and explosive personalities, and adopted attitudes that were always firm, and which generated constant conflicts. Antonio Luiz was characterized by boldness and the will to expand the business, but Benedito was more restrained. The treatment of financial questions was a source of disagreement between them, especially after the loss of Nilton. Benedito’s succession, making Antonio Luiz the effective leader, did not happen in an official and direct way. The consequences of a CVA (cerebrovascular accident) gradually made it impossible for him to continue his functions. The year 1987 can be considered the turning point, when Antonio Luiz started to respond more effectively for Bental’s destiny. This transitional period, with the gradual withdrawal of the founder, the death of the elder brother, the stress of the financial manager, added to the financial crisis that had installed itself, was extremely delicate for Antonio Luiz. He recognized that his academic background in engineering was not sufficient to help him control a highly-diversified company with a high degree of indebtedness. (CII06) [...] when we had a financial crisis, [Nelson] was our financial manager and he was considered the wizard of finance [...] But he started to suffer from exhaustion. [...] then the doctor gave him thirty days rest. When the 30 days were up, he gave another 60. When the 60 were up, another 6 months. And I started to accumulate that financial position also. I was never very keen on it... I know how to spend money, better than count it. I detest the number after the decimal point; for that, there are specialists. […] This was a very difficult moment for me, also seeing my dad’s capacity draining. Then, we started to look for someone to help us. (I08). At this time, the greater part of the companies of the Bental group were not profitable, some of them being run by unqualified family members, without anything being done, out of respect for family relationships. There was clear evidence of the 80 presence of family altruism in Bental. The good intentions of the founder (in generating employment and income for his family) started to generate agency costs for the business, reflected in the opportunism of some of its members who were compromising the company patrimony. Faced with this scenario, Antonio Luiz, after two years at the head of the group, decided to hand over the executive direction to a market professional, from outside and not belonging to the family. After a long negotiating stage, Giuseppe Mancini was contracted, and entered the business establishing a strong, professional management and rigorous control. On his entry, Giuseppe presented a detailed plan of measures that should be taken for clearing up the financial side of the company and for the adoption of a highly professionalized management. In compensation, he negotiated his autonomy in the form of a contract, governed by a fine in the event of breach, proportional to the time of operation, demanding total freedom in decision making. (CII07) I prepared a plan in which I set out in detail everything I wanted to do. And the plan was approved. I asked: ‘I would like to be authorised to do this. These are my conditions for taking on the position of CEO. And if I am not successful you are authorised to dismiss me and I assume this commitment. But I would like total authority to be able to undertake this programme’. [Benedito and Antonio Luiz] signed this contract and I went into action. (I15). After taking over the position, Giuseppe Mancini concentrated on three fronts of action: making the structure of the group slimmer; creating suitable management systems and; defining clearly the role of the family concerning the business. In the view of one of the present directors, Mancini was able to carry out the necessary shock treatment, starting from the adoption of rigorous measures, to reverse the company’s delicate situation. (CII08) Was it a risk? Yes, it was. Because it meant handing over the management of the business to a stranger. I mean, he [Antonio Luiz] knew [Giuseppe], but he was an outsider as far as the family was concerned. So the command of the executive management left the family and went to a professional executive. And this executive arrived and did a lot. He closed companies, sold companies, he changed business, he established rules, he introduced accountability. He established planning rules, set up the strategic planning, created the budget meetings, he dismissed company presidents, he developed a programme for the withdrawal of personnel from the group (at the time we reduced the group from 13,500 people to 6,200), and he 81 dismissed family members who were in the command of companies or that had work positions in the company that he judged were not contributing anything. It was a very painful process; we had some crises even in the family. And at the moment that we were having these crises he took the contract and put it in front of everyone: ‘I have a contract that gives me powers for this. Either you comply with the contract or I go’. I will say this, it was painful for everyone, but I think it was the salvation of the group at the time. (I13). With regard to the scaling down of the holding company structure, Mancini reduced the Bental group from a universe of 64 companies to only 14 and managed to attain an average growth rate of 22% per year between 1992 and 1997. It is possible to infer that the good results were responsible for keeping him in the position, in spite of the manifestations existing against him, above all within the controlling family. This restructuring phase of the holding company is narrated by Mancini as a period of numerous tensions, although essential for the constitution of the new group, in accordance with the explicit statement in the following fragment. (CII09) The expansion process was over the limit. There were 64 companies, totally out of control. There was no structure to run 64 companies. [...] A small-mentality reigned in which the accounting had to be something ‘secret’. We had many difficulties even with the signed contract. There were continual disagreements, but everyone understood that it was the only way to set up the group. (I08). In relation to the creation of a new management system, the concept adopted by Mancini, which they call ‘the network company’, deserves special attention. To this end, results centres were set up, tied to an evaluation and bonus system for all the company staff. In addition to this, he maintained the focus on customer satisfaction, based on the belief that a high level of customer satisfaction would guarantee the maintenance of the market and, consequently, the productivity and cash flow essential to the business. (CII10) This is something that we introduced when I came in 1989. We introduced into the company the system of ‘Network Company’, which is a system of involving the people, considering them not as employees, not as numbers, but as associates. Involving the people in all the decisions of the company itself and of the group. [...]. You should treat the people as an integral part of the company. (I15). (CII11) A more participative system in the decision-making process where all associates have a bonus, have a variable remuneration tied to the result of the company. Then they participate in fact, because they are associates and not employees. He [Giuseppe] had a whole set of activities to introduce, 82 which we call a ‘Network Company’. Network because we function as small results centres and you are responsible for that result of your centre. In addition to your commitment to the result of your centre, you are also committed to the whole for you to receive your variable remuneration. (I11). The third point worked on by Mancini while he was at the head of the group was the question of the insertion of the family into the business. The posture adopted was an aggressive one, and involved the dismissal of various family members that held positions in the company. Initially, an attempt was made to improve the capacitation and formation for some family members that were managing companies in the group. However, in the event of refusal to adapt, they were immediately dismissed. (CII12) And at first we tried to re-educate these partners, we tried to introduce a group management model in those companies that were directed by these partners who were relations, but it did not work. There was much reluctance on their part. And the group ended up by selling its participation in these companies. The funny thing is that some relations who were dismissed, in the end recognized that it was necessary to do what was done. Others feel resentment to this day. (I13). It is possible to infer that such steps were only tolerated due to the moment of strong crisis faced by the group. In this sense, on being questioned about how the shareholder and former family CEO, Antonio Luiz, faced this phase of dismissals, one of the older directors of the group stated. (CII13) He knew that it was necessary to do this. The pill was bitter, but needed to be swallowed. And he suffered a lot. [...]. He came under much pressure from family and friends. Of the following type: ‘aren´t you the master any more in that company? Did you give your company away, did you give it to [Giuseppe]? Who calls the shots there? Is it you or him? But [Antonio Luiz] is a very intelligent person. He knew that what was being done was very painful but had to be done. And that he wouldn´t do it, first because of his big heart, and second because of the emotional involvement that he had with people. So it was very difficult for him, but not for [Giuseppe]. [Giuseppe] did not have this commitment to anyone. He had a commitment to the shareholders to change the group, to transform the group, to professionalize the group. (I13) Some interviewees highlight the exemplary effect of this large number of dismissals of family members orchestrated by Giuseppe Mancini, which gave him the merit of inculcating into the family members of the company the importance of adopting a selection process based exclusively on the criterion of competence, without any granting of privileges. 83 (CII14) Be in no doubt about it. He [Giuseppe] also made the family see the light. The other family members did not want to place their patrimony at risk. They are not going adventuring. They know that today if you don´t have competence, you can´t run a company. (I13) Although the criterion of competence has been accepted and afforded value by the controlling family, there is disagreement as regards the degree of radicalism defended by Mancini, who did not hesitate to remove all family members that worked in the organization. (CII15) He [Giuseppe] was extreme. And we had some clashes with him in relation to this. And in fact it was a lot of ‘stick’. He fired my grandfather’s brother. He fired my father. But they were family members who did not want to adjust themselves to the new management. They did not want to do any training, so they didn´t change in accordance with the business. Now, to go from that to closing the door to the whole family, that´s wrong. (I11) In the view of Antonio Luiz, a family member undergoing a selection process should not be privileged nor, at the same time, undervalued simply by belonging to the family. In other words, in his view, the process of entry into the organization should occur with family and nonfamliy professionals competing on equal terms, competence only being used as the classifying criterion. (CII16) And I started to analyse that situation... you have to professionalize the company. But professional, we all are. If you have a work card you’re a professional. This does not mean that you have competence. If you can have a person of the family competent and come to work, excellent! Now what you can´t do is getting someone who is incompetent and put him to work [...]. Competence doesn´t depend on blood. You may have a competent person within the family and you may have a competent person outside the family. So today, my concept of competence is totally free. I don´t label anybody, in this way, if you’re in the family, you’re no good. (I08) However, it is possible to perceive contradictory elements in the discourse of the interviewee in affirming that in the event of a ‘draw’ in the selective process, the kinship ties would end by assuming the function of a casting vote, so to speak, overturning the apparent impartiality of the process and favouring the family members. (CII17) Today one procures people with competence. Independently of colour, party, family or nonfamliy. Obviously, if you have two competitors, with the same level of competence, one from the family and another from outside, I think that nothing would be better than bringing this family member into the company. (I08) 84 In spite of evincing a certain predilection for family members, Antonio Luiz also admits that the mixture of roles can bring about difficulties for the family company, a position evident in the following section. (CII18) The professional executive, when he gets fed up or when he doesn´t turn out right, goes to another company. For the executive shareholder to change his company is impossible. There is no way for him to do it. So, it is a risk that needs to be well calculated by both parties. Because if it does not work out right, the consequences for the executive shareholder are greater than those for the professional executive. The professional executive will find another place in the market. And the shareholder executive will have to continue to live with the company as shareholder. This aspect I think is a little more complicated. (I08). The fact is that in Bental, starting from the 2nd generation criteria involving professional qualifications for entry of members into the company started to be utilized, restricting in part the entry of family members. From this time also criteria of competence and merit started to be used in the promotion of members, both for family as for nonfamliy members. The main criterion established by the family is consolidated professional competence. In parallel with the mass withdrawal of around twelve family members of the company, Mancini worked on the structuring of the holding company, defining the corporate participation of the different family blocks and developing a shareholders' agreement with clear rules on the entry and withdrawal of shareholders. This process was to be fundamental for the separation of the family and company spheres, maintaining the family members only as shareholders and/or on the Board, and the management in the hands of nonfamliy professionals. In Mancini’s view, keeping the controlling family’s concern focused on the property sphere, tied to a highly professionalized management was indispensable for the greater transparency of the group. (CII19) Each has his or her salary, there are the dividends, and there is absolutely no double dealing. And you know that it is completely prohibited, everything is extremely transparent. The taxman who knows us very well when he enters already knows that [Bental] is an absolutely transparent company. It is all very clean. I always say, it is a glass house, everything is transparent. No one appropriates anything improperly from the company; everybody is controlled including the chairman and all the shareholders. (I15) 85 (CII20) In 1989, we made a shareholders' agreement. This agreement was a little non-participative. When [Giuseppe] entered, he organized the structure of a holding company that did not yet exist. (I11) Figure 4.5 presents the current corporate structure of the Talma Group, comprising three family holding companies which control Bental and that correspond to the three different family blocks. Each block presents unequal capital concentrations that vary from 20.2% to 58.5%. Despite the fact that the family already has members up to the 5 th generation, the capital control is still somewhat concentrated. Figure 4.5: Corporate Structure of the Talma Group Talma 1 Talma 2 21.3 % Talma 3 58.5 % 20.2 % Family Holdings Raizes S/A 100 % Group Holding Bental S/A % = Total and voting capital Source: Prepared by the company. A factor that explains the differences in the share quotas of the three holding companies was the accumulation of shares owned by Antonio Luiz, due to the bonuses received during the time he worked direct with his father. (CII21) Initially, my brother did not work in [Bental] and had a business of his own, having worked here for a very short time. My sister and brother-in-law were never part of the business. And I have worked in this institution all my life. I never had another business. Every time there was a subscription of capital, I increased my participation, through my bonuses which were converted into company shares. When my father died, obviously his shares 86 were divided by three. We were three heirs. And the shares that I had, plus the shares that my father left me, gave me the share control. This was good because it brought about a certain security. A certain unity to avoid the dispersion of the group. (I08) If on the one hand the share concentration present in one of the blocks is capable of establishing a certain unity of command, imprinting greater objectivity on Board decisions, on the other hand, the difference of power share between the majority and the minority group brings us back to the discussion concerning agency theory, abundantly covered by the literature, in that the majority group can utilize its power to influence decisions to the benefit of its interests and to the detriment of those of the minority. This tension between majority and minority is latent in Bental. Starting from the analysis of the section (CII22) it is possible to see evidence that some strategic decisions relative to, for example, the diversification of the group were taken apparently by the imposition of the will of the majority and not starting from the construction of the consensus between the three blocks of shareholders. (CII22) The fact that the company is diversified is due simply to my headstrong uncle, who didn´t like selling anything. It is a profile completely different from mine. [...]. This is his, not the family’s. The group takes after him a lot. Because he is the majority and he is the man. (I11) (CII23) The corporate participation is not equal. You have one branch of the family that has most of the shares. So this part makes the rules. This has already generated conflicts. (I13) Understanding that the adoption of corporate governance mechanisms is a viable alternative for the regulation of agency theory, establishing rules focused on more transparent practices, on better accountability and fairer treatment for the minority shareholders; the Board of Directors becomes one of the main instances charged with this process. In spite of the existence of a board in Bental, its character, initially, was decidedly pro forma, and made it impossible for the minority to have an active voice in the organization. The evolution of the profile of the Board of Directors and the creation of a Family Council are pointed to as fundamental for the definition of roles and a greater harmonization of the relations between the different groups of shareholders. These two themes will be dealt with in detail in the next section. (CII24) When my uncle [Antonio Luiz] left the management he felt a need to set up a board. He set up a board, but it was an advisory board. It was a 87 board constituted just by friends. It was a rubber-stamping board, because they told him what he wanted to hear. (I11). (CII25) It was that board where the friends had a meeting with the chairman during the morning, followed afterwards by a good lunch, with a good bottle of Portuguese wine. Because it was a Portuguese family. And on that board, the friends, because of affinity, did not have the independence to go against the opinion of the chairman. (I09). (iii) Building trust in a contentious environment Mancini had been contracted initially for a period of five years, but after a year and a half of the contract running, he decided to open a discussion to check whether the owning family was satisfied with the results, presenting a positive balance sheet, with strong reduction in debt and great restructuring of the company management model. Mancini’s attitude had the purpose of establishing a relationship based more on trust than on the rigid rules of a contract. Typically, a trust that emerged from his capacity to deliver the expected results. This was sufficient to maintain Giuseppe Mancini at the head of the company for eight years, in spite of the conflicts still existing. Antonio Luiz recognized that (CII26) Conflicts always exist, every day and at all times. But the important thing is the trust in the CEO that was chosen to command the company (I09). The first years of Mancini’s term were especially difficult for Antonio Luiz. In addition to all the autonomy conceded contractually and used by him, in facing the presence of the family in the company, as already remarked, the fact of the inexistence of an established board led Antonio Luiz to remain without a formal role. Kleber relates a typical situation arising out of the difficulty of repositioning his corporate function, that: (CII27) There was no formal board, he stayed in his room. He kept an eye on things ... (I13). Antonio Luiz himself recognised that: (CII28) No longer making executive decisions, the withdrawal from the daily routine and in addition having a CEO in your place: my God, this is traumatic. I took a long time to adapt to this new situation. (I09). 88 In spite of the numerous advances achieved by Mancini in his management, growing conflicts flourished in the relationship between him and Antonio Luiz, and reached an unsustainable level. (CII29) [Giuseppe] went too far in not leaving a space for our Chairman. [Antonio Luiz] is an engineer, a man that was born within [his sector of activity]; he is one of the people that know more [of this area] in Brazil. He has his ideas, he is a machine of ideas. He is a person that wherever he goes he will invent things and [Giuseppe] developed a certain aversion for his ideas. He thought that if they came from him, the ideas couldn´t be any good’. [...] So as not to be run over, as frequently occurs in the family company when the CEO is from outside, he ended up by isolating himself too much. (I13). (CII30) [Giuseppe] did a lot of good in the company. He cleared up part of the debt, but he started to generate a certain conflict. He started to quarrel with my uncle [Antonio Luiz] even in the meetings of the Board of Directors. And this started to become very explicit. At times my uncle wanted something at the strategic level and [Giuseppe] did not. And they had epic arguments at the meetings. (I10). (CII31) He [Antonio Luiz] couldn´t keep quiet. And the executive that came in didn´t like justifying himself very much. [Mancini] liked doing things and at the end of the year show the result. And this led [Antonio Luiz] to suffer much. He dismissed his colleagues, he dismissed his relatives, and he was changing a pile of things that [Antonio Luiz] had created. And up to a certain point he did keep quiet but sometimes, he entered the fray. [...]. And, finally, they had a bad row, and they came to the conclusion that it was time to wind up the contract. (I13). However, even being removed from management, Mancini’s competence was recognized and he was immediately invited to join the Board, which had already been formally structured. Because of his long experience at the head of the group, Mancini is still today seen as having a different level of authority from the other board members: (CII32) He continues on the board. When there is some problem in the group [Giuseppe] is contacted to help to resolve it. He is always heard. When there is the need to take an important decision in the group [Giuseppe] is, much more that the other members, the first to be heard (I13). This fact indicates that, in spite of the traumas left by Mancini during the phase of the rupture, his work has been duly recognized and has generated bonds of trust with the family. 89 (iv) Transition phase with other nonfamliy CEOs After the withdrawal of Giuseppe Mancini from the management, Bental tried two more nonfamliy CEOs. The first of them, a market executive, Hans Mueller worked between 1998 and 2001, conducting the company to another financial crisis. Antonio Luiz comments: (CII33) He was reliable, but he did not manage to produce the results that we expected. He was withdrawn. He stayed a very short time... two years only. He was an extraordinary person, but he had come from a very big company, with a large staff, with many advisors. Perhaps, with access to a lot of resources. It didn´t work out right in a company from the interior, a simpler one. (I08). But Pedro Talma reveals that this decision of Antonio Luiz was not so easily arrived at. Convinced of the inadequateness of Mueller for the position, Pedro pressured his father, and Chairman of the Board, to substitute this nonfamliy CEO. In accordance with the evidence provided by the following quotation, the conflict between father and son reaches a limit situation where the successor threatens to leave the company if the father does not accept the dismissal of Mueller from the position. (CII34) We arrived at a situation where clearly the performance of this professional CEO was very poor. The company entered a terrible financial crisis and I found myself in a very acrimonious debate with our Chairman. He saying that he would not discard the executive that he had nominated, because that was the only executive that he had nominated and that he would absolutely not give way. This conflict even got to the point of placing the company at risk and my relationship with the chairman became intolerable [...] and led to a conflict between two generations: that of the successor and that of him who had been succeeded. It got to the point of me saying: OK then, I’m leaving the business! Because I have a professional life in front of me and I am not going to participate in a catastrophe’. This affected him profoundly and he came to the conclusion that he really would have to substitute Mueller. (I09) In the middle of this critical situation, Antonio Luiz accedes to his son’s request, terminating Mueller’s contract, in accordance with that related by one of the executives heard: (CII35) It was a disaster. The group only did not have a more serious problem because [Giuseppe] had left things very well structured. He had left his followers […] as disciples, with very strong management practices. [Hans] is an excellent businessman, but not a man of operational 90 management. So, it was almost three years of ostracism, after which his contract was finally rescinded (I13). It is important to stress that the rescission of the contract with Hans Mueller happened because of his inadequateness for the position and not because of factors such as lack of ethics or dishonesty. When questioned about opportunistic attitudes related to bad faith on the part of executives, the controlling family members were of one mind. (CII36) No. [Dismissals] linked to bad faith no, but linked to results and values. Bad faith on the part of a CEO no. A CEO’s bad faith we never had. (I11). (CII37) You do not have a CEO that is directing his company acting in bad faith. It is a position of trust. You were chosen and take over a position of confidence and the chap acts in bad faith? I’ve never heard of it (I08). (CII38) Developing business is an obligation. The executive that does not produce results you remove and put in another that does. Now this question, for example, corruption, lack of ethics, is not mentioned in here. Lack of respect. If the chap gives 10 reals to obtain business worth one million in the result, he will be sent packing because he gave the 10 reals. (I09). After Mueller’s dismissal, a career staff member in the company was invited to take over as CEO. Distinguished by a certain stability and little action the management of Rui Pacato extends up to 2005, when there was, once again, a return to family management. (v) The return of the family to management control and the future generations of a 3 rd In 2006, the family returned to the command of the management, in the figure generation successor, Pedro Talma. Pedro is an economist, with diverse professional experiences in other companies and specialization obtained abroad. In accordance with his declarations, he gained positions in the company little by little, thanks to his own merits. In this process of conquest by recognition, Pedro supported various company expansion and improvement projects, demonstrating his ability as an operations manager, up to the time when he was guided by Giuseppe Mancini to assume corporate responsibilities, as Pacato’s direct assistant, who had taken over the position of CEO. 91 The departure of Rui Pacato from the position of CEO consequently set off the succession process in Bental, leading to discussion in the Board of Directors regarding what would be the ideal profile of the new CEO. In this context, the debate about the possibility of contracting a family member to the position was rekindled. Although no formal succession plan existed, Pedro already possessed wide experience of working within the group, and emerged as the natural successor. (CII39) There was no defined plan of succession […]. Surely, this always was in the head of Dr. [Antonio Luiz]: the idea that [Pedro] could be the CEO of the group. This he manifested several times with me confidentially. [Pedro] was prepared in the following manner: he did an internship outside the group; than he did an MBA in the United States. He came later and worked for a time in [Bental] in the area of operations, as a manager. He was not even a director. Afterwards, he started to participate on the executive committee, to be able to see how an executive committee works, what the business management of the group was like. He stayed some 3 or 4 years as a special member of the executive committee, and as assistant to [Rui Pacato]. So this gave him a good background. Then, therefore, although a written plan had not been drawn up there was a logical sequence. Until he ended by being nominated CEO as his father wished. (I13). It is interesting to emphasize that the phase when Bental was under the command of nonfamliy managers was important not only in the sense of inculcating a greater level of professionalization into the business, but also in breaking with the personalistic and patrimonial traditions inherited from the founder. Accordingly, it is possible to infer that the sudden separation between the family and management realized by Mancini accelerated the restructuring of the group, avoiding potential conflicts between family managers and the other members of the controlling family. (CII40) If Dr. [Antonio Luiz] were to this day in the command of management and he were to pass it to the son now, I imagine, from the nature of the son and the way he is, this would work out over time, but it would be much more complicated, costly, painful and laborious for everybody [to succeed in] breaking with what needed to be broken with. It would take a decade. Now the fact of him having found the group, or participated let us say in an evolution of the group with a well-defined management model, with clear rules, with a defined mission, with values defined, and with modern management tools, [...] all this shortened a period of [...], let us say, of ideological disagreements with the father. (I13). In addition to the possible conflicts and disputes for power, the succession in family organizations is commonly marked by the need to legitimize the successor before the family and other stakeholders. Accordingly, a point that attracts one’s 92 attention in Bental relates to the immediate acceptance of Pedro Talma as the new CEO, both by the controlling family and by the other nonfamliy managers. As to the family, the fact of Pedro being the son of the controlling shareholder and of not existing any other family members interested in disputing the position, simplified the succession process. As his father, Antonio Luiz, states, in addition to not having any other potential successors in the family, Pedro’s only sister was never interested in working in the company. (CII41) This is not her line. She doesn´t like it. She likes dealing with art. She likes dealing with painting, she likes dealing with a million things. Except the daily routine of the company. (I09). The acceptance of the new family CEO by the other managers of the group was explained, by Pedro himself, as a result of the construction of his legitimacy during his time in the company. Thus, in addition to having good professional training, the fact of Pedro having passed successfully through various positions and having worked in different hierarchical levels within the group for more than five years, facilitated the gaining of the respect and recognition of his competence by everyone. (CII42) It was surprising. He is well liked within the company. People admire him. They like him. Because he was never a stuck up person, giving orders where he had no business to, never was a person to use power, never was an ostentatious person. [...]. Then this way of his, when he was nominated CEO, all I heard was praise. People saying things like: what a good thing!’. People congratulating, wishing success, people prepared to collaborate. (I13). (CII43) I have a history of preparation over a long period that I think is the ideal. First you have to expose the person to the external world. So that he has a taste of success and develops some self-esteem. And for him to have some self-esteem formed and a successful trajectory. […] The hour that he arrives at the company, he gains the respect of his colleagues. And through this trajectory of conquering different positions within the company, he will conquer his space within the organization. (I09). Another point stressed by Pedro would be his concern in not being seen as the ‘son of the boss’, but, as someone with the capacity to be part of the business. Accordingly, instead of simply imposing his formal authority, the ideal way for the successor would entail the conquest of his space through his work trajectory involving all the members of the organization. In the view of the interviewee, even the 93 prior success of the father is interpreted as an element capable of outshining the activity of the successor in the business. The figure of the person succeeded could represent an obstacle in imposing a comparison, a mark to be overtaken or at least equalled by the successor. (CII44) Generally, the executive shareholder has to conquer his or her space within the company. Because it is not the fact of being a shareholder or even the controller that will make him eligible up to an executive function. So the first thing is that if he does not carve out his space he does not have people’s respect. The second point is the following: the son of a successful father has to seek his own light. Therefore I think that I adopted a trajectory of carving out my space within the company as far as possible outside my father’s shadow. But always remembering that you have to use his experience and success, having him to guide you in the best possible way. (I09). In many family organizations the overlapping of roles, such as the father/succeeded and son/successor, can start friction or conflict situations due to the intromission of affective issues in professional matters. If family managers do not have the capacity to distinguish these two dimensions in a mature way, the confusion of roles can lead to an environment that oscillates between affection and criticism, or even between exigency and an excess of protection. In this case, the successor affirms that in spite of the ideological differences or disagreements between him and his father, the professional issues were always worked on without negative interferences in the family relationship. (CII45) There exists a very great mutual respect between us two. We have already had friction and serious arguments within the company, but, thank God, we have succeeded in leaving all this in here. We leave here and there has never been any resentment, any bad moods, nothing like this. The relationship of respect of father and son and admiration continues, independent of professional conflict or of professional arguments. (I09). (vi) Establishment of formal structures of Governance It is fundamental to emphasize that the idea is not defended here that an absolute separation between the affective and professional spheres in the context of a family company is possible. For this we would have to deny the complexity inherent in such organizations. Although the coexistence among the family members in the daily round of the company is connected in an indissoluble way to the roles and 94 issues present in the bosom of the family, it is believed that the invention of spaces and suitable communication channels is possible so that each type of demand is worked on in its proper context. That is, that the family questions are debated in the field of the family and business issues are restricted to the company environment, minimizing as much as possible the level of friction between these two spheres. Accordingly, among the mechanisms of governance existing, the structuring of a Family Council is precisely the most recommended instance for the creation of a zone of consensus, with the objective of harmonizing family interests with the needs of the company. At Bental, the creation of the Family Council occurred in parallel with the evolution of the Board of Directors and sought to educate the diverse family members in the satisfactory performance of their roles in the organization. (CII46) We needed to create rules of relationship between the shareholder and the company because this is a strong source of conflict: if the shareholders are constantly in disaccord, the business could be finished. We see this happen all around. Then I started to push forward to develop of a Family Council. We brought in a consultant, run a workshop and we have succeeded in writing our family constitution, which deals with the three parameters: family as family, family as shareholder and family as employee. It is a way of regulating this relationship. We wrote the constitution but to practise what was written was rather a complicated matter. (I09). (CII47) The Family Council was created to work on making the family conscious of the difference between being a working family, an investor family, a business and being a family as family, as relations. [...]. Then, to organize the family. To put in place rules, open the debate on the role of the family. Organize the communication with the family, what the family should not see. What should and should not be the family business. To discipline withdrawals, participation, dividends, these things. (I13). (CII48) I think the main the role of [a Family Council] is the integration of the family and of perpetuating and passing family values on to the coming generations. The family history is what really strengthens the ties. [...] And the family tie can increase the longevity of the company. [...] The family has to focus on this: in strengthening its ties. (I10). After the creation of the Family Council its members were elected and the mapping of what would be the expectations of the group of heirs carried out. This established the directives for saving the history of the family and the business; the perpetuation of the culture and the values of the group; the respect for the diversity of interests within the family, and the creation of clear rules for the entry of family members into the group management. The following step was the creation of training 95 programmes focused on the young heirs who would wish in the future to become group executives, as also on those who wanted to open their own business or develop their careers independently of the company. Another programme stressed as fundamental would be the development of the shareholders, covering the knowledge necessary for family members to develop the capacity to understand and interpret the data generated by the Board of Directors. Thus, it would be possible to level the knowledge between the family shareholders and develop them for the active exercise of their role as Board members, through the consolidation of a legitimate interface between the spheres of the owning family and of the management. (CII49) I believe that [the development of the family shareholders] is the principal role of the Family Council. Not principal, but the initial role. Because how can you really have a Family Council with people who can discuss and report issues to the Board of Directors without this preparation? I don´t believe it is possible. So we end up with the idea that this preparation should be, in my opinion, the starting point of the Family Council. (I12). However, in spite of the advances made by the Family Council, various obstacles were encountered. The main one, in the view of the interviewees, was the lack of commitment and interest on the part of the family in involving itself in its activities. (CII50) It is challenging for you to prepare a person for the role of shareholder, who very often does not have an interest in the company, does not have an interest in the business and has a parallel career. So, it is very difficult to contribute to the development of the life plan and preserve the company patrimony. It´s a bit like that, isn´t it? The heir thinks that he has to milk something from the company and not contribute to it. (I10). (CII51) Each [family holding] has a different dynamic [...]. The members of [Talma 3] don´t involve themselves much in the business, they don´t participate. They want the dividends and to do their own thing. [Talma 2], which is my uncle’s, is more involved. It´s him and my cousin. On the other hand my other cousin is not involved. And there in [Talma 1], I am the one most involved. And it is good this way. The involvement is small with regard to the business. So today, there are few who sweat. There is no family involvement of the type: ‘let´s get into this business, or let´s do that’. I don't see the family as an entrepreneurial family in this sense. There is no harmony in this sense. (I11). In section (CII51) it is evident that the growth of the family, allied to the different characteristics present in each family block, may contribute to the lack of harmony, and to the loosening of the family ties that generally maintain the unity of the controlling family in relation to the business. From the moment that three or more 96 generations live together in the daily routine of a company the phase denominated consortium of cousins starts (Gersick et al., 1997) marked by an increase in the level of complexity of the issues that surround the family, property and management spheres. Figure 4.6 below presents the current family configuration, demonstrating the complexity incurred by the growth of the family that has already reached the 5 th generation, representing a universe of 31 shareholders or heirs. This figure also shows the three family holdings (Talma 1, Talma 2 and Talma 3) with their respective members. The controlling family members interviewed are highlighted in green. Figure 4.6: Structure of the Talma Family in 2008 Source: Research data In parallel to the structuring of the Family Council there occurred the evolution of the Board of Directors that, gradually, became the more strategic and directive instance in the organization. At present, the Board of Directors is composed of 11 members: two being honorary members, including the CEO; 4 internal members and; 5 members who are independent. The present profile of the Board is highlighted by the current Chairman in the following section. 97 (CII52) Our Board of Directors is composed of three distinct teams. It has the family members, with representatives of the three family holding companies that have a statutory right to a seat on the board. Besides this, we have professional Board members. For example, [Giuseppe] is a board member to this day. And we also have members who are businessmen. The family members usually have a more passive approach. The professional Board members want to see the figures, want to see the third decimal point. And the businessmen want to know about the business, what is being done, what is the course... So I think that our board is very well balanced. (I09). In addition, the Board of Directors is being advised, since 2004, by three permanent committees: the Audit Committee, the Human Talent Committee and the Corporate Governance Committee - each with its own working dynamic. Other governance practices adopted and that impart greater transparency to the organization were the adoption of external audits and the opening and disclosure of information to shareholders and market by means of formal reports. According to the 2006 Bental Report, the functions of control are exercised by the work of an internal audit, an independent external audit and a statutory audit committee. The external audit is always carried out by a specialized and internationally recognized company. It evaluates the general level of risks of the organization and the level of compliance with laws and regulations. The present configuration of the Bental governance model can be visualized in Figure 4.7. Figure 4.7: Governance Model of the Bental Group in 2008 Family Council Shareholders Meeting Statutory Audit Committee Audit Committee Board of Directors Human Talents Committee Corporate Governance Committee External Audit CEO Internal Audit - - - - - Functional connection _____ Hierarchical connection Source: Prepared by the company. 98 In relation to the activities of the Board of Directors, its schedule is composed of four annual meetings with previously fixed agendas. The first meeting deals with the financial statements and the rendering of account of the internal and external audits. The second meeting is dedicated to the analysis of the first quarter, covering the performance rates and the results attained. The third meeting carries out a strategic analysis of the segments in which Bental is active, discussing the long term directives. In addition to this, issues relative to social responsibility are also addressed. Finally, at the fourth meeting of the Board, the focus is on annual performance and the approval of the budgeting for the following year. This meeting includes also the performance assessment of both CEO and Board members. These four meetings of the Board are given schematic form in Figure 4.8, presented below. Figure 4.8: Schedule of activities of Bental’s Board of Directors st 1 Board Meeting nd 2 Board Meeting rd 3 Board Meeting th 4 Board Meeting   Financial Statements External and Internal Audits   Performance in the First Quarter Human Talents   Results of 1st Semester / benchmarking Social Responsibility/Long Term Corporate Directives    Annual Performance/Budget for Following Year Long Term Plan Evaluation of the Board/Members/CEO Source: Prepared by the company. The effective participation of the Board of Directors in the Bental planning process reinforces the formality and structured nature of their cycle, as the description of the investments approval process demonstrates. 99 (CII53) At the August meeting, we discuss with the board of directors what are the strategies and directives of the operations and of the group. We take suggestions and the board approves or makes some requests for alterations or recommendations. This goes back to each of the companies and from then on they set up their investment plan. Each investment project is, therefore, discussed internally within each business unit. After this, it is brought to [Bental] in an event that is called an investment committee. We analyse investment by investment in terms of rate of return, pay-back, present value and alignment with corporate strategy. Then, after preliminary approval, these investment projects go back to the company that contemplates this in its long term plan and budgeting, checking its coherence. These investments go back again to the holding company for discussion of the final budgeting and of the long term plan. Being approved at the level of the holding company, they are then taken to the Board of Directors for final approval. (I14). It is important to emphasize that the present stage of the Bental Governance Model was not achieved without clashes, conflicts, errors and successes. It should be remembered that firstly just a pro forma board was created, in 1995, which was reformulated along the lines of the Board of Directors in 1998 and from then on, evolved in the direction of best governance practises. One of the principal challenges initially faced was the delimitation of the function of the Board of Directors, removing it from an operational focus towards a more strategic profile, in order to avoid power disputes between the executive management and the board. (CII54) I would say that there is a conflict of interests between the executive management, eager for power, and the board of directors which wants a hand in the operational side. This was our first phase and we tried to define very clear rules for good performance of the executives, of the board and of the shareholders. (I09). (CII55) There was a moment when the Chairman called the attention of all my executive team. Afterwards I had to say to him that he was going over my head and that as Chairman he should deal only with me. (I09). At present, some control and supervision mechanisms of the board over the management have already been created. The evaluation of the CEO would be based on the results achieved and on a corporate climate survey of the whole group. In their turn, the executives are evaluated at the meetings related to the performance of the company’s main areas. In this context, they would be questioned about the positive and negative results of their respective areas. 100 (CII56) The board evaluates the CEO by the results, by the performance of the business. There are not many ‘monologues’. We don´t use any sophisticated instrument of assessment. It is direct: how is the result doing? How is business doing? How is the indebtedness? What was the profit? What is the EBITDA.? This is the evaluation of the board. Also, the organizational climate is taken into account. These are the two relevant things: the result and the organizational climate. (I13). (CII57) The simple existence of a board educates the executives through knowing that there exists a higher body of governance where they, family member or not, will be evaluated. So each executive goes there and renders accounts to the board. (I11) The control exercised by the Board of Directors over the management of the group is pointed to as a source of tension between the members and the executives. Such fact occurs due to the active and even incisive posture of the board members in the demand for results in relation to the group managers. (CII58) My perception is that the executives do not like board members very much. Because they think that the board members keep sticking their noses in where they don´t understand anything, that they don´t know the real situation. And when you have a more active board, the board member opens his mouth. He says what he thinks and what he feels. He asks difficult questions. As a responsible party, he needs to know what is happening. And the executive doesn´t like explaining very much. (I13). In Pedro’s view, the current CEO, the presence of an active board is fundamental to the development of the organization. According to him, many companies in Brazil started to implement Boards of Directors merely with the intention of complying with a market exigency, with the intent only of transmitting an image of transparency to attract new investors. Thus, the primordial objective of the board would be frustrated, that is, a board consisting merely symbolically would not be capable of acting in the strategic plan of the organization, directing and regulating. (CII59) The companies are setting up Boards of Directors because the market requires it. [...] The company speaks like this: I am going to launch shares, the SEC requires me to have a Board’. Then it runs after a lot of big shots, and sets up a board full of ‘nobodies’ just with titles and says: ‘here it is it´s set up!’. It is more something to lull the market than really something to aggregate value. (I09). Although the Bental Board of Directors presents an interesting trajectory, towards an increasingly directive and less consultative role, it is necessary to consider that, due to the family character of the business, the board is also still in a 101 rather unprivileged position in relation to the power equation of the company. This question is made explicit when we analyse the content of the speech of the current Chairman and majority partner of the group. (CII60) You cannot let the board decide. At most you have to take options and in accordance with the experience of the board members they can say: ‘this option is better than the other’. (I09). (CII61) This is because in the end you know exactly that the Board of Directors’ decisions go to the shareholders assembly and it is the shareholders general meeting that decides, it is the majority of votes that decides. Because the majority of the votes belongs to the family [Talma]. Then in the end it is the family [Talma] that decides. The Board of Directors of a company that has the share control belonging to a family, [...] should always take into account before what is the opinion of the family to be able to discuss with it and try to convince per decision. (I15). From the analysis of section (CII60) it is possible to perceive that the advisory nature of the board still remains alive, even if residually, as the actual Chairman does not totally accept the premise that the strategic decisions should be taken within the ambit of the board, but that, in reality, the board ‘at most’ should signal, based on the experience of members, which of the alternatives under discussion would be ideal. In section (CII61) it is interesting to note the conscience of one of the members and exCEO of the group on the primacy of the family in relation to board decisions. In this sense, the board should always seek a consensus with the controlling family members, as the current level of independence would not allow the board to impose a decision without the consent of the family shareholders. However, an emblematical moment for Bental regarding the level of independence of the Board of Directors occurred from the clash between Antonio Luiz and the other board members. The board members decided that it would be important to carry out the sale of a Bental asset. In disagreeing, Antonio Luiz threatened to dismiss the whole Board if the decision were not rethought. In this way, in the case in question, the authority of the majority partner was evoked in order to brake a decision of the Board considered contrary to the interests of the shareholder. Pedro Talma relates: (CII62) There were terrible moments of tension... And the meeting of the Board where the sale of the subsidiary was being decided, it lasted from 8 in the morning to midday. And in the interval he [chairman and majority partner] 102 said that he would not let the sale proceed and that he would dismiss the whole board. Then I said: ‘so you dismiss the board because you have already used all your arguments and the board will not reconsider it. The board is decided to approve this sale. And you will put at risk not only the sale and the financial situation of the company but throw down the drain 10 years of corporate governance. Because from the moment that you do this [...] we are going to send a signal to the market that the one who calls the shots here is the shareholder and not the Board of Directors’. Then, thank God, he reflected and did not dismiss the board. [...] And it was really a proof of the independence of the board, which voted on a question running counter to the chairman’s position (I09). Finally, it is valid to emphasize that the Board of Directors has been gradually advancing towards a prominent position in relation to the strategic directions of Bental, being cited as an important element for the perpetuation and longevity of the company. (CII63) The role of the board is in the performance of the organization and its perpetuity. The board functions as a parameter of evaluation of the proposals of the management for their short, and long term plans of operational strategy. We have four meetings annually and we take to the board both operational problems and virtues, and strategic points to be discussed, approved and monitored by the board. Then, this situation of having someone that is looking at your performance, I think that is very important. (I14). In addition to the independence of the Board of Directors, another fundamental point in the indication of the presence of good practices of corporate governance is the level of transparency reigning. Companies that seek resources in the stock market for the funding of their expansion and development are obliged to adopt a posture of transparency. Bental, even being a closed capital company, values such principle highly, constantly concerned with the disclosure of information to the shareholders and to the market. (CII64) In terms of corporate governance, of transparency of information, we already do this as if we were a publicly traded company. (I09). (CII65) One of our values is transparency. This facilitates things, because our executive knows that he really has to render account. [...] We do not have anything to hide. The principle that information must flow well in the company is not new for us - we have already been practising this a long time. (I11). All the shareholders have privileged access to the information necessary to follow affairs. There are formal quarterly meetings with the family members, including 103 all the generations, where a formal presentation of the economic-financial balances and operational data is made, the forecasts of profits or losses, with the intent of awakening the interest of the new generations in the business where the family resources are allocated. Transparent practices also are adopted that permeate all executive levels, making it possible to follow the attainment or not of the results so as to identify their possible causes. (CII66) We have always been very transparent. By tradition, by culture. [...]. So much so that, for example, it is a characteristic of ours, at the meeting of the board of directors for the principal group executives to appear. They are invited to give information that may be necessary or to make some presentations. (I14). Finally, it is concluded that the development of the governance model and professionalization of Bental, in addition to the transfer of the command of the company from family to nonfamliy executives and afterwards returning to family management, show a case of extreme wealth, based not merely on the longevity of the organization, but also on the level of complexity of the obstacles overcome in more than five decades of history. In this sense, the descriptive analysis of this case had as its objective to map out the kernels of meaning present in the discourses of the interviewees, with respect to the role of governance and of its influence in the minimization of conflicts and the mediation of the succession processes and professionalization at the core of the company. Thus, the main elements emerging from the analysis of content were utilized to aid in the construction of the Family Business Agency Conflict Model, whose results will be discussed in the quantitative stage of this thesis. IV.3. The Prize Case IV.3.1. The firm Prize Construções e Empreendimentos was incorporated in 1959 with the objective of producing slabs for popular housing. Its founder, Richard Goldman, while still a young engineer, had participated in the building of the new capital, Brasília, at the end of the 1950s. As everything had to be erected quickly on that immense site, 104 he developed the concept of industrialization applied to building, which consisted basically in the prefabrication of components such as slabs, stairs and ramps. Inspired by this experience, Richard started his own company, a small industry that was a pioneer in the technology of prefabricated concrete parts and that, 60 years later, was one of the three largest Brazilian companies in the field. Established in Minas Gerais, a state of the Brazilian South-East, with an area of 100,000 m², the company has today a production capacity of 2,000 m³ of prefabricated parts per month, and can proudly point to activities which enjoy acceptance throughout Brazil. The Prize prefabrication system is a model of industrialization and optimization of building which utilizes reinforced concrete structural parts, manufactured in accordance with the works project. These parts are transported to the site and assembled directly from their transports in their definitive positions. The industrialization of civil construction has been very common in the industrial and commercial segments, allowing the building of shopping malls or industrial warehouses in a short space of time, but is still little used in the housing segment. To occupy this space, Prize opened recently a new work front consisting of the construction of residential buildings, not only as a supplier, but also as entrepreneur. The company has closed partnerships with three regional contractors, where Prize offers the building solution and the precast parts. Such a solution reduces the average time for the building of a residential block from a year to six months. This strategic option occurred due to the large repressed demand in Brazil. According to federal government statistics, there is a need to construct 1.7 million housing units a year, in the next five years, just to meet the existing deficit of around 8 million units. Table 4.5 presents the evolution and growth of Prize using parameters such as the number of employees and annual generated revenue, the company projecting invoicing of the order of 200 million reals for the year 2012. 105 Table 4.5: Prize’s Business Performance Evolution of the figures Number of employees Net profit (in thousands) Income (in millions) Sales (in millions) 2006 360 10% R$40 R$56 2007 500 10% R$60 R$70 2008 800 10% R$ 90 R$ 120 Source: Research data IV.3.2. Profile of the interviewees Table 4.6: Prize’s profile of the interviewees Code I16 Name Elton Goldman Position CEO Reason to be interviewed Owner(s) acting as CEO nd (2 Generation) I17 Kate Goldman I18 Neil Brooks I19 Amber Goldman John Goldman Adjunct Financing Director Operational Director Shareholder Owner(s) active in the nd company (2 Generation) Non-family member in senior management Owner(s) not active in the nd company (2 Generation) Member of the owner family not active in the company rd (3 Gen.) I20 Heir Date 5/5/2008 and 9/12/2008 9/5/2008 Duration 2h36 min 9/5/2008 1h13 min 6/6/2008 42 min 24/6/2008 49 min 1h20 min Source: Research data IV.3.3. Family issues in Prize One of the principal factors in understanding the family questions present in the Prize case is the succession. The current CEO of Prize, Elton Goldman, belonging to the 2nd generation of the controlling family, has a successful trajectory in the company, but one that was not without its difficulties and challenges. His formal link with the company commenced in 1982, when he started to work as a purchases assistant. From there he went through the costs, financial and commercial areas, in this way ranging over all sectors of the company. Elton affirms that one of the motives of his growing involvement with the family company, above all 106 from the end of 1980, arose out of the need to take over some functions due to the withdrawal of the founder, who at that moment was involved with the management of a new business. (CIII01) My father has this entrepreneurial nature, and in the 80’s he started up an undertaking in the area of information technology, which was Quartzil Informática. It was a business that was much larger than he had anticipated and which ended up consuming a lot of his time and took his mind completely off Prize. This was the end of the 80’s and I was already in Prize at this time. Therefore, naturally, I could see this situation, his absence, and increasingly focused on resolving the problems of Quartzil. And I felt an obligation to take on the questions of Prize. So my involvement was very intense, but also very spontaneous. (I16). Another important question was the health condition of the founder, who, stricken with Parkinson's disease, gradually had to withdraw from the daily routine of the company. (CIII02) And soon after, at the beginning of the 1990s, he started to have some health problems that for a time greatly diverted his attention from work. This upset things a lot and affected us significantly. (I16). It is important to stress that neither the insertion of the successor, nor even the areas of his activities in the family organization occurred through imposition, or any pressure on the part of the founder. The successor stresses that since childhood he had developed a spontaneous interest in the family company, in observing the father’s routine and his dedication to the business. (CIII03) I was always very involved since childhood, always accompanying my father. Work always held an important place in his life. So I also became involved very early on. [...] I don´t think he used his paternal authority for me to take over a determined position within the company. This sort of thing: ‘From now on you will remain responsible for such and such an area’. He never said anything like this! He never told me where I was going to work in the company. Did he observe me working? Of course he did! And certainly without me noticing it, he must have given some guidance, but it was all very much my own initiative, to go on filling spaces (I16). In spite of having started his career in the company and the withdrawal of the founder not having been an abrupt change, the successor faced several difficulties to legitimize himself with the older directors and other company staff. The professional respect and charisma concentrated in the figure of the founder made it a symbol for the company and its staff. It was thus a hard task to substitute him who had been 107 responsible for the foundation and growth of the company, and resulted in the loyalty of many workers changing into mistrust and lack of identification with the successor. (CIII04) Elton’s yoke is a heavy one. One of the reasons is this: you still have factory staff who would give their lives for Dr. Richard. This is very fine, but at the same time it complicates things! For example, sometimes, and I am not saying this happens, an employee can subvert the authority of the successor, because he is still linked to the figure of the predecessor. [...]. I think people have to realize that organizations must be lasting, but that everything undergoes change. And that we, human beings and professionals, have our expiry dates. (I18). Another factor presented as an obstacle by Elton was the fact of being regarded by many staff as the ‘boss’s son’, who was there merely because of his kinship ties and not because of his competence to execute the work. While this was a factor of great discomfort for the successor, he considers it one of the motives for his persistence in carving out his space and demonstrating his value to the company. In the words of one of the present directors, from the second generation and Elton’s sister, the gradual transition of power from the predecessor to the successor did not facilitate the process: (CIII05) It really didn´t make things easier! Because for a long time Elton worked with him [founder], but he was very young, and was seen by a large number as a boy, and a boy who was even a nuisance there. (I17). (CIII06) I really found things very difficult, because this business of being the ‘boss’s son’ is terrible. It is a heavy burden. And with me it was no different. So when I had a function within the company, the company had already existed for many years. There was already a group there who was at the head of the business, manager, director, salespeople... [...]. I came to work and they said: ‘The club dues are all paid up, go to the club’ or ‘What are you doing here? Go for a walk, go and have a swim...’. It was more or less like this! [...]. But I was also a person who didn´t give up easily. So I wanted to impose myself and to be recognised. (I16). From the middle of the 1990s, Elton was already at the head of Prize, having to live up to the image and brilliant performance of his father in the company, and having to stamp on it his own style of management. And, in addition, prove that he was competent enough to run the business. Before his status was legitimized by employees and executives, there were some conflicts. Elton came up against several instances of resistance from the old executives; men who had worked in the company for many years and that were absolutely trusted by the founder, and who 108 seemed not to believe in the competence of the heir to manage the family business. There was an atmosphere of a ‘veiled boycott’ of the successor’s management. Facing up to this scenario, Elton started to dismiss some executives that belonged to the former management group formed by his father, in order to recycle the team, bringing in new members that he trusted, and at the same time removing staff who were out-of-date or too set in their ways. (CIII07) They were people who had been contracted by my father and who had been in the company for 30 years and that ended up almost becoming part of the family. Then there came a new management, with the need for a lot of changes. And they were not fitted for these changes. They would even contaminate and jeopardize the team that was in there. [...]. And it was not only one person: there were three executives that had exactly this profile from my father’s time. They were really necessary at that time, but with the passage of time they had become outdated, they didn´t keep abreast of new developments, they had that defect of being too set in their ways. They had a very rigid and completely inflexible posture. They really had to leave the company. (I16). One of the most critical moments in this process was the withdrawal of one of the older directors of the company who had made his career at the side of the founder and that, because of this, felt he was the rightful heir to the position of CEO. In this way, the wrangling for power, waged implicitly between the successor and the former director, resulted in the reduction of Elton’s authority, and installed a conflict situation that ended with the dismissal of the former director, in a tense and quarrelsome process. It is interesting to note that this decision, in addition to a practical effect, that is, the resolution of the conflict, also came accompanied by a symbolic effect, that of a demonstration of force, making it clear to all members of the organization who really was at the head of the company, and helped to raise the selfconfidence and the level of authority of the successor. (CIII08) Some years ago, I think that this was a decisive moment for the growth trajectory of Prize, a focused, sustainable and planned one. It had much to do with a change that we made in our management. I removed a director who had already worked in the company for 30 years, the commercial director. [...] he merged his position of executive with acquired rights and was not contributing anymore, he was not aligned. [...] It was not easy to take this decision, nor was it easy afterwards to take the responsibility for taking it, but it subsequently proved to be the right one. This also ended by giving me much more self-confidence (I16). 109 (CIII09) He [the former director] thought that he had the right, that he was the principal executive of the company. The principal executive was [Elton]. [...] but the withdrawal of this person was not a friendly matter. This was not [Elton]’s intent, nor surely the intent of the family. We did not cultivate the objective of discord. It was only that time ran out... the expiry date arrived. We wanted to try another model. (I18). As can be seen, in relation to the old staff, the process of legitimization of the successor proved to be a contentious one. However, as to the family, such process occurred more naturally and spontaneously, even if, with some apprehension and fear. The relative tranquillity and ready acceptance of the successor by the other family members can be explained, at least in part, by the lack of other candidates to the succession in the bosom of family, interested in competing for the position. In addition to this, the long trajectory of Elton in the company was responsible for him acquiring experience, little by little, and being recognized by the family as the natural successor to the position of CEO. (CIII10) I think that in the beginning, [Elton] started when my father was still very active, so we didn´t worry, because Elton had his support there the whole time. But, as my father withdrew, he became unable to be there working all day, and with [Elton] taking over more and more, I confess that I really became a little concerned. If only because of his immaturity [...]. I think that with time he started to see the real situation. He committed many blunders, many, but I think also that my father pretended not to see, because [Elton] was the youngest. He was very patient during his learning process. But I think he learnt as he went along, like learning to ride a bicycle. You just do it. You fall, you get up, and then you just get it. (I19). (CIII11) I think that [Elton] managed to get respect for his proposals and for his strategic vision. He formed an executive group, who adhere to his ideas, I won´t say totally, but the vast majority... I, for example, today consider [Elton] a real leader. He conquered this, although it was not easy. And, this is very important, indeed: the company has to have one. (I17). At present, in addition to Elton, also connected to the management of the company is another family member belonging to the 2 nd generation, Kate Goldman, who acts as deputy financial director. Figure 4.9 below presents the company’s current organization chart, in which is highlighted, in green, the position of the two family members who occupy managerial positions and of the nonfamliy manager, who were interviewed. 110 Figure 4.9: Organization chart of Prize CEO Elton Goldman HR Adj. Fin. Director Kate Goldman Operational Director Businees Director Administrative / Finance Director Neil Brooks Adj. Technical Director Controller Debts Receivables Accountant Sales Budget Manager Manager Plant Manager Project Manager Civil Works Manager Source: Research data As to the 3rd generation, only a niece of the current CEO is connected to the company, doing an internship in the area of architecture. Although there already are members of the 3rd generation with the right age and preparation to take over positions in the company, there exists a certain resistance to their entry, clearly influenced by the vision of the founder, who always advocated the importance of the separation of company and family. Thus the principal orientation to the successors is to seek professional experience outside Prize, to only later try for a position with the organization. (CIII12) This has actually been discussed. We are very concerned about this. We do not want our relations, nephews and sons, to imagine that here is their natural place of work. We encourage precisely the contrary. Yes of course, Prize is a work option, but not the only one. What we want is that they have the best professional training they can get, that they acquire experience outside, and then bring it to Prize. (I16). (CIII13) I think that the family culture is not, and never has been that of making the company a haven for the whole family [...]. Well, now I think that our mentality is starting to change a little in connection with this. But up to very recently, the culture that was implanted by my father was followed to the letter. I think we even exaggerated a little [...] Starting fairly recently, we have seen that things are not exactly like that [...] obviously even a family member, if he is no good, is not going to be kept on in the company just because of that . (I17). On being questioned about the possibility of the creation of formal rules to regulate the entry of family members, the members of the second generation showed 111 that they were conscious of this necessity, underlining the importance of a process oriented towards competence and not family ties as a selection criterion. Regarding this question, the current CEO relates that he even stopped one of his nieces from taking over a position in the company as he saw that it was being done mainly because of kinship rather than using the formal contracting procedures. Such decision, in spite of demonstrating a vision of equanimity and a consciousness of the importance of understanding roles, also generated a certain friction among the family members connected to the company. (CIII14) [In relation to my niece], I felt a very great desire in her to come into Prize and in truth I was the one who barred it. I did this because she wanted it so much that she was already considering herself working in the company. But this had not been agreed with me, or with Kate. So it was a point of great friction, to the extent that she considered her wish frustrated. [...]. We are going to have to have the ability to create opportunity without creating at the same time any conflict of an operational nature. I am absolutely not against having relations working in Prize. I think that Prize is a company of a size which allows this. We have a quite professional management and the people who come to work have to be in harmony with this management. They will have bosses! (I16). (CIII15) I think that we have to establish these rules [of the entry of family members into the company] as quickly as possible. [...].because there can appear problems of the type: ‘I have a right, I would like to, I can’ The next generation has only a notion, very superficial knowledge about the company. First of all we have to present the company to them: this would be the first step. And afterwards, faced with the interest of each one, prepare them for one day being able to enter the company, as a professional person like any other. (I19). The current commercial director, nonfamliy, points out that the market itself obliges the company to orientate its selection process by the criterion of competence, as, otherwise, the contracting of unprepared family members could lead to a reduction of efficiency and result in losses for the company. This being the case, provided that the prerequisite of competence is taken into account, there would not be, in his opinion, any restriction related to the contracting of relations for managerial positions. (CIII16) I don´t think that we can have this reservation that just because you are a relation you can´t be admitted. I think that this is a thing that the market has as a premise and we have seen it here. This has nothing to do with it, the chap can be the son of the boss and be perfectly competent to take over the business, as he can be the owner himself and not have the competence to run it. So I don´t think we should mix these things up. People have to prepare themselves. If they want to be an entrepreneur, if they have a 112 commercial flair and professional competence, so much the better if they are a relation. But it can´t be by kinship: Kings, queens, princes, that’s not the way to do it! (I18). In the view of one of the members of the 3rd generation, who today is making a career in the financial market, totally independent of the family company, and who demonstrates the desire to one day enter it, it can be perceived how it has strongly internalized the question of the need for qualification, in declaring that being a member of the family even represents a great source of pressure and that a career in Prize should not be initiated by a managerial position, but from a trajectory similar to that of the current CEO, who started at the operational levels, climbing up the ladder until he reached management positions. (CIII17) I think that if you are part of the family, the responsibility of working in a family company is even greater. I definitely do not agree that being a family member opens doors. You are evaluated in a different way; you are more criticized than the others. And I think that for you to bear this greater responsibility you have to have experience. I think that I have to be more qualified, more prepared, with more experience of the market. [...]. I also think that if I really join Prize, I do not imagine myself starting immediately as a director. I think that you have to start learning: being introduced to the company; knowing the business; seeing what its organizational structure is like. And I do not expect taking over the business, asking for a position, from the beginning [...]. but I think that as I already have experience, I do not need to start from scratch. But neither do I expect to start above everyone. I don´t think that’s the way. You have to follow some steps just to get prepared, to be qualified. (I20). Another point that emerges as a factor relevant for the longevity of the company in the hands of the family, concerns the organizational values set up by the founder and the feeling of pride in the company shown in the interviews of the members of the three generations. These aspects, in spite of being subjective and intangible, are responsible for the interest of the family in providing continuity for the business and passing it over to the coming generations. (CIII18) I think that we are very attached to Prize. To this brand Prize, to this image Prize, because it was built up with great dedication. And I think that before these values became the fashion, ethics, transparency, we have always been like this, which came a lot from the personality, from the character of my father, his posture. [...] We were not there, but Prize was very much inside our house. What my father did, the concerns, the priorities. This was very present. So I think that it is a thing that even not being there within the company, ends up with us building together. (I19) 113 (CIII19) When you treat the business as your own, that was built by your grandfather with much sweat, an entrepreneur, you are reflected in him and you do not want this business to die. So, I don’t want it to die. I would like one day for my path to cross that of Prize, just to maintain this legacy that my grandfather gave us, that my uncle brought this far. And if it is not me, perhaps it will be my cousin, perhaps my sister or perhaps two or three together. (I20). IV.3.4. Professionalization of the company One of the noteworthy features, cited as a true stigma in many studies on family-owned companies (Gonçalves, 2000a), is the existence of practices arising out of the absence of a clear delimitation between company assets and family patrimony. That means, in many cases, it is common for the controlling family members to appropriate goods or resources of the company, utilizing them for private purposes and thus generating a real depletion in the company cash. In the case of Prize, one point that attracts one’s attention is the clear division between the family patrimony and the company resources, present since the founder’s management and passed with emphasis to subsequent generations. (CIII20) In our family, no-one shoots the traffic lights. This sort of thing does not exist in the company: ‘Ah, I’m going to use the company car. I’m going to ask a staff member to do something for me.’ This liberty does not exist [...]. Everything is separated. When it was only my father, the business was already not mixed up in this way. ‘Ah, so-and-so coming round to do a little something for me, or I´m going there to get this and that, I’ll ask for an advance’. This does not exist. [...]. Naturally he passed this on to us. So, this fear of it becoming an informal business, where everyone does what he likes, I don't see this at all. (I19). As regards the professionalization of Prize, it can be seen that in the presence of the founder there reigned a high level of informality in the management practices, a lack of delegation. And in spite of occurring to a lesser degree, these aspects are still present in the management of his successor. (CIII21) There was little formalization of things in the company, just because my father participated in the whole process. The company was small. Therefore, he himself spoke, he calculated, checked, sold, did the marketing, he did everything in the company. And logically that has merit, but as the company grows, it becomes impossible to manage. And I think that he had to adjust himself to this. For a long time, delegating functions, delegating responsibilities was a difficult thing for my father as well as for [Elton]. (I17). 114 (CIII22) I think that today, [Elton] is already more sure of himself as a CEO, and succeeds in delegating more, although he also started with that same profile as my father. [...]. he, for a long time, did not delegate. It was that personal thing, just him. I think that there has already been a lot of progress in his behaviour. (I19). Recently, Prize developed its strategic planning, in an attempt to draw up their objectives aligned to strategic guidelines, capable of involving the whole management team. In the view of the current commercial director, it is still necessary to make progress in the internal alignment, to raise the level of commitment of the collaborators and, in parallel, move in the direction of greater decentralization. (CIII23) We have held some seminars, contracted some specialists and we did some alignment work. We have also developed a strategic plan. This is something that went well. I think that [Elton] is showing much business maturity in seeking out advisors, in relinquishing the exercise of power alone, in asking the opinion of someone who can collaborate. This is important. (I18). In order to tie the strategic planning to a set of targets, an evaluation and incentives system was created based on the participation of the staff in the net profits of the company. Some of the adopted evaluation criteria are: gross margin; production goals; sales; revenue with the respective invoicing; level of customer satisfaction and; brand reputation. Although it represents a significant advance in the professionalization of the company, it was seen that the subjective character of the evaluation had to be minimized, starting from the development of a much more formal and objective performance measures. (CIII24) The first formal evaluation was done last year, through a movement of the company for profit distribution. All levels were evaluated. Even the directors were evaluated by [Elton]. I still think that we need to create better criteria. The evaluation is a little subjective. You have the parameters, but they need to be placed in a firmer rule. I don’t think that it is only the numbers that we have to evaluate. It should be the numbers, as well as, the behaviour. (I18). In order to attain the necessary balance and focus, the current CEO is trusting to the growth of the company as a trigger to carry out a continuous renewal of his team, looking for professionals in harmony with his business vision, and results orientated. (CIII25) Seeking the growth of the company is an interesting issue because it makes possible for you to form a team aligned with your strategic planning 115 and the challenges that you proposed. When the company grows you need people attuned to what you are proposing to do. And these people have to arrive committed to your challenge, your purpose. Therefore, as the company has grown I have found it easy to form this team and also of breaking down a lot of resistance. [...] So this dynamic of company growth I think that has been the vehicle that has enabled me to attain a management that reflects my style of running the business. (I16). IV.3.5. The formalization of the Board of Directors In addition to the discussion on the succession process and its peculiarities, and the trajectory of professionalization of Prize, a treatment of the Board of Directors as the principal instance of governance becomes necessary. At first, a Board of Directors was set up in which only the controlling family members participated, resulting in the discussions being conducted in an informal manner. Accordingly, the Board represented a pro forma instance, working only in a rubber stamping capacity, with all the decision-making power concentrated in the figure of the CEO. As things changed, Elton invited people of his inner circle to be part of the Board and, although still maintaining a much more advisory than directive character, the simple opening up to the recommendations and opinions of the external members already represents a significant step in Prize detaching itself from a personalist management concentrated merely in the vision of its executive family members. (CIII26) I think that the difference is that between wine and water. I think that the Board has become professional. Because a Board formed just with family members is very informal, with little requirements, with few targets. When the external members came in with their experience from other companies and other scenarios, I think that it became much more professional, and you can see this clearly through the improvement in the company performance. (I19). (CIII27) Well, I think that this is a first sign of the professionalization of the company. Because we know that when we have a long-standing, traditional, family company, with capital safeguarded by the family itself, you are a company that has an owner. Then the owner’s voice ends up being almost a rule or a law. And the formation of a Board with external members, this, in some way, helps a lot, because we have the obligation of discussing the positions of the owner with professional Board members. (I18). On being questioned regarding what should be the tendency of the Board – if it should remain consultative or move in the direction of a directive profile – the 116 interviewees asserted that the expectation is to build a directive and active Board, which will be done gradually with the maturing of the family shareholders, in the sense of accepting directives that up to a point could run counter to the family view of the business. (CIII28) This is the situation when you have a family company: you have management by reason and management by emotion. This is natural and we need to see this with great clarity and with great tranquillity. I have the consciousness of being the director of a company that has an owner. But when you set up a 100% professional board, soon the will of the Board will end up being different from that of the owner. And I think that this is a point that requires more developed maturity from the family of shareholders, which means living with the controversial, living with differences. And of having to accept, even against one’s own wishes, the consensus of personnel that are there to make some contribution to the company. (I18). (CIII29) This requires us to submit to opinions of others, because the company cannot live with this personalist management, and decision making process: the chairman wants it that way and that’s the end of the story. I think that this phase has already passed. (I19). The positive perception of the experience with the new Board is given as an incentive to implement in the future a formalized Family Council, making available a channel of communication open to all family members involved with the company. This instance would thus be charged with dealing with family questions, such as the policy of entry and opening up of opportunities for the coming generations, filling an important gap making sure the company endures, through legal instances and the search for consensus among the different family blocks. (CIII30) I think that the Board that we have already formed, now only tends to be institutionalized. The meetings will continue in the way in which they are happening with the right frequency. And I think that with time we will also start to think about moving towards a Family Council, a formal one, to insert the new generation and prepare them. Because I think that this is fundamental, something that hasn´t occurred with our generation. After the founder, we came in, but without much preparation, some even without any preparation whatever. So, I think that this will open the way to preparing this new generation right away, who are nephews, our sons and daughters. (I19). According to the interviewees, the profile of the external board members represented a fundamental aspect in making the experience of the new Board a success. They brought with them a new vision, based on their vast market experience. Another positive point was the capacity of the Board to accelerate the 117 speed of the decision making, perceived as usually slow by one of the current nonfamliy directors. In addition, the possibility of benefiting from the support of the external members is given as a positive reinforcement, raising the level of confidence and allowing less hesitation in the definition of the paths to be followed by the company. (CIII31) There are three external board members, who have been contributing a lot. Firstly because they come with wider vision, not confined only to the company, a good vision from the market perspective [...] this brought an interesting evolution. I think that we were very fortunate, because we got a person who has a concrete batching business, which has to do with our business and a person from the financial area, which I think was a very interesting addition. (I17). (CIII32) We have members of great quality that have experience of this market that is new to us. We are facing an extremely globalized market which has characteristics different from that of 50 years ago. These board meetings are very productive to give focus, to give direction to the decision, because one of the characteristics of a family company is that you do not have the same speed in the decision taking of companies that are already 100% professionalized. Therefore, it is certainly very positive either for us, executives, or for the shareholders. Because their convictions or ours when corroborated by other people from outside, who have experience in the market, in the banking area, in the environment of big industries, brings greater certainty that the right path is being followed. (I18). Another gain with the structuring of governance has been the increase in the degree of understanding of the distinct roles, disciplining the activities of those who occupy more than one chair in the organization, which is the case of Elton, at the same time CEO and shareholder. This leads him to adopt more transparent practices in relation to the other shareholders. (CIII33) With the gradual withdrawal of my father, I increasingly took over the function of principal executive, because of the fact of being a shareholder and being very involved with the business. I nominated myself CEO, although I wasn´t. [...] really I only realized how much I was involving myself with the business much more than would be required by my responsibility as shareholder, when the family started to become interested in the business. When I saw that that business was not only mine. I had never thought of running Prize in a different way, as owner of only a part of it. I always ran it as if it were mine, as if it was my life there. And I think that was very important, after I became conscious that it belonged to the family, I tried to act more carefully. (I18). (CIII34) I think that we have made great strides on this question [regarding disclosure]. Due to the formation of the Shareholders Council and the Advisory Council, I think that there has been a great change, as today we have enormous transparency. Not that this generated conflicts before, but it 118 did generate a certain ill feeling... Because it was so natural for there to be no accountability for the family members within the company. This had a rather, let us say, heavy implication. Not everyone was in the company, so there had to be accountability. But today, this has all been peacefully settled. (I17). IV.3.6. The property sphere Finally, it is important to highlight some changes that occurred in the property sphere, influenced by the process of succession and by the greater formalization of the Prize governance mechanisms. In spite of being a company that has lasted for more than 50 years of activity, its corporate structure can be considered simple, being a closed capital corporation, composed at present of only four siblings, shareholders, belonging to the 2nd generation of the family, in accordance with that indicated in Table 4.7 following. Table 4.7: The Prize’s capital structure Shareholders % Capital Amber Goldman 25% Rose Goldman 25% Kate Goldman 25% Elton Goldman 25% Source: Research data Coexistence between the partners is seen as harmonious and well aligned. It is possible to infer that this, at least in part, is due to the reduced number of shareholders and to their kinship. Thus, taking as a reference the model of Gersick et al. (1997), we could classify the current phase of the company as a ‘company of brothers’, in which the potential for conflict is less than that commonly observed in more complex organizations, like those ‘consortiums of cousins’, distinguished by a greater number of generations occupying the property sphere together. Accordingly, although each partner possesses a degree of differentiated access to the daily routine of the company, the nonfamliy director relates that the alignment between the 119 partners is sufficiently satisfactory to guarantee that no disaffection interferes in the company management. (CIII35) I think that there certainly exists cohesion as regards the ends of the business, but with a differentiated involvement. We have [Elton] who is in charge of the operations, as CEO. There is [Kate] who is involved in the financial issues. There are the other two sisters who are a little more removed from the daily routine. And I don´t see any disharmony from the point of view of the shareholders regarding the main goals. There are clear objectives in the sense of trying to be a national reference regarding building solutions, of sustainable and increasing growth to create value, to attach value to the business, of seeking a different space in this market. We do not see any disagreements among the shareholders. Even if they existed, they didn’t get as far as here in the company. If they existed and didn’t get here, that’s already something to be thankful for, because it doesn’t create turmoil. But I do not even believe that they exist. (I18). Another point capable of playing an important role in the maintenance of an atmosphere of harmony between the partners is the figure of the founder and the values disseminated by him. Although he no longer exercises an active function in the company, due to bad health, he continues to enjoy enormous respect and represents the family’s moral foundations. (CIII36) I think that this [disagreements] doesn’t occur much because of the figure of my grandfather. I think although he doesn’t go into Prize anymore, he still makes everyone go more or less in the same direction. But, when my grandfather dies I don’t know. [...] I am definitely not saying that the day my grandfather dies the family will collapse, but I think that he is what has maintained it united to this day. He is that person over there, who today is very fragile with health problems and so forth, but who is the one that is an example for everybody. (I20). (CIII37) I think we are a very united family, but without that exaggeration: ‘every Sunday, we all have lunch together’. There’s not much of that. What I think keeps us together a lot are the values, passed down by our parents, by my father and by my mother: the example of dedication, of work, of something worth doing, doing it well, of wanting things to come out right. I think that this is the great unifying factor. (I19). Another element that emerges as relevant is the payment of dividends to the shareholders. Although there is no record of conflicts in relation to this question, divergent orientations can be seen between those that are merely shareholders and those that also occupy managerial positions. According to the vision of the partnermanagers the first concern should be to reinvest profits to maintain company growth, only secondly would the focus return to the remuneration of the shareholders. 120 (CIII38) The distribution of dividends is a demand from the shareholders not working in the company. It comes more from my two sisters, and I accept this as the way it should be. But, for those who are involved in the company, our main concern is always in reinvesting, and I am very conservative in relation to this. I think that the company comes first, before the family members’ needs. (I17). The question of return for the shareholders has been receiving greater attention on the part of the current management, because, due to the custom of passing over share quotas to the coming generations, new shareholders and new demands will arise, making the drawing up of the shareholders' agreement an interesting option for the regulation of the entry into and withdrawal from the company (via sale of shares). Accordingly, there already exists an expectation on the part of the CEO that the next generation will be prepared to take on the role of future shareholders. (CIII39) My dream really is to succeed in the professionalization of the company management. What I want is to be able to offer security for my children by creating, consolidating a company that is sustainable from the business point of view. One that generates results and dividends. For my kids, my expectation is that they can fulfil the function of shareholders of a good business. (I16). In relation to the possibility of going public or even of a future merger or sale, different reactions from different shareholders are encountered. For one of the shareholders the option of going public is accompanied by the sentiment of apprehension, above all because of the need for transparency and the loss of confidentiality of the organization’s operations. On the other hand, there is also the idea that these options would be an almost inevitable path for the company to follow in the future. (CIII40) I think that this is a natural path. Now I do not understand this subject profoundly [going public]. But, at first it scared me a bit. It’s the same old story: ‘This business is ceasing to be ours’. (I19). (CIII41) I really don´t think that we have any pre-defined idea of maintaining control indefinitely. I think it is a tendency for us to have partners and participate in another economic group, using Prize and vice-versa. I think that this is an almost natural path. And I think it is good. I feel the lack of it today, given this very competitive, rapid market, sometimes capital intensive, human resources intensive. Sometimes when I see some companies and say: ‘Goodness, if we could use that from that company it would be an enormous gain for Prize’. Therefore in this sense, as a means of increasing our competence, I think that this opportunity is very welcome. (I16). 121 (CIII42) If it is for the survival of the business, because although Prize is a family company, the business of Prize is to make a profit, to make money. The family itself also depends on this. So I would prefer seeing Prize among others, in a merger with another company, or Prize being passed to another administration, another management than see it going downhill, declining. That is, I don´t even see this scenario. But if this situation were to prevail and if I were a manager, I would definitely not be afraid to take this decision. I think that the company might be family and all the rest of it, but in spite of everything it is a company. And it has to be run to make money. (I20). Finally, it is interesting to emphasize the vision of the interviewees on what would be success for Prize. In this regard, replies were received related to results, to pride, to tradition, to confidence and honesty of its members and to the possibility of perpetuating family values starting from the business itself. (CIII43) Results, certainly. But this business, as it was set up didn’t produce the results of the best businesses in the market. Consequently it is not the results which are the main thing in the business. I think it has a lot to do with carrying on the tradition of a business created by Dr. Richard: an entrepreneur, innovative, persistent. A long-lasting business. And they take pride in this. They are very proud. I think of course that the desirable pride in money enters into the equation as well, but the pride in building up a lasting business sometimes exceeds even that. (I18). (CIII44) One thing that I like, that I see a lot of in my business today is the following: it is being able to say that those people over there are honest. This is the characteristic of a family company that I think is a point in favour. The people do not want to compromise their names or the name of their company, and I like this in Prize. (I20). (CIII45) And I think that what is more important is that the values themselves continue the same. In terms of strategy, the perspectives in relation to the market segment can change, new products being introduced. So what I think is important is that the values, they can remain over generations. And I think these values are very good and have to remain. (I17). IV.4. The Case of Minas Agro Group IV.4.1. The company The history of the Minas Agro Group is strongly linked to the enterprising activities of its founder Eurico Fontes, who started his career in 1976, when, at 23, and recently-graduated as an agricultural technician, he moved to the southern region of the state of Minas Gerais, Brazil. From then on he made a successful career in the sales division of a large multinational company. 122 Soon after, he entered as a minority partner in a company specializing in the warehousing and trading of cereals. A short while after his entry into this company, he proposed the diversification of the business, widening its range to include the production of fertilizers. In this context, in addition to maintaining its activity in the trading of rice and beans for the state of São Paulo, the company commenced production and trading of fertilizers for the whole region in which it was situated, characterized by an essentially agricultural economy. This company lasted for approximately 15 years, until, in the middle of 1992, it was wound up due to personal reasons alleged by the majority partner. At this moment, Eurico received for his 10% participation, one farm, one liquor distillery and, in addition, a reasonable amount of capital. Still in 1992, this financial amount was invested in the creation of a new fertilizer industry that was the germ of the Minas Agro Group, which structure can be seen in Figure 4.10. Figure 4.10: Enterprises of the Minas Agro Group Source: Research data After going through a long developmental phase, from 2002, Minas Agro established a strategic partnership with BUNGE fertilizers, a well-known multinational company in the Brazilian agribusiness field. In this new phase, the company 123 abandoned the production and the trading of fertilizers, concentrating only on the process of mixing the components of the fertilizers produced by BUNGE and on delivery of the final product to its destination. Inspired by his entrepreneurial spirit, Eurico allied the development of this undertaking to the creation of new businesses such as Minas Logistics, which was created solely to meet the demands of Minas Agro, but which, with the passage of time, gained space within the group. Thus, besides coping with the intense distribution movement of its own business, in off season periods in the fertilizers sector, the company started to offer services to new markets. It should also be pointed out that, together with the development of the fertilizer and transports businesses, Eurico also administered the distillery and the dairy cattle farm, occupying himself with a broad range of activities. The successful trajectory of Eurico Fontes was abruptly interrupted in 2006, when he was killed in a car accident, leaving the business group and the family abandoned. IV.4.2. Profile of the interviewees Table 4.8: Minas Agro’s profile of the interviewees Code I21 Name Position Heloísa Fontes Patrícia Fontes CEO I23 Vivian Fontes Not active in the company I24 Gabriela Fontes Administrative Director I25 Cristina Fontes Commercial advisor I26 Thomas Sampaio Board advisor I27 Débora Brandão Legal advisor I22 Finance Director Reason to be interviewed Date Founder’s wife and the present CEO Owner active in the nd company (2 Generation) Owner not active in nd the company (2 Generation) Owner active in the nd company (2 Generation) Founder’s sister in senior management 11/09/2008 1h30 min 11/09/2008 1h21 min 12/09/2008 43 min 11/09/2008 1h14 min 12/09/2008 1h00 min Non-family member in senior management Non-family member in senior management 12/09/2008 1h13 min 12/09/2008 1h01 min Source: Research data 124 Duration IV.4.3. Obstacles, clashes and challenges in the succession Without doubt, the most noteworthy aspect of this case revolves around the succession process, triggered by the sudden death of the founder. The victim of a car accident, Eurico Fontes was only 53 years old, and he left his widow and three daughters responsible for a business group whose activities and decisions had been concentrated solely in his person. In spite of two of the successors, Gabriela and Patrícia, having graduated in administration and of the desire of the founder of one day having them in the management of the business, the absence of a process for the preparation of the succession was responsible for the fact that they had not developed greater links with the company, and did not know the business. The other sister, Vivian, was fully occupied with her graduation in medicine, making her even more distant from the family business. (CIV1) The girls were no longer living with us. Gabriela, who is the eldest, had been in the USA for two years, where she was doing an MBA. Patricia, the youngest, had already lived there for six months, doing a course of English, but it was already settled that she would also return to continue her studies there. And Vivian, who is the one in the middle, was studying 4 medicine in Barbacena . (I21). (CIV2) This loss [...] brought much insecurity in its train, because [Eurico] was absolutely the centrepiece. He was very active. So everybody was unhinged. His daughters had never been here... They were very much removed from the company, and he also didn´t insist on it. [...] if I was his [Eurico] daughter I wouldn´t leave him alone. If they had stayed at least two years here in the company living together with him, they would have learnt the way. [...] so the two attitudes reinforced each other, he didn´t want it much and neither did they. (I25). The lack of succession planning in Minas Agro represents a typical case. Leone (2005) points this out as being a recurring trait in the context of Brazilian family-owned companies. In spite of the explicit interest on the part of the founder in preparing the daughters to take over the business, it is probable that the feeling existed that there was still a lot of time in which to plan the succession, causing certain complacency, and resulting in the absence of any formal planning. The positive side of all this is that the expectation of the entry of the daughters into the family business was preceded by the search for greater maturity and 4 Barbacena is a city in the state of Minas Gerais, in Brazil, which stands out as a center of education. 125 professional experience in the market, before entering Minas Agro, in accordance with that stated by one of the successors in the following section. 5 (CIV3) So I had some job opportunities in Belo Horizonte and he [the founder] was very pleased with this. He liked the fact that I was going to learn a little more, acquire a businessperson's viewpoint. Learn to have the discipline of an executive [...]. And he said: ‘I think that now is not the moment for you to return here. I think that you are still very green’. And went on: ‘you have to acquire more experience, broaden your perspectives. At the right time you will take over the administrative part of the company and I will go off to the farm’. (I22). But this distance existing between the successors and the daily routine of the company, on the occasion of the sudden death of the founder, brought with it, in addition to their natural hesitation and perplexity faced with the urgent need of providing continuity for the family business, an additional difficulty, which was their lack of connection to it. In this context, the emotional impact brought about by the loss of the founder, was added to their feeling of insecurity faced with the responsibility of taking over the management of a business group whose reality they knew nothing about. (CIV4) When the unexpected death happened, our aunt at the end of the wake said: ‘Now it’s up to you. Tomorrow we are going to be there together, we are going to pray with the staff and we are going to push this boat forward’. This strength of hers, at that moment, precisely on the day of the burial made a great impression on me and on my sisters. And I believe on my mother also. Because we were very shaken, but we hadn’t been able to think much. We went to bed and on the following day went to the company to work. We arrived here and there were a lot of people we had never seen. So, we got down to it, on a wing and a prayer, without thinking if it was going to come out right, or what could happen more into the future. It was a sudden succession. (IV22). Faced with such an unexpected situation, and Heloísa, the wife’s founder, not having had the preparation necessary to take over the company, the agreed decision was that Patricia and Gabriela would help their mother in running the business: (CIV5) However much she lacked expertise in fertilizers, alcohol, cattle, she knew what it was all about in general. She didn´t have the ability to run a company of this complexity, but we resolved to be near and also to find people who could help us. (IV24) (CIV6) I was living in the USA and was not going to return very soon. I was doing my MBA and was intending to work in some company there, to first get 5 Belo Horizonte is a Brazilian city, the state capital of Minas Gerais. 126 experience, and afterwards return, if I did return [...] but at the start the impact was very great and at no time did I think of leaving my mother and sister alone. I thought that I had to hold their hands and we would march forward together. (IV24). (CIV7) I had concluded a course of business English in Orlando. I returned, but was already looking at MBA courses to do in the USA and was also considering the possibility of going to China to brush up the Mandarin I was studying in Belo Horizonte. (IV25). The decision that Heloísa would remain at the head of the company was intended to calm all the staff because of the respect that she inspired, because of the correctness of her treatment of everyone and the devotion that she had always dispensed to philanthropic works: (CIV8) [...] her presence, is primordial, is essential, in here. She shows respect for people and commands respect from them, because she is seen as the matriarch, the founder, even if she has not previously had a role in the management. (I24). (CIV9) We set up a foundation to help needy patients with cancer, of which I was the first president. Here in the city we don´t have oncology treatment. Our patients have to go to larger centres to receive this treatment. And we, together with the local government authority help out with travelling expenses, the purchase of drugs, and free psychological assistance... Everything is done with volunteer work. And we raise money through telemarketing, donations and charitable dinners. (IV21). In addition to their lack of preparation as regards knowledge of the business itself, another factor that generated great difficulty for the successors was the absence of formal mechanisms of control and management. As Eurico had centralized the decisions and memorized all the information relative to the business, when he died he took it with him, leaving the owners completely in the dark. One of the solutions found by the Fontes family was to resort to hiring an adviser to help them in the company direction. Thomas was chosen and his initial mission was to carry out an investigation on the feasibility and financial health of each company pertaining to the Minas Agro Group. (CIV10) Our father concentrated many of the things in his head. He kept everything to himself as regards company matters, even with his own family. We did not participate directly. We knew some things, but we did not participate in the construction phases of his business. (I22). (CIV11) My father was a centralizer. He took care of everything: he ordered things to be done, but he kept on top of things to check. I think that there are 127 remains of this and the girls are trying to change it, to ensure that each employee takes more responsibility for his work. (I23). (CIV12) Because he [Eurico] was a very entrepreneurial person, intelligent, he kept many things in his head, so my initial task was to make an assessment of the economic /financial situation and a budget plan of all the six companies. (I26). (CIV13) He [Thomas] has been with us for two years. And they [my daughters] trust him enormously. He is very intelligent and is a person who likes new challenges. And he is always following the market in all senses. So thus, with him coming here, things started to become easier (I21). Heloísa’s trust in Thomas actually predates his entry into the company and had been established from ties built up at the charitable foundation that she was president of. (CIV14) In the Foundation there was a person who carried out volunteer work very well and who was our executive director. And I was looking for someone who could keep an eye on and orientate the girls [...] he had worked for 30 years in the Bank of Brazil and was now retired. [...] At the Foundation he did volunteer work, unselfishly, without earning anything. I could see the love that he had for what he was doing and in this he seemed a little like me. I believed that he could do a good job, would work and take this responsibility regarding my daughters [...] and it worked. (I21). In addition to Thomas, other long-serving staff also facilitated the process of settling down the new managers, sharing their experience and establishing a means of access to some strategic information left only verbally by the founder. (CIV15) I felt secure with her [Débora]. First, because she was a person who did not withhold information. On the contrary, she had an open attitude which led us to trust her work. [...]. She [Débora] must have been here 11 years. She worked directly with my father and was his right arm. [...]. So when we came in, she undertook the role of passing over much information which had been only in his head. She helped us a lot, doing some coaching work with us. (I24). In spite of the trust deposited in the nonfamliy managers at present in the company, when questioned about the possibility of contracting a nonfamliy CEO to run the business, the successors say that they feel insecure about this option. The lack of experience and the relative degree of ignorance of the handling of the business can have contributed to this attitude of insecurity, as without the existence of formal mechanisms of control and monitoring of the company results, the effective supervision of the acts of the contracted manager would be jeopardized. 128 (CIV16) We see difficulty in finding a reliable person. Just the fact of a person who is not from the family coming in, would cause an impact. Our insecurity has to do with the results that he would produce here in the company. We are studying this possibility, but we are not totally convinced that we are going to go for this idea [...]. For this to happen we need to be very secure, and today I really don´t feel comfortable with this idea, of putting in a person who will in practice take all the decisions of the company and map out the path for the business to follow. (I22). After taking over the Minas Agro group, the new managers sought, in addition to diagnosing the situation of each company, to promote the separation between the family patrimony and that of the organization. This helped directly in the elimination of abusive practices regarding company funds, avoiding the use of company assets for personal ends, and reducing the calls on the company cash arising out of the payment of family expenses. (CI17) As the business was all administered by him... he mixed up everything. The butcher’s bill was paid here in the company. It was everything in one sole cash. Today we are trying to separate this. [...] We want to form a group of companies with the family assets outside the business, totally disconnected. (I22). The separation of the family patrimony from that of the company is clearly a simpler process than the separation between professional subjects and family ones. The full understanding of the social roles, permitting each question, professional or personal, to be dealt with in its appropriate environment, is stated to have been a difficult process that demanded discipline and perseverance on the part of the new managers. (CIV18) At the start it seemed impossible to do it. Because work subjects were always dealt with at home and family subjects at work. Everything was mixed up and it did not seem possible to separate them. After several negotiations we succeeded in doing it. We no longer talked about work at home, and at the company we no longer talked about family matters. (I22). (CIV19) If we [the two sisters] had an argument at home for personal reasons, we brought it into the company. And here also we didn´t talk to each other, even if we had to decide on a work question. We went through an initial phase, behaving like this. But today we have learnt to deal with this. We are much better, through dialogue, and the interference of my mother. (I24). (CIV20) In the beginning when we were going to participate in important negotiations, my mother corrected me saying I had to speak in a different way. She poked me in the middle of the meeting for me to change my attitude. So all the time she said: ‘my dear, not like this’. And I totally lost my 129 credibility. So I was trying to speak properly about a certain matter and at the same time she was assuming the role of mother. I felt very small. (I22). In accordance with authors such as Dupas et al. (2003) it is important for the family to be able to develop a ‘zone of consensus’ with the purpose of minimizing the conflicts, directing the contentious questions towards amicable resolution. This notion that conflicts are part of a natural process of maturing, provided they are treated properly, is seen by one of the sisters, as a situation that was starting to exist in the Fontes family. (CIV21) The conflicts at home are making us grow more and more. [..] My sisters are very united. One is always drawing attention to the defects in the other and of themselves. Thus, one helps the other to grow. [...].Our family environment is very good. Of course it is not always absolutely marvellous: naturally there are conflicts, but by and large it is excellent. (I23). Another important question, implicit in the next quote (CIV22), relates to the building of the legitimacy of the successors. The distancing between the successors and the company, added to their youth at the moment of the succession, generated various difficulties, such as the low esteem by the body of the staff and the inevitable comparisons with the founder’s management. In this first stage, the search for legitimacy was accompanied by actions focused on the change of corporate culture rooted in the organization. (CIV22) When we started, it was very difficult to try to do something new, because the nonfamliy managers always used to say: ‘but in your father’s time we didn’t do things like this’. The search for our legitimacy was a constant struggle, as it was very difficult for them to believe that we were competent to succeed him. Their view was of: ‘two little spoilt girls, who lived abroad, travelling the world, want to come in here and pretend to be the bosses?’ So it was very complicated, we had to tread very warily (I24). (CIV23) It was very difficult for them. The first three months I was with them the whole time. As the girls were very young and considered spoilt, rejection was very pronounced. Staff members said: ‘I am not going to be told what to do by a few girls who are well brought up, but who know nothing’. It is difficult for you to change the way these people think and the rejection exists to this day. They are trying to change this, putting people they trust in. (I25). (CIV24) There are conflicts. Because they [Gabriela], [Vivian] and [Patrícia] are girls, are young. And they were seen by the staff as children. Only those children have grown up and there has been the loss of the father, the patriarch, but the staff has stayed the same. And the girls have started to command. So the first thing we worked on here with the girls was the 130 construction of their legitimacy. We worked eight months on a series of questions that marked this legitimacy. (I26). (CIV25) When we came in, there were many people that he had trusted and who felt secure with him. He was the one who said what had to be done or not. Everyone had a totally passive attitude, trusting that things would settle down from one hour to the next. This feeling of security that he had taken care of everything was gone. (I22). Another factor that influenced the process of building legitimacy of the successors concerns the question of gender. Authors such as Jimenez (2009) provide evidence that, in family organizations, the women, traditionally, assume roles disconnected from management such as those of wives, mothers and family leaders, limiting themselves to acting, in the majority of instances, only in the family environment. In the case in question, the reality is no different. Thus, in addition to the youth of the successors, the fact of being women is highlighted as an element that generated additional difficulties, until they succeeded in having their role of managers recognized by the body of the company staff. (CIV26) There still exists very great discrimination with relation to the role of women. When a woman is seen at the head of a business, I have already heard many times: ‘[this business] is not going to last long. In two years it will all be over’. Today things are already different, they are already saying: my goodness they are even changing things. It´s even working!’. I mean, I think that when a woman assumes a professional role, she tends to be better than the man. [...] in summary, the woman is more organized than the man in ideas and processes in general. (I27). The resistance of the staff, seen even at the management level, was translated into uncooperative attitudes. They ceased to share experiences and solutions and made decision-making on the part of the successors difficult, when they didn´t deliberately sabotage it. In some cases, such resistance changed into a clear posture of aversion, generating typical agency conflicts, which led to the withdrawal of four long-serving executives in the first two years of the new management. (CIV27) We had some difficulties with people in here. Even with a lady staff member who was from HR and who thought herself the owner of the company. She ruled the roost in here [...]. She also acted as administrative manager and was in conflict with the girls because as the girls assumed their roles, they started to call her decisions into question. (I21). The withdrawal of staff who were not capable of bringing themselves into line with the new style of management practised by the successors could have 131 contributed to the composition of a management group aligned to the vision of the new management, nourishing a climate of cooperation and commitment. Nevertheless, the loss of former middle managers, who dominated several fundamental processes in the company, is pointed to as one negative aspect brought about by the entry of the new managers into the business. (CIV28) After he [founder] died, the girls had difficulty in retaining the middle management who he had selected. And even difficulty in contracting new ones, even a little because of ignorance of the business itself. [...]. There have been significant losses. [...]. I think that at times patience has been lacking in analysing and accepting, even on the part of the partners who were coming in. Today I believe that they even feel, in part, the loss of some employee. [...]. The knowledge and length of service, knowledge of the business of this middle management are important in this transition [...]. (I27). Regarding governance, there is still no Board and no formal mechanisms in place. However, the recent creation of a committee of management, composed of Heloísa, Patrícia and Gabriela to deal with strategic subjects, can be considered an embryonic element for the construction of the governance structure of the Minas Agro group. The structuring of governance constitutes an essential stage with the purpose of arming the controlling family with elements for the overseeing and control of the group management. This option becomes especially important given the declared interest of the founder’s widow and current chairwoman of the group to assume gradually a counselling role. In addition to this, the possibility also exists, not entirely rejected by the successors, of eventually passing the management over to a nonfamliy professional and themselves occupying only seats on a Board of Directors. (CIV29) And initially we would want to create a board. [...]. After we discussed it a little, we decided to start with a committee. And we still don’t have external board members for it to be a board of directors. So, we are going to start like that, only with us three, a less ambitious format. And little by little we will build it up, making rules, defining the days of meetings, the manner of taking the decisions. So, all the important questions will be dealt with in this committee. (I24). (CIV30) We want to direct the important decisions, the more strategic ones and those concerning the direction of the business, the analysis of investment, of new businesses, the renegotiation of important agreements, and the formation of partnerships, to be dealt with in the committee of management. (I22). 132 In addition to mechanisms for the monitoring of management, governance also could help in the succession planning of the organization. The definition of who will eventually take over the position of CEO, after withdrawal of the matriarch of the business, can be made difficult by a possible dispute between the two successors who are already introduced into the business and who possess quite different profiles. In the view of Débora, the ascent in the hierarchy of any one of the two to the position of chairwoman, would lead to the need for changes in the arrangement of the current power equation in the company, and could lead to conflicts between the managers. In this sense, the formalization of a plan of succession could help in the definition of clear criteria for the choice of the new CEO, minimizing potential conflicts and friction between the family members involved with the company. (CIV31) One completes the other and I can’t see one with a position hierarchically superior to the other. How would the choice of one of them as chairwoman be? I think it would be difficult. [...]. There exist many conflicts among them. They complement each other, but with one acting in one area and the other in another, because if one interferes in the other’s area, there you have a problem: ‘You decide there and I decide here’. One of them is extremely financial and the other extremely human. (I27). The finding that the succession is not just a one-off event, but, above all, a process, is perceived by one of the successors who, after two years, does not consider that the succession process has come to an end. In this sense, in accordance with that pointed out by Lambrecht (2005), the succession does not terminate only by the transfer of the position of the person succeeded to the successors, but, considering it from a procedural perspective, it only becomes effective after the accumulation of experience and the full recognition of the successors as legitimate managers of the business. (CIV32) At the start we thought that this succession had already happened. But we simply took over as the managers of the business. And a little later, we perceived that the succession had still not happened. Because the founder, the founder’s wife, who has participated since the start of the construction of the company, she is still here in the company. So she is still part of the business [...] That is, the process is ongoing, and has still not being completed. (I22). Finally, it is valid to emphasize the interest demonstrated by the new managers in the maintenance of the property ownership and in its transmission to the coming generations. It can be seen that the values left by the founder and 133 disseminated both in the bosom of the family as also in the organizational context, have been capable of support the family cohesion and the identification of the successors with the business. The pride in the success of the company and the interest in preserving the memory of the founder through the continuity of the company in family hands, play a vital part in the perpetuation of the business and attaining longevity. (CIV33) It is a mixture of feelings. At the start, I felt more the question of not leaving them alone. But afterwards it seems that the business was getting into my blood. I started to see that I had the potential to work here, that there was something to conquer and something to improve. I arrived at the conclusion that I could make progress as a professional right here. [...] I started to see everything my father had constructed and that I should not let it fade away from one day to the next. And also it started to become part of my dream, when I saw that I could develop here. (I24). (CIV34) I don’t want the business to go right only for a while. If I have the ability and the competence of run the business giving it long life, for me that will be much greater success. And if my children one day participate in the business, this is something that will certainly give me great pleasure. (I22). 134 CHAPTER V THEORETICAL MODEL AND HYPOTHESES V.1. Building a theoretical model Theory can be defined as a systematic explanation of the relationship between a set of variables, or also as an explanation of a particular phenomenon (Kerlinger, 1980). For Strauss and Corbin ‘theory consists of plausible relationships produced among concepts and sets of concepts’ (1994, p.278). Gill and Johnson (1997), in turn, define theory as a formulation regarding the cause and effect relationships between two or more variables, which may or may not have been tested. Therefore, from these definitions, it can be stated that the theory seeks to relate events or concepts abstracted from the reality that surrounds man, in order to explain it in a systematic way. Theories can be developed inductively or deductively, or obtained by a combination of both arguments, as proposed by grounded theory. As seen, the grounded theory methodology is a field that aims to generate theoretical constructs that explain the action in the social context under study. The researcher seeks to identify processes that are happening in the social scene, from a series of hypotheses, which, linked to each other, can explain the phenomenon, combining inductive and deductive approaches. Strauss and Corbin (1990) say that all the procedures of grounded theory are designed to identify, develop and relate concepts to formulate the theory. After observations, the steps for the construction of the theory are: (i) collection of empirical data, (ii) coding or data analysis, (iii) definition of the theory. (Stern, 1980; Glaser & Strauss, 1967; Strauss & Corbin, 1990). (i) Collecting empirical data It has already been stated that the collection of empirical data within this research was conducted through semi-structured interviews involving actors selected 135 by the four companies studied and from them, specific case studies were built, having been presented in the previous chapter. (ii) Coding or data analysis The analytical process conducted in this stage aims to build theory, using the methodological rigor necessary to help researchers to identify concepts and discover categories from the data collected, besides suggesting propositions of how these categories are linked to form a theory (Strauss & Corbin, 1990). After collecting the empirical data, the researcher examined the transcripts of each interview, line by line, identifying and labelling the key concepts found with a word or sentence that expresses its meaning (Warburton, 2005). Concepts are recognized to be the basic units of analysis since it is from conceptualisation of data, not the ‘raw data’ per se, that theory is developed. ‘The incidents, events, happenings are taken as, or analysed as, potential indicators of phenomena, which are thereby given conceptual labels’ (Strauss & Corbin, 1990, p.7). Only by comparing incidents and naming like phenomena with the same term can the researcher gather the basic units for theory. The second step in the theory building through grounded theory is the creation of categories, which are wider and more abstract than the concepts they represent. Strauss and Corbin define them as ‘the cornerstones of developing theory [...] provid[ing] the means by which the theory can be integrated’ (1990, p.7). Bacellar (2005) suggests creating, initially, as many categories as possible, guaranteeing the coverage of all important aspects of the phenomenon in question. At the end of the process, however, there is the need to funnel them, finding common elements that may set up a theoretical framework. Indeed, a theoretical framework is formed from the proposition of generalized relationships between a category and its concepts and between discrete categories (Pandit, 1996). Glaser and Strauss (1967) have named this proposition of linking as 'hypotheses'. However, Whetten (1989) sees that the term 'propositions' is more appropriate since they involve conceptual relationships, whereas hypotheses require measured relationships. Pandit (1996) concludes that 136 since the grounded approach produces conceptual and not measured relationships, the former term is preferred. If, however, the proposed theoretical model becomes object of a subsequent quantitative research, as is the case that occurs in this thesis, the relationship between the constructs may be termed as hypotheses. In summary, data analysis involves breaking down, conceptualising, and reassembling it in a new way. These steps are conducted through three levels of coding process: open coding, axial coding and selective coding. The different phases of Strauss and Corbin’s grounded theory coding sequence could be illustrated as follows: Figure 5.1: The grounded theory analytic process Axial coding Open coding Concepts Categories Fracture & label the data: In vivo coding Classify concepts: Properties Dimensions Subcategories Drill down categories: When? Where? How? Why? Uncover relationships among categories: Mini-frameworks Conditions & Consequences The paradigm Selective coding Discover the ‘core’ category Develop theoretical framework Direction of analytic sequence Source: Adapted from Harwood (2002, cited in Warburton, 2005) Open coding The process of open coding is the first step in data classification. At this stage, the data collected undergo a continuous process of fracturing and labelling. Each 137 fractionated datum (incident, event or happening) is compared with another concept and the similar ones are grouped together and receive the same conceptual label. This process of grouping concepts at a higher, more abstract level is named categorising (Pandit, 1996; Warburton, 2005). Categories and subcategories can be described depending on the number of grouping levels it is desired to achieve at this stage. The entire open coding sequence could be illustrated as follows: Figure 5.2: Open coding sequence Conceptualising (abstracting) ‘Labelling’ ‘Fracturing the data’ Categorising Discover relationships between concepts Develop properties & dimensions ‘Properties’ are attributes of category. ‘Dimensions’ represent location of properties on a range Source: Strauss and Corbin (1998, cited in Warburton, 2005, p.8) Axial coding – creation and development of concept The second stage of data classification is axial coding, where there is a reorganization of information seeking to correlate similar open codes, summarizing them in their central idea. At this stage, a tentative conceptual framework is generated using the data as a reference. The researcher tries to discover the main problem in the social scene, from the standpoint of the actors or participants in the study and how they deal with the problem. Comparing all the data as they are received, the researcher makes a choice about the relative permanence of the problems in study. The reduction, the next stage in the procedure, is an inductive process of grouping the codes into categories. Here the categories already formed are comparatively analysed in the light of new data that arise, in order to try to identify 138 those related to the most significant. This process reduces the number of categories, as they become more organized. This grouping of categories is a theoretical form of analysis, where, as the integrations emerge, the grouped categories form others more general. The vital step in this process is to discover the central variable that best explains the action on the social scene (Cassiani et al., 1996). The axial coding is the means that helps the researcher to achieve the integration of the categories. The goal is to gather data by developing connections between the categories and subcategories (Pandit, 1996). Selective coding – modification and integration of the concept At this stage, I sought to establish the relationship of the core category chosen as the pillar of the emerging theory - with the other categories which validate it. The aim is to identify the causal and intervening conditions, context, consequences, covariances and contingencies that form the theoretical relationships in which the categories are related to each other and to the core category (Munhall, 2007). Following this procedure I was led to develop the theory. After this step of coding the data, one passes to the next phase which is the reduction of the categories. (iii) Delimitation of the theory The reduction of the categories is the way to delimit the emerging theory, when the researcher can discover consistency in the original group of categories and can then formulate the theory with a small group of concepts with a high degree of abstraction. The theoretical saturation of the categories occurs when: no new or relevant data emerge; the development category is dense; and the relationships between categories are well established and validated (Strauss & Corbin, 1990). 139 V.2. The use of Atlas.ti The software Atlas.ti was developed by Scientific Software Development to help in the qualitative analysis of data, through a variety of tools that make it possible to manage, extract, compare and examine significant data within a text. Most of the texts to be analysed are extracted from transcriptions of interviews carried out during the exploratory research, where investigation is required to formulate theories based on replies that offer insights into complex, and perhaps up to then, unknown, questions. Based on the analysis of the texts collected it is possible to establish hypotheses which can be objectives of the research investigation, and the Atlas.ti is, accordingly, a tool of great utility for the development of the grounded theory. It can organize the diverse information through key words listing the common characteristics. It also provides a general view and permits the researcher to check the critical points and those of greater importance in the conclusion of the theory. It is developed from the creation of a hermeneutic unit where the primary data are grouped, including all data collected during interviews and field notes. These data are then broken down into quotations (relevant stretches) that are given codes (two numbers: one representing the source document, where the quotation derives from, and the other the numerical order of the quotation in that document) and annotations forming schemata (graphical representations of the relations networks). All these elements are commented to facilitate subsequent understanding of their meaning. After finalization of this process, the programme generates two reports: one shows the history of the analysis and codification process and the other describes and comments the elements of the theory. As far as is known, in grounded theory generation, recurrent themes are identified in the data and coded as concepts. The used coding sequence used the openaxialselective series. It is important that the codes and categories that label emergent concepts are derived directly from the data. Authenticity is enhanced by the use of in vivo codes which are labels named from the actual verbatim words found in the data (Strauss & Corbin, 1998). 140 V.3. The process described Following the method professed by Strauss and Corbin (1990), the first step was to open code all transcriptions of each interview using the Atlas.ti. Then, the subcategories were determined that capture the main theme of each paragraph and identify the more general category to which they belong. After that, the axial and selective coding were carried out to establish causal relationships, describing the phenomena, their contexts, intervening conditions, action/interaction strategies, and consequences (Strauss & Corbin, 1990). During the theory development process few subcategories that were relevant to the emerging theory were identified which were not captured by the initial categories previously selected. To bring these subcategories into the theory, a return to the cases for an additional round of open coding was necessary. Using an iterative process, the new chosen subcategories were recoded within the remaining categories and incorporated into the theory through axial and selective coding. Table 5.1 gives an overview of all the relevant categories and subcategories contained in the proposed theory developed through a grounded process. In it, are presented the quotations relative to the interviews, containing the main elements that cause tensions in the four companies surveyed. Of course, the literature review that was intended to expand the knowledge of the researcher was, over the whole process, reaffirming the importance of each subcategory and suggesting others, according to studies conducted previously by other scholars, and providing new elements that were incorporated into the interview guide in subsequent interviews. After the development of the fourth case, I considered that the saturation point was achieved, as little new data seemed to emerge from the field research and the relationships between categories appeared to be well established. Although it is difficult to exactly measure when the condition of theoretical saturation is achieved, Table 5.1 also shows how frequently each category and subcategory occurred in the cases. Most of each subcategory relating to the theory appeared multiple times leaving enough evidence to support the emerging theory. 141 Table 5.1: Coding scheme for all relevant categories and subcategories and the correlated quotations 142 143 144 V.4. Proposed theoretical model To propose a theoretical model the key variables must be identified and described. Then, the linkages between these variables have to be proposed. And finally, the explanation about the reasons of the existence of such relationships has to be presented (Voss et al., 2002). Wacker (1998) confirms these steps proposing as a general procedure for theory building: define the variables; limit the domain; propose relationships (model building); and predict the theory. Eisenhardt (1989b) also contributes to the theory building stating that the process of theory testing involves measuring constructs and verifying relationships. In an iterative process, like the one proposed by the grounded theory, the concepts, categories and relationships emerge shaping the theoretical model, during the coding process. This model shows as core category, the agency theory construct. Twelve other constructs that relate to this core category have been identified, three of which with direct relationship: alignment, trust and impartiality. Each of these constructs shown in Figure 5.3, and their hypothesized relationships, are elaborated in the subsequent section. Figure 5.3: Theoretical model construct on the basis of Grounded Theory H2 Alignment H3+ H4 H1 + + - Family and Business Interaction Management Control + H7 H6 - Agency Conflict Degree of Board Professionalism Trust + H8 + H10 H5 Transparency H9 - H12 Altruism - + Impartiality H16 Source: Research data 145 Competence Opportunism + H11 + Legitimacy H14 H13 H15 - Comprehension of social roles V.5. Validation of hypotheses and constructs through a comparison of the literature In this final step it is time to compare the emergent theory with the extant literature and examine what is similar, what is different, and why (Eisenhardt, 1989b). Certainly there is always some relevant literature to refer to. Voss et al. (2002) highlight the importance of reviewing the findings against the available knowledge to avoid reducing their degree of reliability. Thus, each of the constructs empirically raised must be contrasted with the existing theories for completion and validation. It must be said in justification that, because of the volume of constructs proposed in this theoretical model, the literature review carried out here cannot be allembracing, but at the same time it is not conducted only superficially. V.5.1. The Alignment Construct Alignment is one of the most obvious and expected constructs to compose this model, since it is the basis of the theoretical discussion of agency conflict: to what extent the objectives of the agent and principal are distinct or in agreement. For Mayer and Schoorman (1992) goal alignment reflects the overlap of goals and values of the owners on one hand and the management on the other. In this model, the construct alignment is intended to represent the relevant elements in the reality of a family business that may exacerbate the agency conflict in the absence of shared aims. Carlock and Ward (2010) suggest that family and business are different entities: the first related to caring and the second to money. Thus, besides the economic results which are the usual meaning for the existence of a firm, it has to be recognised that family owners may also considered social, psychological and even the spiritual objectives. In fact, the ‘three circle model’ suggests that strategy in family-owned firms must consider the different factors which are specifics for this sort of company, taking into account this dual identity of the family firm, and its business and social interests. (Ibrahim et al., 2004) For Schulze et al. (2003b) alignment in family-owned firms is more difficult to be achieved because family members are locked into the firm. This requires a greater 146 effort to get a clear comprehension of the individual goals, to avoid the consequential divergences related to a lack of alignment: Conflict can occur when family members are not clear about what goals are being pursued or when differences in priorities are not communicated as part as the family’s planning and governance processes. […] The multigeneration family needs to confirm its shared, often multifaceted, values – and then agree on a vision that drives strategic thinking and planning for the family and business. (Carlock & Ward, 2010, p. 14). Besides alignment related to values, vision and strategy, other facets have to be considered. The preparation of the succession, the criteria for admission of family members, the definition of the role of shareholders, the remuneration of members of the family, among others, are issues of great importance. Figure 5.4 shows the aspects that arise from the four cases developed for this dissertation and which deserve to be researched as potential drivers in the increasing of the agency conflict. Figure 5.4: Categories of the Alignment construct Source: Research data Based on the understanding that a family business consists of family and nonfamliy members who, as individuals, try to maximise their chances of success and minimise their chances of failure, and who will therefore search for a more cohesive way to operate to the benefit of the company, it is argued that: Hypothesis 1: The greater the alignment among the family owners and between them and the managers, the lower the agency conflict. 147 Succession Although usually what matters in a succession is the effectiveness of the outcome of the transition, i.e. the ‘successor´s ability to keep the family business healthy by means of sustained growth and continued profitability’ (Venter et al., 2005, p.289) some authors suggest that the alignment with the succession process by the predecessor, successor, and other family member, is also of high importance for the future harmony within the company (Handler, 1989b; Sharma, 1997). This idea is also defended by Santiago (2000) who points out the importance of a leadership transition to be done so smoothly that the change does not disrupt the family or the business. To Venter et al. The more informed the family is about the generic issues, phases, and activities associated with generational transitions, the better able it will be to understand how succession has manifested itself in its particular case (2005, p.298). Churchill and Hatten (1987) have developed a life cycle approach to designate the succession process between predecessor and successor in a family firm. Four phases are proposed: (1) owner-management stage, where the owner is the only member of the family directly involved in the business; (2) a training and development stage, where the descendants learn the business; (3) a partnership stage, where father and son share experiences working together; and (4) a power transfer stage, where responsibilities shift to the successor. Negotiating with transparent deadlines for the beginning and completion of each of these steps is central to not creating expectations that will not be met and the consequent emergence of tensions. However, the most common attitude of the founder (and all who pass through this process) is to avoid committing him/herself to the implementation of the succession. Lansberg (1988) regrets this reality, noting that according to his research, the lack of adequate succession planning is one of the main reasons for the death of companies. The author points out that in addition to long-term benefits such as longevity and the evolution of family business, also in the short term the 148 company gains in efficiency by reducing the tensions stemming from uncertainty over the succession process. Vision & Values In a family business, a shared vision involves family members’ collective ideas about the future of the firm, including desired business domains, desired growth rates, and financial performance (Mustakallio et al., 2002). A shared vision provides a common framework which allows assessing available information and focussing on relevant issues, thus promoting coherence between stakeholders’ expectations and opinions on organizational goals (Ring & Van de Ven, 1994). Also, a shared vision reduces the threat of opportunistic behaviour and helps establish a social norm of reciprocity, which reinforces commitment to jointly agreed-on decisions (Uzzi, 1996). Haugh and Mckee (2003) suggest that in family-owned firms commitment and loyalty are shared values that promote better integration, not only circumscribed by the family innner cicle, but extending to include nonfamliy members who have become part of the team of employees. And when all actors in a family-owned firm share a common vision, the sharing of information intensifies, increasing alignment and improving decision quality (Dyer & Singh, 1998). According to Leana and Van Buren (1999), a shared vision provides the members with the basis to be ‘good agents’ and to commit themselves to collective goals and actions, playing down their individual wants. Furthermore, Haugh and Mckee’s (2003) findings show that shared family values encourage autonomy and reduce the need for close supervision, avoiding costly and inconvenient monitoring mechanisms. Strategic alignment According to Dooley and Fryxell (1999), gaining commitment to strategic objectives is a challenge in all kinds of companies because strategic decision processes are replete with conflicts. Astrachan and McMillan (2003) suggest that the 149 first step in avoiding such inherent difficulty of achieving strategic alignment is trying to identify what ‘success’ means for the stakeholders. This is particularly important since the main objective of the strategy is to offer economic and non-economic benefits to shareholders, employees, customers and the community (Andrews, 1991). Once objectives which meet the expectations of stakeholders in a balanced way have been defined, you must define a strategic plan that integrates, coherently, these objectives, policies and actions of an organization. (Quinn, 1991), Strategic planning is one of the most widespread tools which allows managers to construct alternative future scenarios, identifying the positioning of the company and its opportunities and threats (Boyd, 1991). It includes business and corporate plans, their approval at board level, setting an annual capital and operating budget, performance targeting, and performance assessment processes (Grant, 2003). Following all these steps, strategic planning in a family firm, would promote cohesion among managers, board and family members aligning the family’s expectation with the business’ needs, and as a consequence focusing efforts and preventing conflicts: Designing a business strategy aligned with the family’s vision will ensure that everyone shares a clear picture of the company’s future and the resources available to support the strategy. In a family business, management needs to consider the family’s values and vision carefully to ensure that the owners are agreed on their intentions and commitment ... [this] is a critical concern for management in planning the firm’s longterm strategies and investments. (Carlock & Ward, 2010, p.4041). Risk taking The study of agency relations brings in its wake not only the discussion of control problems but also includes issues related to risk sharing. Jensen and Meckling (1976), seminal authors of agency theory , assert that agents (managers) are usually seen as risk averse and the principals (the owners) are more willing to accept risk as greater risk is associated with higher potential return. For them, this distinction between the propensity to take risks or not leads to agency conflicts: managers making rational decisions, seeking to serve their own interests and 150 hedging against risks, produce decisions inconsistent with the interests of the owners. Eisenhardt (1989a) explains this misalignment concerning risk-taking, since agents have less ability to diversify their employment, while principals are able to diversify their investments. Kiser (1999) adds that the risk-averse agents tend to reject contracts that lead them to take the most risks, preferring contracts with fixed wages and even low values, but with less variation. The effective alignment of interest between owners and their agents in taking worthwhile calculated risks is, therefore, essential. Remuneration & Incentives Mitnick (1973) defines the incentive systems as a set of mechanisms, financial or non-financial, to align the personal goals of the agents to act in accordance with the preferences of the principal, superseding their own goals. For Eisenhardt (1989a) the best instrument to achieve this is a contract between the principal and the agent, based on common goals, which motivate agents to act according to the interest of the principal. His argument is that this type of contract aligns the main agents as the reward of both is now based on the same actions. An alternative approach that enhances the level of commitment of managers with the strategic objectives of a company is to make them shareholders by distributing shares as performance bonuses. Jensen and Meckling (1976) observed that as managers’ ownership of the firm’s stock increases, the interests of managers and outside shareholders become more aligned. It is interesting to note that Schulze et al. (2001) examining the effect of incentive compensation on the performance of closely held family-owned firms, found that nonfamliy managers are influenced by pay incentives, but not family managers. This means that variable pay does not serve as a governance mechanism in familyowned firms, where the family members are leading the company. In another study, McConaughy (2000) compared the CEO’s compensation of family-owned firms in two distinct situations: those with family and those with nonfamliy CEOs. He found that 151 family CEOs have less need to receive additional incentives for higher commitment. Also, the findings of this study suggest that family controlled firms have to pay nonfamliy CEOs more to get comparable performance. O’Reilly and Chatman (1996) show the reasons behind this reality, when they compare agency and stewardship theories. While in the first, the formal control mechanisms, such as reward or punishment dominate, in the second, the social control mechanisms, such as shared values and achievement of expectations attitudes, impel the managers to higher performance. As family managers typically tend to fit the model of stewardship, it is natural to find a reduction in agency costs through compensation and incentives, when the CEO is family. Corporate Cohesion To ensure the perpetuation of a company, especially post-succession, the founder needs to define in advance how it will be managed and what are the rules of coexistence among the members, so that its goals and individual interests prevail. Therefore, it is necessary to establish an agreement between those who run the destinies of the company, how they relate over time. It is normal to establish official corporate documents (articles of incorporation or statute) with the minimum conditions of validity of the business in its legal form. But rather than presenting the minimum requirements for legal validity of the society, the formatting of a corporate agreement becomes, therefore, a means of disciplining power sharing, with limits on how members will exercise control. A shareholders' agreement, in addition to regulating the buying and selling of shares and voting blocks, as required by law, must express the goals, vision and the agreed criteria about: succession of patrimony and management; levels of decision; entry and exit of partners; distribution of dividends; private use of goods and services of the company; creation of private business of the partners, capital investment and financing; tag along and drag along clauses; rules for the valuation of the company; dispute-resolution mechanisms; transfer of shares; mergers and acquisitions, joint ventures, partnerships, bankruptcy and creditors agreement; distribution of dividends 152 and executive compensation; and board and the chairman selection (Friedman & Friedman, 1994; Alvares et. al, 2008). Moreover, being a multilateral agreement, in addition to the above themes that serve as parameters of content to be discussed among the shareholders, the agreement must have the full force of law with respect to contracts, dealing with topics such as confidentiality, notifications, modes of termination, and period of validity, irrevocability and irreversibility. V.5.2. The Level of Independence and Efficiency of the Board Construct The role of the board of directors has been central to the corporate governance studies in public firms, not only because it is a product of regulation for this kind of company, but also because of its importance in diminishing the distance between shareholders and management. On the other hand, literature specifically addressing the boards of directors of family-owned firms, most of them privately held, is still rare. However, those who investigate its beneficial adoption in such firms, recognise that it is one of the most important organizational mechanisms in mitigating agency conflicts. Ward (1991), for instance, emphasizes the value of boards in family corporations, addressing both business issues and the needs and interests of the owning family members. Hill and Snell (1989) see the board of directors as a corporate governance mechanism which helps to harmonize the interests of family owners and managers. Fama and Jensen (1983a) also register the utility of a board of directors in family businesses as a formal control mechanism, to align the interests of both groups. After all, the board of directors represents the interests of shareholders and seeks to align the interests of agents and owners through corporate governance mechanisms such as reporting, auditing, and policies (Jensen & Meckling, 1976). These considerations result in the proposal of the following hypothesis: 153 Hypothesis 2: The greater the level of independence and efficiency of a board of directors in family-owned firms, the greater the alignment between owners and managers. The Level of Independence and Efficiency of the Board construct in this study considers the following elements: Figure 5.5: Categories of the Level of Independence and Efficiency of the Board construct Source: Research data Structure of the Board of Directors In speaking of the structure of a Board of Directors, three features should be considered: 1ST SIZE OF THE BOARD OF DIRECTORS According to Boyd (1990), to be effective, the board of directors of a company should opt for a number of members that will enable it to work properly as a team, without being too big or too small. Though a larger company and a more complex one require a board with greater experience and therefore more complementary skills, authors such as Jensen (1993) and Lipton and Lorsch (1992) suggest that large boards can be less effective than small ones. Ward (1991) also suggests that smaller boards are more desirable for family businesses since individual responsibility tends to dissolve in larger groups. The idea is that when Boards become too big, CEOs are more likely to have control of it, thereby increasing the agency problems (such as director free- 154 riding), so that the board assumes a more symbolic role. In this respect, Andrade and Rossetti (2011) suggest a sizing with an average of seven members, with variation of plus or minus two, depending on the current and future characteristics of business. Nash (1995) and Newell and Wilson (2002) also share the concept that the most effective boards range between five and nine directors. 2ND PROFILE OF THE BOARD MEMBERS Heidrick (1998) defines the ideal director as ‘a friendly, independent thinker who expresses his or her thoughts on a compatible basis… the ideal director is not a yes man or woman’ (1998, p.272). In fact, a board member who acts in an independent manner will counsel and demand the CEO during the decision making-process increasing the potential for optimal decisions both for the business and for the owners (Daily & Schwenk, 1996). For Nadler (2004) more than independence, the main attribute for a board member is the proper competence required for the function, not only accounting and public-reporting expertise, but also all knowledge related to the company, its environment, and its industry. His or her business experience must be sufficiently broad or deep that it allows him or her to deal with consultative challenges. But, besides cognitive attributes, the behaviour and posture expected from a board member must be also considered. Andrade and Rossetti (2011) highlight integrity, openness, a capacity to listen, outspokenness in expressing his or her points of view and constructive dissatisfaction, as those especially desired. Heidrick (1998) suggests that the best board member that can be chosen is intelligent, well-informed, analytical, and a team player. For Li (1994) the corporate board, with its mix of expertise, independence and legal power is a potentially powerful governance mechanism. 155 3RD STANDARDS OF OPERATION The Board should establish clear rules of operation. For instance, one must establish an annual schedule of meetings with the guidelines being prepared by the Chairman and adjusted with the directors and CEO. As for frequency, no more than six, and no less than three meetings per year are recommended (Gallo, 1993). Moore (2002) claims six meetings per year in alternate months as a good pattern for most companies. Gregory and Simmelkjaer (2002) present as the European average, eight meetings per year. Ward (1991) sees as the usual procedure in family business, board meetings at least four times per year, supplemented by additional monthly executive committee meetings. As for the decision-making process, they shall be taken by consensus, resorting to a vote only in extreme cases. Each member must have one vote. Adequate preparation for the meetings of the Board by its members is critical to its effectiveness. To this end, certain information must be submitted in sufficient time, at least five days in advance. (Gallo, 1993; Alvares et al. 2008). Representativeness versus competence level The main board tasks are approving investments for growth; counselling, monitoring and evaluating the CEO performance; hiring, firing and setting compensation for the top management; reviewing corporate strategy, goals and policies; and ensuring audit procedures. Moore (2002) suggests that the content of board meetings should include all key issues produced by the CEO and other senior executives to the board, such as the results of operations, the status and outlook of financial and strategic plans, proposed or rejected business deals, the current financial forecast and early signals on changing trends, the economic and competitive environment, public policy issues, shareholder matters, and any proposals for board approval. Cowling et al. (2001, p.372) added that the board of directors ‘can link a firm with its external environment and secure critical resources, including prestige and legitimacy’. 156 To accomplish these expected roles, Aronoff and Ward (1996) remind us that a board member might be ‘an experienced, objective, knowledgeable outsider, usually an active chief executive, who brings to the table background in the particular strategic issues facing the business, as well as sensitivity to the special concerns of family-owned businesses’. (Aronoff & Ward, 1996, p. 30). Likewise, Mallin and Melis (2012) argue that for shareholders to protect their investment, they should constitute a competent Board ‘to manage the company and to ensure that effective strategies are in place for the company’s overall corporate performance and long term sustainability’ (p. 171). A board of directors, therefore, is a place reserved for the experienced professionals whose level of contribution can be effective. Thus, it is recommended that rather than being represented by directors who may not have the necessary level and are chosen simply because they are family, the ideal is that each block of a family company nominates independent directors that they can recognize as their true representatives. Aronoff and Ward (1996) also draw attention to the fact that it is best to involve independent directors on the board, even in the first or second generation, since it can later be much more difficult when the council is already filled with representatives from each branch of the family. Decision-making forum The Board of Directors usually has been identified for two important roles: oversight of management and counsel in the strategy process (Mace, 1972; Pearce & Zahra, 1989). This monitoring and control role is relevant to the protection of shareholders’ interests, but it is in the strategy process where it could really make a difference. Bathala and Rao (1995) recognise the board as the authority to give sanction to the major policy initiatives within a company. For these authors, it is the final instance in the decision-making process. 157 Board of Directors’ Independence According to Hermalin and Weisbach (2003) the most important factor determining a board’s effectiveness is its independence from the CEO. Of course the CEO would prefer a less independent board and as much power concentrated in his/her hands as possible, and a Board of Directors as dependable as possible. This is the case in most family-owned firms, where the CEO determines the functioning of the board and where his/her intense involvement in the selection process of directors may affect its independence (Finkelstein & Hambrick, 1996). Thus, it is fundamental to select board members whose commitments to his/her role is not impacted, at all, by any imposition of the CEO. Felton and Watson (2002) maintain that they should be free from commercial or personal ties that could prejudice their capacity to query and challenge the top management. Heidrick (1998) defends the idea that the presence of external directors contributes to increasing the independence of the board in family-owned firms. For him, a family member, an employee, or an outsider who depends on the family, acting as a board member, usually tells the CEO what he wants to hear. Kesner and Johnson (1990) see this kind of director adding concrete advantages to a family firm, when he/she has the necessary skills to act as a board member, a high capacity to understand the owning family and the nature of its relationship with the business, and a high independence to counsel and decide with freedom. External board members Regarding its composition, it is recommended that boards should be composed of a mixture of insiders and outside members. In the case of family-owned firms, Voordeckers et al. (2007) present three different configurations: family boards, composed solely of family members; insider boards, composed of family members and nonfamliy managers; and outside boards, with the inclusion of external board members. Outside members usually refers to those directors who are not part of the owning family or of the management (Ward & Handy, 1988). The advantages of each class of member are that internal members usually bring a better understanding of the organization’s activities and the outsiders usually 158 come ‘with expertise in capital markets, corporate law, or relevant technology who provide an important support function to the top managers in dealing with specialized decision problems’ (Fama & Jensen, 1983a, p. 17). Outside directors also are more likely to challenge established assumptions about strategy, inducing managers to reposition themselves regarding things they say they are certain about (Pearce & Zahra, 1989). This leads to a better strategic decision process and increases the company’s chances of survival (Daily & Dalton, 1994). Practice, in fact, has shown that family-owned companies that set up a Board with experienced external members, have shown superior results and guaranteed the perpetuation of their business (Aronoff & Ward, 1996). To Bathala and Rao (1995), external directors act as an important force for minimizing agency problems between managers and shareholders. Brickley and James (1987) found in their research that the presence of outside directors tends to reduce managerial consumption of perquisites and Weisbach (1988) proved that the higher the proportion of outsiders on a board, the more prompted is the board to substitute the CEO after a period of poor corporate performance. Not all authors are in accord on the value of outside members. Ford (1988), for instance, sees no value for privately-owned firms to adopt outside directors. His findings reveal that outside directors seem to have less value because of their lack of specific knowledge of the firm and their lack of availability to the firm. Jonovic (1989) claimed that an outside board member does not provide the best governance structure in the majority of family businesses due to their unique governance challenges. But the fact is that most authors agree that the mixture of external and internal members on the board may provide independent expert opinions, from the former and deeper understanding of the family culture from the latter, allowing all relevant views to be represented. Thus, although a mixed board seems to be the best option, a survey conducted by MassMutual Financial Group and Raymond Family Business Institute (2003) regarding the family business reality in the US shows that family members still have a majority on the board. This is due to the fact the entrepreneurs and family 159 owners usually resist bringing in outsiders because they do not want someone directing their actions or revealing family secrets (Nash, 1995). Heidrick (1988) see it as traumatic, at least initially, when family managers have to share information with and consider the viewpoints of independent outsiders, thus making it more difficult for them to adopt a more independent and, consequently, a more professional board. V.5.3. The Family and Business Interaction Construct In family-owned firms, the effects of good social relationships have to be considered, as in most cases board members, managers, and owners belong to the same family. Harvey (1999a; 1999b) points out that traditions, loyalty, altruism and bonding relationships compose a family system which normally influences the operation of the firm. For Mustakallio et al. (2002) these social interactions among family members contribute to create a common vision of the family firm’s mission and goals, which can improve the quality of the strategic decision-making process and managers’ alignment with the strategies. Furthermore, recent literature shows that family involvement within the firm plays a crucial role in aligning individuals with its goals by reducing transaction costs, by the establishment of better information flow, by the promotion of creativity and knowledge generation, besides the formation of alliances (Arregle et al., 2007). Also, Child and Rodrigues (2003) endorse the idea that close social interaction between managers (family members or otherwise) promotes mutual identification, friendship and even close ties. These ideas allow the following hypothesis to be postulated: Hypothesis 3: The better the interaction between family members, and between them and the company, the greater the degree of alignment. Considering the role healthy patterns of family interaction can play in reducing agency conflict, this construct is discussed from the categories highlighted in Figure 5.6, as follows: 160 Figure 5.6: Categories of the Family and Business Interaction construct Source: Research data Family Council One of the key roles that corporate governance in family businesses takes is to expand the interactions among their members by creating links with the company and the property (Mustakallio et al., 2002). For this purpose it is essential to promote from informal spaces, such as get-togethers, to formal family meetings, creating opportunities for family members to meet and discuss issues of their interest. Among the instances of governance capable of promoting harmony between the interests of family and business, the Family Council can be highlighted. According to Davis (2003), Family Council is a result of practices adopted by European familyowned companies that for generations has been used it as a mediating forum, focused on preserving family cohesion and avoiding the appearance of conflict that can negatively impact the company. To Mustakallio et al. (2002) a family council helps extend social interaction, strengthening the family bonds and increasing family members' identification with the company. From the definition of rules and agreements, the Family Council becomes a space for rational and orderly discussion of issues related to the role of family in relation to the business. Lansberg (1988) adduces how important a Family Council is in a succession process, helping the family open the communication channels so that family members may discuss succession together and in different subgroups, reducing the possibility of undesirable resistance. 161 Bernhoeft and Gallo (2003) maintain that the family business also needs to develop a system of education for its members, seeking to strike a balance between individual interests and the demands of the organization, and strengthen the Family Council as a bridge for communication between the company and family. Among its main tasks, they cite:  Managing all issues and family interests in its relations with society and company;  Attending to the values and the family history and their transfer between generations;  Acting as a representative forum of the families in their relations with the Boards of Directors and the Shareholders’ Committee, and vice versa;  Planning and coordinating meetings and family events;  Creating and managing training programmes for the family members developing their roles as shareholders;  Creating and managing vocational, professional and personal programmes for the young heirs, as a complement to the orientation of each family. Family Harmony Given the diversity of actors and the interests involved, the family business can be understood as a scene of constant negotiations, disputes and rearrangements of power, where different actors are immersed in games of interest, while seeking to build its legitimacy with different stakeholders (Pereira, 2010). Such tensions are enhanced by the concept of interrole conflict presented by Kahn et al. (1964). According to these authors the idea of interrole conflict occurs when the pressures that an actor suffers and which are associated with a particular role, will influence his or her performance with other social groups to which he/she also belongs. In a company family interrole conflict occurs, for example, when problems and responsibilities related to the work, interfere with the duties of the 162 family, and vice versa. Smyrnios et al. (2003) believe that the presence of interrole conflict in family businesses ends up being an inevitable reality. And therefore, disruption of family harmony, bringing rivalries and resentments to the business environment, will result in direct negative consequences for the company: Personal conflicts that dominate the interactions and remain unresolved distract from the grant-making business at hand. Bystanding family members and the nonfamliy administrators are reduced to keeping the peace or skirting controversial issues. A great deal of time and energy must be devoted to conflict management. In these situations the work of the foundation suffers, and valuable board members may become discouraged and withdraw. Nonfamliy administrators rarely avoid being caught in the family's political web, and, inevitably, their ability to perform routine tasks is affected. (von Lossberg, 1990, p. 378). Sharma et al. (2001) see the lack of harmony as leading to misalignment regarding the vision and to disagreement on the way to achieve it. For them, ‘the more harmonious a family is, the more likely it will be that its members will want to continue to pursue a shared vision in the interests of all concerned’ (Sharma et al, 2001, p. 28). Family interest in the firm Dyer (1988) has suggested that the family business culture is an important factor in determining the success of the family business beyond the first generation. The strength of the culture and family norms determine how the family contributes to the success of the business (Harvey, 1999a). In this sense, shared values can directly influence the degree of interest of the heirs and therefore the longevity of the organization. Tapiès and Fernández (2010) point out that values are passed from generation to generation through the process of socialization and education of younger members. For the authors, the younger generation internalizes the values observing, listening and living with the previous generation. Kets de Vries (1993) confirms that this idea that socializing with the company since childhood, facilitates greater learning about the business and a greater bond that is established with the history and the present values of the company. 163 After all, they have been in contact with the business from early childhood onward. Breakfasts, dinners, walks, family gatherings, and summer jobs create opportunities to learn more about the business […]. This kind of knowledge may give family members a head start, compared with executives entering the business later on. Early training like this helps to explain the sometimes puzzling appointments of very young family members to senior positions. (Kets de Vries, 1993, p. 313) It is important to consider that the gradual increase in the number of offspring can lead to a loosening of the kinship that has bound them up to a certain point, making it more difficult to perpetuate the values for future generations. In view of Tapiès and Fernández (2010), the formalizing of family values by drafting a family constitution could help in the preservation and dissemination of these values for future generations. Thus, the organization's history, guided by the values that were significant for it, could be documented and passed on to successors, contributing to the construction and/or maintenance of a shared vision of the business. Kinship ties for perpetuation of the firm To achieve longevity of the family business, and escape from the statistics which appoints the usual short life span for family-owned firms, it is essential to increase the levels of interest and commitment of the successors in continuing to run it. Handler (1989a) calls attention to the precise need for the new generations to be integrated into the family business to ensure good long-term performance of their companies, while Cabrera-Suárez et al. (2001) highlight the need to transfer knowledge and information from the previous generation, to ensure the perpetuation of the family business. Of course issues such as the competence and training of successors are also extremely important. However, the basic premise underlying the perpetuation of the company is really the identification of the successors with the business they will become responsible for running in the future. The absence of heirs or successors with an interest in taking over the family business could mean the end of it, or at least indicate the need to seek nonfamliy successors to take over the management. It is important to emphasize, as evidenced by Rossato Neto and Cavedon (2004) that parental pressure can occur to force the children to take positions in the family business, in order to keep future generations in charge of it. But the level of commitment of the heirs, compelled to take over the 164 business, would hardly be the same as that of the founder. Thus, there is a risk that many members of the family business may have other types of motivations for the organization, with a lower level of involvement or disconnected from affective values. Some people work because they have to work, not because they want to work. This is the beginning of the demise of family businesses: the values are unclear, the loyalties divided and motivation is based on money. The succession plan for the family business depends on the commitment of people and the ability to transmit it (Rossato Neto & Cavedon, 2004, p. 6). However it is not merely the absence of successors which can negatively impact the business. Lima (2000) indicates the existence of a deep sense of contradiction in the family firm divided by opting to follow meritocracy or the norms of dynastic succession. Authors such as Stewart (2003) corroborate this view warning that when the importance of kinship ties is given precedence over the logic of efficiency, much harm is generated for the company. The most obvious distortion is nepotism, which will be discussed in more depth within the altruism construct. In the presence of nepotism, criteria of experience and ability are relegated to the background, often in favour of hiring unqualified relatives that cannot be easily dismissed (Lomnitz & Pérez-Lizaur, 1987; Whyte, 1996). Despite these disadvantages, Stewart (2003) states that kinship would also offer benefits able to offset its risks. The author points out that the benefits of kinship are linked to emotional aspects, affective support and cohesion in the family. The relatives would represent a source of social support, financial resources, advice and information and business channels (Stewart, 2003). For Davis and Herrera (1998) and Walsh (1994) cohesion represents the most fundamental measure of effectiveness in a family business. Cohesion reflects the level of mutual support and how strongly the members of the family group are connected, influencing the quality of relations between the company and family. Family systems with high cohesion are characterized by strong affective ties of interdependence, with the family spending most of their time together (Olson & Gorall, 2003). This factor can create a sustainable foundation of trust, confidence in raising the family directors and consolidating relations between the board and the CEO. 165 Integration of the family with the firm Harvey (1999b) argues that a reciprocal involvement between family members and the company can help to create shared visions of business capable of generating long-term value. In addition, the durability of these interactions creates a shared language and a collection of narratives that provides a knowledge base common to all family members (Nahapiet & Ghoshal, 1998). Barnett and Kellermanns (2006) defined three different levels of interaction between family and business: family influence dormant (low interaction); facilitating family influence (moderate interaction); restrictive family influence (intense interaction leading to excessive levels of family influence in the company). In the first case, where there is low interaction between family and business subsystems, although the company is not subject to the total control of the family, the opportunity of deriving benefits from family experience and knowledge is lost. (Astrachan et al., 2002). This may relieve uncertainty about the role of nonfamliy executives, but on the other hand, these businesses do not enjoy the positive aspects that family culture could provide, including the establishment of a strong sense of identity with the firm, generating a greater commitment to their values and goals (Barnett & Kellermanns, 2006). At the other extreme are the family businesses that present an intense interaction between family and business. Barnett and Kellermanns (2006) indicated that this high level of influence can result in loss of autonomy of nonfamliy managers, creating agency problems unique to the family owned firms. In this sense, the ideal would be to reach a situation of moderate interaction between family and business, where the company would benefit from the involvement of the family, avoiding at the same time abuse of power on the part of family members (Barnett & Kellermanns, 2006). 166 V.5.4. The Management Control Construct The main purpose of any management control system is to ensure that the behaviour and the actions of all workers are aligned with the company objectives and strategies. Thus, management control assumes an important role in the strategic alignment process, inducing managers to act to reinforce the strategy of the firm. This is exactly the reality demanded by agency theory, for which the principals must align their agents in relation to their interests. One of the main ways to make sure that this goal is achieved is to design an appropriate system of management control which can provide the proposed strategic direction to all stakeholders (Slagmulder, 1997). This concept leads to the following formulation: Hypothesis 4: The more developed the system of management control in the family business, the higher the prevailing alignment. The control mechanisms to be discussed in this section relate to issues such as: the effectiveness of boards of directors; the existence of economic indicators and financial control; the level of protection of the interests of shareholders; the level of autonomy of the CEO; and the attitude of shareholders in terms of their level of activism monitoring the business. Figure 5.7, introduced below, shows the structure of this construct: Figure 5.7: Categories of the Management Control construct Source: Research data 167 The effective controlling role of the Board of Directors The Board of Directors, as previously discussed, has important functions. Corbetta and Tomaselli (1996) point out the roles of advice and guidance for the CEO, the responsibility of establishing links with the external environment and the role of representing the company to the community as the most important ones. In family-owned firms, its importance is also recognized in acting as a bridge between the company and the family. In their turn, Forbes and Milliken (1999) highlighted the role of the Board of Directors acting as a control and monitoring centre, hiring, compensating, disciplining and dismissing executives. The performance evaluation of the executives and the consequent decision concerning the ways of providing incentives can be based on the monitoring of their behaviour or the observation of performance resulting from their pattern of behaviour (Johnson et al., 1996). Baysinger and Hoskisson (1990) advocate the existence of two levels of control by the Board of Directors: strategic control and financial control. The strategic control usually includes subjective criteria, related to the drivers of future performance of the organization, while financial controls propose the use of objective indicators, representing the past and present performance achieved. CEOs have to have their performance regarding the delivering of results evaluated at both levels. Board of Directors of family-owned companies are challenged, however, to apply such assessments to family CEOs with exemption since they typically give ' the benefit of the doubt to the incumbent (e.g., bad luck or unfortunate circumstances rather than incompetence) when interpreting ambiguous performance data' (GomezMejia, 2003, p. 229). Presence of effective economic/financial control Companies that choose to create boards with family members, representing their own family branches, and who do not have the proper qualifications, are common on the Brazilian scene. They are often not even able to understand and interpret economic indicators and financial data presented by the CEO or the Board 168 of Directors in their balance sheets and reports. Thus, one has to understand that the Board is a place reserved for experienced professionals who can lend their expertise to the company's development. In addition to the formation of a professional board to exercise proper control, Eisenhardt (1989a) believes that effective information systems should be developed, avoiding the principal being deceived by the agents. For her, the better informed the principal is, the more he/she will make the agent work in line with his/her interests. But though the principal possesses the best mechanisms of control over his/her agents and has the right to dismiss at will, the question that arises is related to doubts about the ability to execute a perfect control, either because of lack of time or skill. Even, therefore, using quite non-lenient monitoring, the owner may not have sufficient financial sophistication to understand the operations and financial data (Ang, 2000). To minimize risk of agency problems due to lack of control, several economic and financial parameters can be established such as: reporting and detailed statements by the CEO; generation of reliable indicators of the company's financial health, based on best accounting practices; the existence of communication channels available to shareholders to provide additional information about the business. Information systems, an essential part of any control mechanism, serve to communicate the strategic outcomes throughout the organization. This prevents vertical information asymmetry, which together with the lack of agreement on goals are the main causes of strategic misalignment (Verhaegen, 1986, cited in Slagmulder 1997, p. 132). Protection of the shareholders’ interests The agency problem can become worse when the principal is unable to adequately monitor or control the behaviour of agents. For Jensen and Meckling, (1976) principals can reduce divergent interests by increasing incentive alignment and diminishing the agency costs associated with bad decisions by increasing 169 control. For this, they may resort to a series of mechanisms both motivating and aligning the agents, such as compensation systems, and establishing a Board of Directors focused on monitoring the company's executive group. Pearce and Zahra (1989) argue that the increased level of monitoring of managers would be able to reduce information asymmetry between agent and principal. This would encourage the owner-managers to examine their industry, seeking opportunities and monitoring trends in its competitiveness. Also according to the authors, monitoring via the Board often relies on the contributions of outside directors, who are more likely to challenge established assumptions about the strategy, forcing managers to refine their positions. Moreover, the experience of these board members can also raise the quality of strategy discussions. According to Mitnick (1982) the principals seek to control the behaviour of their agents, and the agents, in their turn, must seek to control or manage the elements of their environment to meet their key objectives. Thus, the agency models usually focus on incentive systems that produce these controls. In respect of an incentive that exists between a principal and an agent, the principal must seek to use rewards and encouragement indicating the behaviour that is expected from the agent. To do this, the principal must assume the costs of identifying the agent’s actions that would satisfy his/her preferences and the costs of monitoring and ensuring compliance. The definition of models of executive compensation and incentives tied to company performance is emerging as a major form of internal control. For Steinberg (2003), fixed and variable compensation and stock options are internal mechanisms that contribute to minimize the problem of agency. Authors such as George (2002) warn that executives should not be responsible for the definition of compensation systems, emphasizing the importance of setting up a compensation committee, submitted to a Board composed of independent directors, to define the incentives and remuneration of top management. As noted by Oliva and Albuquerque (2007), the performance bonuses meet short-term goals while the stock option programmes, long term. The adoption of stock options is not exempt from risks; however, it is interesting to note that giving shares to executives, fulfils the role of bringing them closer to the interests of shareholders. 170 That is, the variable remuneration of the managers would be affected by fluctuations in share prices in the market, ensuring closer alignment between the interests of executives and shareholders (Oliva & Albuquerque, 2007). Note that many of the instances, regulatory and self-regulatory, that are designed to monitor the agency relationships, carry in themselves other agency relationships. These instances are represented by internal or external auditors, regulators, departments of internal affairs, government regulators, insurance companies, investment advisers and rating agencies, all acting on behalf of some set of entities. They may, therefore, also incur agency problems, the existence of shirking, co-optation, bribery or perhaps simply a lack of monitoring of the right things. As Shapiro (1987) asks in an escalating cycle of agents overseeing agents, it should be asked: who monitors those responsible for monitoring? Level of the present autonomy Autonomy is a basic human desire for independence, freedom and free will and that is usually found in situations of self-employment (Matthews et al., 1994). However, several authors have shown that if you can provide autonomy in the business environment, positive work attitudes and other outcomes, such as job satisfaction, are achieved (Hackman & Oldham, 1976; Karasek, 1979; Karasek & Theorell, 1990; Lawler & Hall, 1970). Although some employers fear that greater individual autonomy affects performance, some research suggests that it could even increase productivity (Bailyn, 1993a). For the agreed level of autonomy to be implemented without the appearance of unnecessary tension, the outcomes should be properly negotiated, the same way that checkpoints should be established. In order to regulate the level of autonomy granted to managers, Mitnick (1973) suggests the use of surveillance to restrict the freedom of agents, reducing the set of possibilities for discretionary choices. For him, within a principal-agent relationship, such policing mechanisms must be implemented to an extent which depends on the degree of freedom that it is willing to give, and the cost of implementation and maintenance of monitoring the agents. 171 It is important, therefore, that the level of control established serves as a support for the alignment of the goals of principals and agents, reducing the risk that the granted freedom ends up as an agency conflict. Laissez faire control Ang (2000) shows that agency costs tend to increase with more fragmented ownership structure for a company. As the number of shareholders is increased, they start to take a more passive attitude regarding the monitoring of companies. That is, the decreased ownership of these investors, can lead them to take less responsibility for the direction of the company, not recognizing the burden of monitoring the management. The criticism of the passivity of shareholders was reported by Robert Monks and Nell Minow who advocated the thesis that organizations that have shareholders who are dedicated to the effective monitoring of management tend to add more value and generate more wealth than firms that do not have the same follow-up (Monks & Minow, 2004). These shareholders rely on the law of least effort, understanding that their return will be constant, regardless of their attitude toward business. However, Monks indicates that these returns tend to rise if the shareholders assume an active posture, participating and monitoring the actions of the executives (Andrade & Rossetti, 2006). Of course, this phenomenon is less prevalent in family businesses, whose owners also usually accumulate the function of managers. But even there, where ownership is dispersed by the presence of a large number of family members or other types of shareholders, there is a risk of a looser control. Dyer (1988) points out that the major risk of a laissez-faire culture is that employees may not act consistently with the family's basic values and assumptions. They can lose sight of the objectives of the company, and the business can run out of control. Finally, the last discussion within this construct is related to how management control impacts on disclosure and transparency. Ge and McVay’s research revealed that 172 poor internal control is usually related to an insufficient commitment of resources for accounting controls. Material weaknesses in internal control tend to be related to deficient revenue-recognition policies, lack of segregation of duties, deficiencies in the period-end reporting process and accounting policies, and inappropriate account reconciliation (2005, p. 137). The authors remind us that the Sarbanes-Oxley Act has increased disclosure requirements related to the effectiveness of internal control, the focus of SOX 3026 being on disclosure of controls and procedures, while SOX 4047 focuses on internal control. This allows the proposition that: Hypothesis 5: The more developed the system of management control in a family business, the greater is the level of disclosure and transparency. V.5.5. The Trust Construct Trust is understood by Sundaramurthy (2008) as a fundamental element in the competitiveness of social organizations, especially given the increase in the levels of complexity and uncertainty present in the market. For the author, trust refers to a personal belief that the individuals engaged in processes of exchange will use their best efforts to honour their commitments without wanting any advantage for so doing, and so assuming opportunistic attitudes. Ring and Van de Ven (1992) propose that trust is a key element on which governance should be based, precisely because it would be capable of reducing the risk of opportunistic behaviour in principal and agent’s transactions. Complementing this idea, Lyons and Mehta (1997) define trust as a meaningful concept for the parties to a social relation if and only if at least one party is exposed to an element of behavioural risk. Behavioural risk is experienced by one party when he or she is uncertain about the behaviour of the other as it may materially affect him or her, and when such exposure is a 6 Section 302 of the Act mandates a set of internal procedures designed to ensure accurate financial disclosure. The signing officers must certify that they are ‘responsible for establishing and maintaining internal controls‘. 7 Under Section 404 of the Act, management is required to produce an ‘internal control report’ which must affirm ‘the responsibility of management for establishing and maintaining an adequate internal control structure and procedures for financial reporting. 173 matter of common knowledge between the parties. In other words using Oliver Williamson's (1985) jargon, behavioural risk can be understood as vulnerability to opportunistic behaviour (1997, p. 240). In accordance with Beccerra and Gupta (1999), the theme trust has been discussed by different sciences, such as psychology, the political sciences, economics and administration. These distinct fields of study seek to analyse the question of trust from different perspectives: trust between individuals; between organizations; between individuals and organizations; trust as a general characteristic of different societies; and lastly, trust as a personal attribute. For Nooteboom (1996), we trust someone, if it is probable that he/she cooperates even not being forced to and not having any material interest in doing so. This means that trust is associated with the non-egotistic forms of cooperation, based on the loyalty to a partner out of ethics, the ties of friendship or kinship, and not on the basis of coercion or self-interest. Among the principal benefits of trust, brought out by empirical research undertaken by Beccerra and Gupta (1999), the following should be mentioned: o Improving individual performance; o Acting as a “lubricant” of the social system; o Promoting interorganizational cooperation; o Developing competitive advantage and; o Reducing conflict. The relationship between the levels of trust present in the organizational context and the reduction of conflicts is found to be extremely important for the discussion proposed by this thesis, concerning agency theory in family-owned companies. That is, in accordance with that pointed out by Raskas (1998), the ties of kinship existing in this type of organization promote trust via a greater sharing of information and a sense of common purpose built up among the family members, which would result in lesser propensity to opportunistic behaviour. Wilkins and Ouchi (1983) share the same opinion, adding that trust creates a common interest and shared expectations which then facilitate the tolerance of both partial goal conflict and temporary periods of inequity within the relationship. 174 This discussion allows the formulation of the following hypothesis: Hypothesis 6: The higher the trust existing in the family and business spheres, the lower the agency conflict. This section will adopt the perspective of a discussion on the implications arising out of the level of trust deposited in family managers, in nonfamliy managers and between shareholders and managers. In addition, the question of the channels utilized for the identification of external executives is also discussed, starting from the network of relationships of the owner family. Such elements can be visualized in Figure 5.8 below. Figure 5.8: Categories of the Trust construct Source: Research data Trust in family managers In family-owned companies, trust can be comprehended as a form of social capital imbued with considerable power, instilling in family managers attributes such as high professional integrity, and ends and values common to those of the founders. LaShapelle and Barnes (1998) maintain that trust can act as a mechanism of integration among the members of different generations of the family involved in the business. Thus, trust would be created and maintained by means of, and in addition to, the family, catalysing the succession processes and strengthening the alignment of the family managers with the business (LaShapelle & Barnes, 1998). Sharma (1997) complements this affirming that trust in the abilities of the successor 175 demonstrated by the predecessor possesses a significant influence on stakeholders’ satisfaction with the succession process. In the context of a family company, when trust is guided by ties of affection, the company tends to be run by informal agreements, based on a high level of faith in the intentions of the other party (Gomez-Mejia et al., 2000; Pollak, 1985). In spite of trust being indicated as a source of social capital for family-owned companies, it is necessary to state that it also possesses a negative side. As Shapiro (1990) points out, this fact becomes evident when the trust existing among family members is used as a justification for the selection of managers from the family, instead of experienced professionals present in the market. In addition to this, family members frequently tend to be generously remunerated, independently of their demonstrated merit. To this must be added the fact that, the stability in the company that the family enjoys, and the less rigid surveillance exercised over it, can raise the risk of shirking and act as a stimulus to other opportunistic practices (Shapiro, 1990). Trust in Nonfamliy Managers For Raskas (1998), trust, in family organizations, is the result of a number of shared experiences and common values, experienced through the proximity of the family members and the power of the family social structure. The nonfamliy managers, on entering the organization, are not always able to readily identify the role of the values disseminated by the family, or perhaps even share them. In addition, because they have not participated in the construction of these values, these executives do not generally enjoy the same level of trust as that shown to the family managers. This question is reinforced by the research carried out by Chrisman et al. (1999) which indicated that the more complex the business, and the greater the number of nonfamliy managers involved in management, the greater is the level of concern demonstrated by the controlling family in relation to the loyalty and involvement of the executive body. Another risk to be considered concerns the play of interests, contrived by individuals not belonging to the family who are capable of reading the family context 176 with the purpose of taking advantage of its weak points. Several researchers such as Babic (2003), Lodi (1998) and Oliveira (1999) are unanimous in affirming that familyowned companies can suffer through hired professionals who on identifying the weak points existing in the relationships between the family members, stimulate conflicts only to gain more space and power for themselves within the organization. Beccera and Gupta (1999) argue that trust is not only based on the power of family ties. It can be based on the level of competence demonstrated by the manager and by a standard of behaviour imbued with honesty and ethics. In other words, a nonfamliy manager who does not have the benefits of kinship to gain trust, may conquer it either through ethical conduct or through competence in managing the resources of the principal. Nooteboom (1996) divides trust precisely into: o Behaviourial trust: the intentions of the agent not to defect willingly; o Competence trust: a partner’s ability to perform according to the intentions and expectations of a relation. Child and Rodrigues (2004, p. 143) in defining the role of corporate governance as a way to ensure "that the managers run the company honestly and efficiently, delivering over the fair and acceptable return to those who have invested in it" make clear that both sides in the trust, whether behavioural or competency based, must be present concurrently. Potts and Matuszewski (2004) defend the idea that the alignment of the executives with the ethical values of the company and their conduct at the head of management over a period of time are fundamental factors in establishing a relationship of trust. In this sense, the contracting of well-known executives in the market represents a chance of forming teams based on criteria of competence, and minimizes the risk of family ties being overvalued to the detriment of the ability and talent necessary for the function. Embeddedness Faced with the risks associated with the contracting of nonfamliy managers, pointed out in the previous section, it becomes relevant to discuss what would be the 177 best channels utilized by the owning family to select and hire these professionals. It should be understood that the families usually are part of a wider social field, establishing in their networks relationships constructed on the basis of trust. Thus, the principals operate in a context which involves multiple relations not only those related to the company: managers, customers, suppliers and other stakeholders. Consequently, it is natural that seeking for executives for their company, they reach out to these social network, which produces, therefore, a series of expectations, obligations, reproduction of social standards and inter-relations with other players (Veliyath & Ramaswamy, 2000). Granovetter (1985) studied how behaviour and institutions are affected by social relations, and this brought into evidence that even in modern organizations, they play an important role. Shapiro (1987) and Cook (2001) portray a quite different strategy in dealing with the agency problem, using the systems of selection, control and sanction of the agents. The authors point out that agency relations are introduced into a permanent structure of personal relations and that these networks of trust can be utilized as a strategy to minimize the risk of wrong selection in the agents’ recruiting process. The agents possess registrations of their career and reputations that can be accessed by the principal through the members of trustworthy social networks, in which both are embedded. However, Shapiro (1987), warns that conflicts of interest are not necessarily extinguished by contracting agents from personal networks, although their effects tend to be attenuated. The author maintains that agents and principal belonging to the same social network are more likely to share similar interests and values, than when they belong to unrelated groups. In addition, the agents would be more likely to act in a more committed, honest and altruistic way when invested with responsibility through the indication of ex-employers, friends, family, neighbours, colleagues, members of their church, and so on (Shapiro, 1987). Shapiro (1987) goes on to say that the monitoring of agents’ behaviour would also be facilitated in close and continuous relationships, where the agents are routinely supervised by the principal or his/her associates. Thus, the social networks 178 offer a rich variety of sanctions for the offending agent, such as: disgrace, ostracism, loss of reputation and other forms of social reproof. Although embeddedness is recognized as a source capable of transferring the trust of the social relations into the company, it also has its dark side. Shapiro (2005, p. 277) explains that because embeddedness is often an excuse to relax vigilant recruitment and monitoring, it provides cover, not only for wayward offspring or relatives, but also for confidence swindlers to feign social intimacy and thereby enjoy unfettered opportunism. For Heidrick (1988) the CEO and the family board do not need to rely on their own social network to fill the strategic positions in the company. In fact, the author suggests that they preferentially ‘avoid involving a close friend who may be as eager to please as family members or employees’ (1988, p.274). Trust between Shareholders and Managers A clash can be seen in the literature between those that adopt blindly the utilitarian presuppositions of agency theory, underpinned by relationships based on formal agreements and whose management relies on rigorous control mechanisms (Corbetta & Salvato, 2004) and those that, following the presuppositions of stewardship theory, focus on the trust systems to manage relational risk, through organizational structures which promote reciprocity, cooperation, social insertion and trust between those involved (Donaldson, 1990; Mayer et al., 1995; Sundaramurthy & Lewis, 2003). Control is about influencing the behaviour of the agents ensuring that they act in an effective and cooperative manner (Das & Teng, 2001; Lebas & Wiegenstein, 1986). Trust on the other hand, is about allowing the other party to act with full discretion ‘based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party’ (Mayer et al., 1995, p. 712). The option for each of these mechanisms to the detriment of the other, can lead to a dysfunctional organizational dynamics (Sundaramurthy & Lewis, 2003). 179 Focusing only on control stimulates myopic decision-making and encourages a polarized board-CEO relationship with insufficient openness for advice interactions (Roberts, 2001; Westphal, 1999). Falk and Kosfeld (2004) also make it clear that the decision of the principal to implement rigid control mechanisms can be interpreted by the agents as a sign of distrust, and may reduce significantly the motivation and willingness of the agents to do their jobs properly. Thus, before adopting any mechanism of governance based on a view that “one-size-fits-all”, the principal should evaluate carefully the psychological aspects and remember that “what works well to control or motivate an opportunistic manager may not work well to control or motivate a steward” (Lee & O’Neill 2003, p. 212). Along the same line of reasoning, Ghoshal and Moran (1996) argue that hierarchical controls can raise the level of opportunistic behaviour and at the same time, diminish the positive feelings that the agents nurture in relation to the company. They assert that the more incentives and sanctions provided, the greater will be the tendency of the agents to behave in an opportunistic manner. For the authors, monitoring and rewards can alter the behaviour of the individual in accordance with that intended in many cases, but their attitude towards the organization will also be affected negatively. In this sense, organizations fail when they are unable to create a social context necessary for building trust and the exertion which are essential for maintaining cooperation, as in inducing the presence of trust, the necessity of monitoring itself is reduced (Ouchi 1980; Dyer, 1997). On the other hand, to base all relationships on trust only can lead to an environment of groupthink where the principal will rarely challenge the positions adopted by his agents (Sundaramurthy & Lewis, 2003). Thus, Bammens et al. (2009) propose that control and trust should be used in a complementary manner. Davis et al. (1997) recommend that such mechanisms should be utilized alternatively, depending on psychological and situational factors. And, finally, Sundaramurthy and Lewis (2003) reinforce the idea that both mechanisms, not only can, but must coexist. 180 V.5.6. The Disclosure and Transparency Construct Transparency and disclosure are concepts essential to corporate governance, guaranteeing an effective allocation of resources in society and lessening the information asymmetry between a company and its stakeholders (Popa and Peres, 2008). Patel et al. (2002, p. 326) confirm this idea stating that ‘higher transparency and better disclosure reduce the information asymmetry between a firm’s management and financial stakeholders - equity and bond holders, mitigating the agency problem in corporate governance’. Andrade and Rossetti (2006) also highlight the fact that the practice of disclosure is closely associated with the search for the minimization of asymmetric information in the principal-agent relationship. For these authors, an agent in his/her fiduciary role has an obligation to act with full transparency to the principal, revealing any information that is relevant. In addition, the agent must disclose to the principal what his/her interests are, when divergent, not taking advantage of his/her position to obtain approval of his/her actions due to lack of adequate information. Therefore, transparency is central to producing reliability, fundamental for less conflicting relationships within the family business system, as posed by Child and Rodrigues (2003, p. 351). Greater transparency in the form of more open reporting of corporate data, monitoring of behaviour, and reporting of concerns, […] provide a more constructive alternative to traditional hierarchical control. It also helps to avoid breaches of trust stemming from unfair, even unlawful, practices that give certain groups undue benefits. This discussion allows the following hypothesis to be postulated: Hypothesis 7: The higher the level of disclosure and transparency of management transactions, the higher the level of trust prevailing in the family business. It is important to note that transparency occurs whenever information is accurately presented, not only including the firm’s financial situation, but also its performance, ownership structure and information regarding the governance of the 181 corporation (Mallin, 2002). Thus, the discussion of this construct will consider the degree of access and the level of dissemination of information present in the family enterprises, as well as questioning the level of transparency in the succession process. The requirement of executives to adopt transparent behaviour is also considered. Figure 5.9: Categories of the Disclosure and Transparency construct Source: Research data The degree of access to information Financial disclosure and business reporting are critical elements for the functioning of an efficient capital market (Popa & Peres, 2008), since they are important information sources for different stakeholders, especially those for public companies. Actually, disclosure is compulsory for this category of firms, since they are required by law to not only file financial information publicly on a periodic basis but also to disclose other information on the company, provide detailed data on new issues of securities and report any trade by insiders. (Arruñada, 2011, p.379). Companies may resort to two classes of publishing variants through which they can reduce the informational asymmetry towards their stakeholders: mandatory and voluntary disclosure. Mandatory disclosure is determined at national regulation level through stockexchange commissions, accounting or government authorities, which usually uses an 182 international referential of business reporting. Its aim is to produce quality information to fulfil the stakeholder’s request in accordance with the applicable laws and standards (Popa & Peres, 2008). Despite its high cost of implementation, mandatory disclosure also encourages the executives to analyse and understand the business results, which means that its cost-effectiveness can be worthwhile (McCahery & Vermeulen, 2006). Popa and Peres (2008) state that it is possible to measure the benefits achieved by the adoption of disclosure for the firms listed through the cash flow and the transaction price increases. Eccles and Mavrinac (1995) studies show that the largest benefit is the increase of the firm’s credibility. They also included in their list of advantages the increase of share value, the increase of the number of potential investors, greater access to capital, the reduction of share volatility, the increase of share liquidity, the improvement of relations with the suppliers and the reduction of political interventions to regulate the market. On the other hand, voluntary disclosure is an additional offer of information, which is made public through the company’s free choice, aiming to attract investors. This is applied especially to non-listed companies, the case of almost all family businesses surveyed in this dissertation, which in most countries, like USA, Japan, Brazil, and others, are not obliged to disclose any financial information (Arruñada, 2011). European Union countries regulate their financial disclosure differently in a very distinct way, requiring all companies to formally submit and publish their accounts. Over recent years, the Commission of the European Communities has made efforts to simplify the business formalities, an attitude, however, that may be counterproductive. Arruñada (2011, p. 378) argues that 'mandatory publication of accounts is not only an issue of reducing the costs of operating businesses but also of easing businesses' access to credit’. After all, says the author, mandatory disclosure of accounts by private companies relates to investors’ protection and credit information, thus helping their trading parties (mainly banks and suppliers) by estimating their credit risk, which in turn, expands firms’ access to credit and reduces its cost. For him, ‘publishing the accounts may be more credible and less costly than communicating them individually to contractual parties or handing them only to those parties who request them explicitly’ (Arruñada, 2011, p. 389). 183 McCahery and Vermeulen (2006, p. 32) suggests which information should be supplied in order to demonstrate accountability to the investors: (1) the company’s objectives; (2) principal changes; (3) balance sheet and off-balance sheet items; (4) financial position of the firm and its capital needs; (5) board composition and company policy for appointments and remuneration; (6) forward-looking expectations; and (7) profits and dividends. Companies that adopt voluntary disclosure using the cost-benefit analysis state that ‘management would publicly disclose up to (or towards) the point where the perceived reduction in the agency costs of equity capital equalled the increased costs of public disclosure to markets and the public domain’ (Holland, 1998, p. 30). Certainly, voluntary disclosure could be an important part of the present corporate reporting practice, for its proposal to cover deficiencies in some of the content of traditional financial reporting, thus increasing the investors' confidence and the market efficiency as a whole. However, Popa and Peres (2008) observed a negative correlation between the volume of information disclosed voluntarily and the reliability of the reports published. This means providing a significant amount of information does not necessarily mean a greater commitment to transparency. In the case of family businesses, a growing number have realized the benefits of a higher level of voluntary disclosure. When comparing family and nonfamliyowned companies, Chen et al. (2008) concluded that the family shareholder tends to have greater influence than other types of investors, regarding the level of transparency adopted in disclosure of the company's results. Silveira (2006) agrees with this view by pointing out the results of a study conducted by the University of Texas at Dallas, with family and nonfamliy-owned companies belonging to the S&P 500. The survey revealed that nonfamliy businesses are less transparent about their governance practices, disclosing less information voluntarily. It also found that family businesses present higher quality financial information and are more likely to warn investors about bad news occurring in the business. The results reinforce the idea that the conflict between managers and shareholders, which is typical in the U.S. market, is lower in family-owned companies, as the closer alignment between ownership and management makes these companies less averse to disclosure of 184 bad news that could affect their profits during the period. Finally, the study also shows differences in the levels of transparency among family businesses whose founder is still active, serving as CEO. These tend to be more transparent than the companies that are in the second or third generations (Silveira, 2006). The level of dissemination of information The level of transparency of the business is intimately associated with the existence of efficient channels of communication and dissemination of its results. Widespread access to information, whether for individual or institutional investors, is a legal requirement (IBRI, 2007). It is worth noting that in the case of family-owned firms where ownership is concentrated in the hands of a single family group, the challenge of being transparent is related to the ability to harmonize the interests between different groups of shareholders, if any, leveraging the asymmetry information and avoiding the expropriation of minority shareholders by the majority (Andrade & Rossetti, 2006). The Brazilian Securities and Exchange Commission (CVM) makes clear in its rules that everyone should be informed at the same time and with the same quality of information, whether through the media, public meetings, websites, etc. The key is that the principle of equal and simultaneous access is respected and all audiences receive the same treatment (IBRI, 2007). In its publication named Guidance on Good Practices in Corporate Governance Disclosure the UNCTAD (2006, p. 29) recommends that ‘relevant information should be available for users in a cost-effective way, preferably through the websites of the relevant government authority, the stock exchange on which the enterprise is listed (if applicable) and the enterprise itself. […] Some degree of harmonization of the location of corporate governance disclosures would be desirable to make the relevant data more accessible. Two possible approaches include putting all corporate governance disclosures in a separate section of the annual report, or in a stand-alone corporate governance report. […] Traditional channels of communication with stakeholders, such as annual reports, should be supported by other channels of communication, taking into account the complexity and globalization of financial markets and the impact of technology’. 185 It also important to note that the CEO and the Investor Relations Director, under the supervision of the Board of Directors have a responsibility to ensure that the disclosure is accompanied by maximum transparency. They should also provide security to the market, tending to safeguard the company's key strategies, without harming the disclosure of their real situation. Regarding instances of governance, shareholders can count on the support of an Audit Committee in order to ensure that the information provided by executives and the board of directors is reliable. As noted by Furuta and Santos (2010), the Audit Committee is a supervisory body mandated, but non-permanent, and should be installed at the request of shareholders. Among its main responsibilities are: to oversee the actions of administrators; submitting opinions on the management's annual report; provide feedback on the proposals of the management bodies to be submitted to the general assembly, relating to changes in capital; issue of debentures or subscription bonuses; distribution of dividends; merger or division; examination of financial statements for the fiscal year; among others (Furuta and Santos, 2010). The degree of transparency in the succession The timing of the succession is still commonly treated by many managers as a family and sporadic transient event, which can lead to great difficulties and failures at the time of its completion (Lambrecht, 2005). Together with the lack of planning, there is a predominance of patriarchal criteria defined unilaterally by one who will be succeeded without the participation of other family members. As Bernhoeft and Castanheira (1995) warn, the lack of a discussed and transparent process can lead to: conflicts of interest among members of the owning family; personal disharmony and distress; isolation and lack of synergy; tense and destructive criticism among stakeholders. Succession planning, a central theme in the discussion on governance in family businesses, has a great potential to assist in the orientation and conduct of succession, contributing to the structuring of a transparent and balanced process (Lansberg, 1999) The item of transparency can be reached from the discussion of the criteria of succession within a family council, open to all members of the owner 186 family. Among the main issues to consider in succession planning, it is possible to highlight: setting rules for the succession; guidelines for training and preparation of potential successors; criteria adoption to guide the professional management of the company; mandatory retirement rules; and definition of policies for financing the economic security of older members (Bernhoeft & Castanheira, 1995). Many authors, such as Lambrecht (2005), emphasize the importance of establishing a participatory orientation, guided by shared criteria for succession. The guidance issued by the UNCTAD (2006) recommended the adoption of transparent procedures to appoint successors, considering that the succession process should be subject to disclosure. It cites a number of codes that reference the succession process, namely:  OECD Principle IV.D.2, which stresses that overseeing succession planning is a key function of the board;  Dey Report (Canada), which considers it an important stewardship duty of the company; and  Vienot Report I (France), which recommends that the selection committee be prepared to propose successors at short notice. The ideal, taking into account the recurrence of the need for succession along the trajectory of the family business, is to consider succession as an ongoing process that satisfies the criterion of transfer to future generations, translating it into a multigenerational transition plan (Lambrecht, 2005). Thus, the family should be able to identify, understand and develop its own way of transferring the organization and its symbolic and real assets over generations. This would imply that rather than thinking about succession, the family, together, should create a standard for the conduct of their succession processes, which is able to guarantee the perpetuation of the company. Transparent behaviour of the executives The level of transparency and information dissemination within the organization is not limited to structuring well-articulated instances and mechanisms of governance but also includes the quality of communication established between the 187 agent and principal. As IBGC (2009) points out, the CEO must ensure that communication is done with clarity, substance prevailing over form. The language should be accessible to the audience in question, addressing both the positive and the negative aspects in order to give people a correct understanding of the organization. However, McCahery and Vermeulen (2006) note that executives tend to be reluctant to disclose sensitive information. Beccerra and Gupta (1999) understand that, in order to adopt the level of transparency required by good practice without the use of gimmicks that effectively bypass it, it is necessary to build trust in relations between managers and shareholders. The authors point out that when there is a strong bond of trust in relationships, there is a tendency to establish an open communication where people do not feel difficulty in sharing information, while relying on the information they receive. In this context, the executives would not be fearful that the information sharing could be used against them. In the opposite case, Beccerra and Gupta (1999) show that in organizational contexts marked by low trust in relationships, executives tend to protect themselves and avoid sharing information that might harm them, adopting more conservative positions in running the company. This type of conduct with its lack of transparency, coupled with a management unable to take risks, clearly goes counter to conventional maximization of the interests of shareholders, which may lead to the problem of agency. In family-owned firms, this question takes on particular characteristics due to the differences in status between family and nonfamliy managers. V.5.7. The Legitimacy Construct The dynamics of family organizations, as stated by Lopes (2008), is notably marked by the peculiarities of the family institution. The emotional aspects of kinship, characterized by solidarity, affection, trust and stability, coexist with conflict, competition and disaggregation typical of capitalist dynamics. The emotional-affective dimension frequently ends by forging relationships in the context of work, such as 188 instances of power and competition, and reflects the standards of management and organizational culture (Lopes, 2008; Davel & Colbari, 2003). In the arena of interactions that arise from the arrival of new generations within the family business, the definition of their roles, their commitment and competence to lead the company's future are crucial for the continuity or disruption of family businesses in the long term (Grzybovski et al., 2008). The expectations and concerns about their ability to take on such a role are only settled with the passage of time, when heir-successors can legitimize themselves with the family and the company. In the Weberian sense, legitimacy can be defined as the acceptance of the exercise of power, i.e., the capacity that an individual shows to cause the other to accept his/her command. And, according to Weber (1971), this can be achieved through tradition, charisma or by law. In the context of family organizations, this issue typically arises precisely when a new family member is incorporated within the company. Despite having the legal authority under the law of inheritance and future ownership, he/she must earn the respect of family and other stakeholders (Barach et al., 1988). These arguments lead to the following formulation: Hypothesis 8: The lower the legitimacy of family members in their entry into the company and at the moment of the succession, the lower the trust prevailing between owners and managers. In this section, the process of building the legitimacy of successors will be discussed from two perspectives. The first deals with the challenges of integration of the successor, the trend of the prevalence of kinship ties in the selection process, the difficulty of overlap with the image of the founder and/or predecessor. In the second, the level of formalization of the succession process is discussed, taking into account the importance of formal succession criteria and the question of recognition of family managers by the stakeholders. 189 Figure 5.10: Categories of the Legitimacy construct Source: Research data Successor’s legitimacy The mere choice of successor by the predecessor is seen, in many family businesses, as a sufficient condition of a satisfactory succession process which will ensure business continuity. However, the process of succession does not end when leadership is transferred from one generation to the next. The successor needs to be legitimized, confirmed by various important stakeholders: key employees, investors, bankers, suppliers, and distributors (Harvey & Evans, 1995). Bernhoeft (1991) highlights how important it is for successful transition that the successor wins his/her own space, presenting results in his/her own right. Barach et al. (1988) comment that even with the legal authority conferred by the family name and the fact that he/she will be the future owner of the company, the successor must be able to prove his/her ability to other managers and gain their trust. These authors summarize this by saying that the parent can designate his/her son or daughter as successor, but cannot force acceptance by everyone in the organization. The successors are challenged to achieve credibility, showing that they are able to perform the job as well as their predecessor had done, reaffirming the idea that the authority and respect inherent in their position have been achieved not simply inherited (Barach et al., 1988). It is not uncommon to find successor resistance from the ‘upper echelons’ in family businesses. Lansberg (1988) explains that such resistance could be based on 190 reluctance to shift from a personal relationship with the founder to a more formal relationship with the successor, a threat to employment due to a change of organizational structure, or to a fear of loss in autonomy and influence. Harvey and Evans (1995) confirm this idea saying that Senior, "old guard," key functional experts within family businesses are the most likely candidates to lapse into circumventing the new power structure. They have a long-standing relationship with the founder […] They have earned credibility and social capital […] These key employees feel that they have earned the right to do what is best for the company and, tangentially, for the family. Their loyalty and feelings for both the business and the family make them quasi-family members. Without attributing any negative motives to these individuals, it can be said that they can undermine the fledgling power base of the next generation. […] The conflict that evolves from this lapse in the chain of command can affect other employees. A key factor in this process is the role played by the predecessor. It can act as a facilitator or make the process even more difficult for the successor. Leone (2005) states that, sometimes, the predecessor is in a dilemma with himself, when he is confronted with doubt and uncertainty generated by the transmission of power. Lansberg (1988) states that the readiness of the founder to leave the family business to a successor may be slowly consumed if the lack of competence and preparation of the potential successor becomes evident. Goldberg and Wooldridge (1993) confirm that the owner-manager's reluctance to leave the command of the company will actually be greater if there is uncertainty about the successor’s skills, willingness and desire to take control. Conversely, if the founder trusts his/her successor, there is a much greater chance that this will effectively take over the business (Sharma, 1997). However, even though the successor has the right profile and has completed a successful succession plan, and can, therefore, be relied upon, the founder can still continue to resist, and moving away from the centre of the arena, prefer to continue to influence the decision-making in the family business. Living under the shadow of the founder generates feelings of frustration and resentment that emerge among the founder’s offspring when they are repeatedly compared to the founder (Dyer, 1986). The higher the charisma and prestige achieved by their predecessor, the greater the challenge of gaining credibility with employees and family members. This continuing influence or interference of the 191 founder in the family business produces conflict, and threatens family harmony (Davis & Harveston, 1999). Research conducted by Davis and Harveston (1999) demonstrated that the conflict is higher in family-owned firms led by the third (or later) generation than those led by the founder or the second generation. This trend is to be expected, due in part to an increase in the number of members and the complexity of the family and the work relationships involved (Aronoff & Ward, 1994; Kets de Vries, 1993). The Figure 5.11 below shows this reality. Figure 5.11: Conflict across Generations Source: Davis and Harveston (1999, p. 318) Another significant discovery of these authors in the same study is that conflicts are manifested more in the presence of the founder’s generational shadow than in its absence. In fact, the presence of the founder still makes it difficult for his/her successor to assume the leadership role effectively, earning the respect and recognition of all. These findings suggest that ‘overall conflict is determined less by the transition or conveyance of leadership of the family firm to the second generation than by the refusal of the founder to let go of the control of the family’s business’ (Davis & Harveston, 1999, p. 320). It is important to note that the problem of generational shadow is sensitive only from the founder to the second generation. From second to third generation and 192 beyond, no kind of significant impact has been found by researchers. That is, although the level of conflict in the third-generation or later - is higher, its cause is not the constant presence or absence of previous generations in the day-to-day business. Figure 5.12 demonstrates this: Figure 5.12: Conflict and the Shadow Effect Source: Davis and Harveston (1999, p. 318) Finally, research carried out by Churchill and Lewis (1983) found that an owner-founder’s gradual withdrawal from the firm, combined with more delegation of responsibility to managers, helps reduce resistance to succession. However, very often there has to be a real withdrawal of the predecessor, because in family businesses, due to the family bond, it is very difficult to delimit the spaces of the predecessor and successor. And even though there is no intention on the predecessor’s part of interfering with his/her successor, the successor needs physical freedom from his/her parents to fully develop confidence and self-image (Handler, 1990). Thus, the idea of changing the predecessors' office space, parking space, conference rooms, or other physical attributes is a symbol of power change that sustains the legitimacy of the successor (Harvey & Evans, 1995). 193 Legitimacy of family managers Heirs, by definition, achieve their positions in the family through birth, not merit (Berenbeim, 1990). But in order to be successful in executive positions in the family business, rather than receive them as an inheritance, they must make themselves worthy. Therefore, it is important to highlight the role of formal preparation and conduct of a planned and transparent succession process. Barach et al. (1988) call attention to the career planning for the next generation as an essential part of the implementation strategy of any family firm. Bernhoeft and Gallo (2003) argue that career planning for members of family businesses in fact will assist in the process of their taking over the top management functions. A leader must acquire knowledge of aspects of his/her particular business (i.e., mission, philosophy, products, production processes, suppliers, clients, financing, technology, and so on); knowledge about the industry in which his/her firm operates (i.e., competitors, financial sources, leaders and important contacts, regulations, history, present and future markets, and so on); besides leadership abilities (i.e., communication, motivation, and other managerial abilities) (Aronoff & Ward, 1994; Foster, 1995; Tsoukas, 1996). Another important recommendation in the training of the heirs is that before working in the company, they should have worked elsewhere before to gain experience. What should be avoided is that they consider the family business as an automatic haven in the event of not succeeding in finding other better positions (Jonovic, 1982, cited in Barach et al., 1988). For Barach et al. (1988, p. 55) work in nonfamliy-owned firms may provide a broader perspective on business and develop capacity to adapt to radical environmental change. Furthermore, achievement on the outside can win the entrant credibility and respect when joining the family firm. Berenbeim (1990) also argues that the heirs should win legitimacy through their achievements in other institutions, where they should compete on an equal basis with other professionals. By walking this path, the heirs tend to gain greater respect from employees and their families and, considered more important by the author, develop greater confidence in their own abilities. 194 But in fact, what usually happens is that most successors join the family firm upon completing their education. Jonovic (1982, cited in Barach et al. 1988) notes that almost 85% of all successors go to work for the family firm immediately after graduation. Even those who choose to do an MBA, usually say they don't intend to join the family firm when they graduate, but as soon as they get their degrees, the majority does precisely that at some point. A recommended option is that descendants should be gradually exposed to the reality of the company, through summer jobs and lower category jobs (CabreraSuárez et al., 2001). Barach et al. (1988) report that the vast majority of family members use these entry models, which are considered effective in achieving credibility with important stakeholders. For the authors, this manner of entry can also provide the successor with valuable knowledge of the firm's operations which helps generate a feeling of confidence and respect. Cabrera-Suárez et al. (2001) also point out that the successor will gain benefits from being familiar with the business and his/her employees, facilitating future acceptance, gaining credibility and building relationships that may be quite important in the future. Figure 5.13: Comparison of entry strategies for succession in family business Low-Level Entry Strategy Advantages Disadvantages 1. Intimate familiarity with the nature of the business and employees is acquired. 1. Conflict results when owner has difficulty in teaching or relinquishing control to successor. 2. Skills specifically required by the business are developed. 3. Exposure to others in the business facilitates acceptance and the achievement of credibility. 4. Strong relationships with constituents are readily established. 2. Normal mistakes tend to be viewed as incompetence in the successor. 3. Knowledge of the environment is limited and risks of inbreeding are incurred. Delayed Entry Strategy Advantages Disadvantages 1. Successor's skills are judged with greater objectivity. 2. Development of self-confidence and growth independent of familial influence are achieved. 3. Outside success establishes credibility and serves as a basis for accepting the successor as a competent executive. 4. Perspective of the business environment is broadened. Source: Barach et al. (1988, p. 53) 195 1. Specific expertise and understanding of organization's key success factors and culture may be lacking. 2. Set patterns of outside activity may conflict with those prevailing in the family firm. 3. Resentment may result when successors are advanced ahead of long-term employees. Barach et al. (1988) present the advantages and disadvantages of an heir opting for immediate entry into the family business via the lower category jobs or delay this entry a little seeking alternative development outside the family business. Figure 5.13, above, presented an overview of this. Barach et al. (1988) also underscore the value of the interaction between predecessor/successor of the same family as a way to bring the successor to become aware of the predecessor's mental processes, ideas, experiences, and so on (Cabrera-Suárez, 2001). Clearly, therefore, the good relationship between them, characterized by respect, understanding, and complementary behaviour is critical to the success of the succession process (Handler, 1989b). However, as noted by Cabrera-Suárez (2001, p. 43) it must be taken into account that father-son relationships are usually characterized by a certain degree of ambivalence; on the one hand, a son identifies himself with his father and wants to be like him and, on the other hand, there are feelings of envy and rivalry that originate in the son’s childhood when they competed for the mother’s attention. In the research carried out by Barach et al. (1988), two patterns of time to achieve credibility with stakeholders by the heirs of a family business were identified: five years for those who followed the non-innovative path and two years for those using innovative behaviour. This, regardless of whether or not they had prior outside work experience. The credibility achieved through innovative behaviour is an excellent alternative in times of change. The authors comment that: there are pitfalls in attempting to be innovative before credibility is well established, but the executives who achieved credibility through initial innovative behaviour brought something to the firm which improved current operations. It is probably easier to do this during times of high flexibility, when the firm can afford to pursue new ideas, or during times of turbulence, when change is necessary (Barach et al., 1988, p.55). Finally, it is essential that the rules of entry of family members in the company are well agreed with all the existing family blocks, with objective and formal criteria, widely disseminated within the family. Lambrecht (2005) states that each family business is responsible for discussing and developing its own criteria, but in general should seek to identify whether the candidate has an entrepreneurial spirit, what is 196 his/her qualification via formal education, external experience acquired in other companies, his/her previous experience in their own family business (i.e. summer jobs) and qualification to perform planning and formal agreements. V.5.8. The Competence Construct Faced with the inexorable process of economic globalization that can be seen at the present time, Gonçalves (2000a) points out that companies, especially family ones, have been obliged to make themselves efficient, in their search for ways of surviving in a world of fierce competition. Competence has become the watchword, representing one of the principal competitive advantages of organizations, and many of those family-owned companies where an informal attitude, paternalism and nepotism were the norm, have been condemned to extinction (Gonçalves, 2000a). Drucker (1977), in relating competent bureaucracy to the modern organization, proposes the substitution of the tycoon by technicians and professional administrators as directors of the private company. In fact, several authors advocate the presence of "outside" directors and managers that can contribute to the company with new expertise and capabilities that are not present in the family business (Westhead & Howorth, 2006). This line of thought leads to the stigma that the family company has been stamped with, for a number of decades: that of a presupposed incompetence (Gonçalves, 2000a). However, for several authors, professionalization is not necessarily synonymous with removing family members from company management. Some family businesses, for example, which have carried out successful transitions, had invariably well prepared their heirs (Morris et al., 1997; Weinstein, 1999). And this is very important, as shown by Ward (1987), for whom successor development is one of the most important features in guaranteeing successful generational transition. In fact, the research carried out by Venter et al. (2005, p. 289) supports the proposition that ‘there is a positive relationship between the preparation level of the successor and the perceived success of the succession process’. 197 In this sense, in accordance with the argument of Lodi (1998), professionalization is the process by which the managers of the traditional or family organization adopt more rational, modern and depersonalized administrative practices; they assume a determined code for training or conduct; they start to substitute intuitive methods by impersonal and rational ones; and pass beyond the archaic or patriarchal forms of contracting work, relying more on salaried personnel. This discussion leads to the following formulation: Hypothesis 9: The lower the level of competence of family managers, the lower is the level of trust present in family-owned companies. This section will adopt two perspectives. Initially, I discuss the aspects related to the level of competence of the managers in family organizations. Subsequently, I deal with the relationship between the levels of competence present in management with the performance of the company, in accordance with that laid out in the following figure. Figure 5.14: Categories of the Competence construct Source: Research data Capacity of the Managers The discussion on the capacity of the leaders in family organizations, usually leads to the debate on the distinguishing features existing between family and nonfamliy managers. For scholars such as Rosenblatt et al. (1985) and Ward (1988), family-owned companies have a unique type of human capital. This can be 198 explained, firstly, by the fact that the name of the family is intimately linked to the company, so that the family members feel naturally more motivated and committed to the business. This type of connection inspires loyalty with family members disposed to dedicate long working hours in very flexible patterns, to achieve the success of the business (Rosenblatt et al., 1985). Additionally, the family members have been, for the most part, introduced to the family company when they were very young, and understand the nature of the business, its customers and competitors, and have received training in the daily routine from the family leaders, who have the knowledge and the abilities essential for running the business (Dyer, 1986, 1992). This type of introductory process can become a significant source of competitive advantage, creating a body of highly motivated family managers that know the business profoundly, and who are also disposed to sacrifice themselves to achieve the growth of the company. On the other hand, it is necessary to bear in mind that family-owned companies possess a limited base for recruiting potential managers. This can result in the family not being able to meet the needs of the organization with the necessary number of managers with sufficient talent to fill the key positions. Companies that need a high level of sophistication or knowledge specialization, such as those with activities in technological sectors, for example, know this situation, at first hand. Thus, perhaps the family does not possess successors who are sufficiently qualified to take over the positions of management, and have to have recourse to nonfamliy managers, selected in the market (Dyer, 1989). If nepotistic practices are present in the organization, the risk is run of filling management positions with incompetent individuals, affecting company performance negatively, inhibiting company growth and generating true disadvantage in terms of human capital. Chittoor and Das (2007) defend the notion that the option for nonfamliy professionals to fill positions of management is justifiable when one considers the number of executives that exist on the market, with similar or, frequently, superior competence to the family members. The authors argue that the professional succession of the management increases the performance and the longevity of the organization, as the contracting of a nonfamliy professional can resolve disputed 199 management positions between the family members, in addition to stamping a new dynamic on the organization, in separating management and property. Installed competence for the delivery of results As has already been pointed out, a high organizational performance can be reached by both family and nonfamliy managers, and the results attained should represent the fundamental criterion for measuring the agents’ level of competence. In the view of Filbeck and Lee (2000), successful management requires the development of strategic plans and systems for monitoring company performance. For Smyrnios and Walker (2003), the contracting of nonfamliy managers that already have competences in strategic management, can represent the appropriate way to lead the company towards successful management. By contrast, the contracting of unprepared family members can generate resentment and act as a discouragement to senior nonfamliy managers, who will start to regard their competencies and merits as going unrecognized. Chittoor and Das (2007) think that the type of manager (family or nonfamily) necessary to raise the performance of the company will depend on the moment in the life cycle of the business. Thus, for these authors, the first years of a family organization represent the phase in which the company would need more of the entrepreneurial drive, present in the figure of its founder, than the managerial skills that can be found in the market. However, starting with the development and expansion of the business, the need for greater managerial capacitation becomes ever more critical, and is the appropriate moment to seek out competence among the members of the family or market professionals to take over the management of the organization. In harmony with this perspective, Miller and Rice (1967) point out that the strong personal ties existing between the family members can reinforce and provide support for the family leader of the company, contributing to better results. On the other hand, the authors also say that, in moments of transition or rapid market changes, management guided by personal ties can be disastrous for the company (Miller & Rice, 1967). Thus, the ties traditionally existing between the family 200 managers and the business can develop into greater resistance when confronted by situations that require a break with family traditions, as well as into less disposition to run risks than nonfamliy managers who do not share these ties. V.5.9. The Opportunism Construct This construct is central in Agency Theory, which describes opportunism as a direct consequence of the asymmetry of interests between principal and agent. For Jensen and Meckling (1976) human beings are rational and utilitarian, very much focused on their own interests and goals. According to this view, it is unlikely that the objectives of others are able to mobilize any individual to the point where he or she is as effective as they would be in the pursuit of their own interests. For Williamson (1985, p. 47) opportunism is explained by the tendency that mutually reliant parties have to ‘mislead, distort, disguise, obfuscate, or otherwise confuse’ each other. Williamson (1975, p. 6) himself had already defined opportunistic behaviour as ‘self-interest seeking with guile’. Lado et al. (2006) states that the presence of opportunism exposes an exchange party to various risks, including from shirking to the exploitation of benefits from the exchange relationship. Given this reality it is important for both parties to develop safeguard mechanisms against such opportunism. Two kinds of actions are considered: ex ante and ex post. Ex ante opportunism is associated with higher costs of writing and enforcement of contracts intended to restrain deceitful behaviour of each party. Ex post opportunism imposes monitoring, the use of incentive systems to promote alignment and other forms of achieving reliable commitment (Williamson, 1985). Alternatively, some authors argue that improving trust between exchange parties is a strategy to be adopted to reduce opportunistic behaviour. Sutton-Brady (2000) for instance, equates opportunism as being inversely proportional to trust. Lado et al. (2008), in turn, present the conventional view that sees opportunism and trust as opposite poles, where the positive effects of trust tend to counteract (or reduce) the negative effects of opportunism within principal-agent relationships. 201 Nooteboom (1996, p. 14) understands that the ‘renunciation of opportunities for opportunism will enhance trust’. Morgan and Hunt’s (1994, p. 25) studies, on the other hand, support the assumption that ‘when a party believes that a partner engages in opportunistic behaviour, such perceptions will lead to decreased trust’. Based on these authors, I can postulate the following: Hypothesis 10: The higher the risk of opportunistic behaviour of managers, the lower the level of trust that prevails in the company. Considering the peculiarities existing in family businesses and the diversity of possible manifestation of opportunistic behaviour, this construct will be elaborated from the following categories: existing attitudes of shirking; white-collar crimes; distortions in executive compensation systems; patrimonialism practices (confounding business and private matters) and, expropriation of minority interests, as shown in Figure 5.15. Figure 5.15: Categories of the Opportunism construct Source: Research data Shirking Assis and Neto (2011) discuss the presence of shirking from the perspective of the concept of gaming, which refers to the situation in which the agents learn the ‘rules of the game’ and start to ‘play’ in search of their own interest, even if this does 202 not correspond to the objectives of the principal. This is manipulation and bad use of the system, especially on the part of the agents. According to Hood (2006), specialists have identified at least three types of gaming and opportunistic behaviour concerning targets. The best known occurs when the agents think that the next targets will be influenced by the better performance that occurred in the past (the ratchet or snowball effect). Thus, they start to believe that if they work very hard during the year, they will only harm themselves, as an even higher target will be set up in the following year. Consequently, the agents tend to reduce productivity in the current period to be more comfortable in the future. A second type of gaming is known as the lower limit effect or threshold effect. It occurs when equal targets are set up for all the units of the system, resulting in there being no incentives to excellence and, also in those that have a better performance being encouraged to reduce the quantity and quality of their services in the direction of the minimum level defined by the target. Finally, the third type of gaming consists in the distortion of the objectives – ‘hitting the target and missing the point’ such as situations where response times are reduced but with a loss of quality of the service. In these conditions, the target definition process can resemble a game between agent and principal, in which whoever is the smarter wins. And the objective of better performance loses in this game. Anti-gaming mechanisms can minimize the occurrence of this phenomenon. Hood (2006) cites the definition of detailed specifications in relation to the indicators and targets, the realization of audits and the exemplary punishment of ill-intentioned agents. However, the implementation of these mechanisms becomes more difficult when we come to consider the reality of family-owned companies, where, frequently, the managers belong to the owning family. Thus, objective evaluation criteria and severe policies of punishment can be left aside, with the intention of maintaining family harmony, even to the detriment of the performance of the organization. 203 White-collar crimes As pointed out by Shapiro (1990), one thing is certain about the agency theory: there will always be agency conflict. The central question is to identify what are the major potential risks. The author tries to distinguish the main forms of manifestation of opportunism by agents. For him, besides shirking and exploiting perquisites, the term guile does not seem to be clear enough to express what agents are up to when they act opportunistically. The extreme boundaries of opportunistic behaviour could be made explicit by the expression white-collar crime. This term was coined by the criminologist and sociologist Edwin Sutherland in 1939, defining such a crime as one committed both by high-status individuals and also by corporations and other legal entities (Strader, 2002). Shapiro (1990, p.250), in turn, defines white-collar crime as ‘the violation and manipulation of the norms of trust — of disclosure, disinterestedness, and role competence’. Thus the white collar crimes in this study are understood as practices of misuse of foreign currency, fraud, misrepresentation and any other illegal means that agents can rely on in order to retain their executive positions, or embezzlement. That is, in this case it is not competence or lack of competence of the agents that has to be kept in check, but moral and ethical conduct. Coleman (1987), based on the interactionist theory of motivation argues that the criminal behaviour subsumed as white collar crime, would be a consequence of the confluence of opportunities of action by the agents coupled with the proper motivation. For the author, the motivation could be explained by the dissemination of a culture of aggressive competition in corporate contexts. The opportunity in turn, would be inversely proportional to the existence of effective control mechanisms (Coleman, 1987). Executive compensation Opportunistic practices related to compensation are found mainly in public companies where, normally, the definition of who is the controlling shareholder is not clear. In many of these organizations, the pay policy is defined by the agents themselves, leading to self-granting of benefits and the provision of exorbitant 204 salaries, by the executives. Moreover, such gains can still be derived from earnings manipulation or artificial valuation of the stock price. According to Oliva and Albuquerque (2007), directors' remuneration should be structured to be bound by the results through smart incentives that contribute to the generation of long-term value. Besides these conditions, it is expected that the person concerned has no power to decide on his/her own remuneration, leaving it to the effective surveillance of other parties. Daily and Dollinger (1992) think that managers are likely to use the firms’ profits to increase their own salaries, hire more staff to make their jobs easier, assign themselves extra benefits and perquisites to increase their own wealth and satisfaction. Owners will seek to maximize firm value and realize gains directly (Demsetz 1983). Wood Jr. and Picarelli Filho (2004) list the core elements of the compensation system as being: (1) fixed remuneration (functional, by skills, competencies and benefits including pension), (2) performance pay (variable, including payment in equities) and (3) the special form (created by the scope of the exceptional results). The conclusion is that the definition of remuneration of the executive is a strategic issue for the company. Thus, the recommended practice by IBGC is the establishment of a compensation committee linked to the board to take responsibility for the subject. Patrimonialism Following the Weberian sense, patrimonialism can be defined as a form of exercise of domination by an authority whose legitimacy is based on traditionalism – ‘this is like this, because it always was” - and whose main characteristics lie in the individual power of the ruler, who, supported by his/her administrative apparatus and recruited solely on the basis of personal criteria, exercises political power in a given territory (Weber, 1999). The remarkable feature of this doctrine would be the absence or weakness in the distinction between the boundaries of public and private goods. In the family 205 business, the meeting between these two spheres, the private (family) and public (the company), may lead to the practice of patrionalism, which is named as a characteristic feature in many Brazilian family-owned firms. In this sense, the patrimonialist vision would allow the family to manage the personnel and company assets as if they were the same. Thus, the lack of boundaries between family assets and the company makes many family leaders consider the organization's assets, including those included in fixed assets, as personal property, while the liabilities are defined as obligations of the company (Gonçalves, 2000a). In an extreme degree, a patrimonialist attitude of managers may even lead to failure of the family business, or even can lead to major distortions commonly associated with family businesses, such as authoritarianism nepotism, use of personal confidence, and not of competence, for the choice of collaborators, and also as a form of paternalistic relationship with employees (Gonçalves, 2000b). Interests of the minority shareholders La Porta et al. (1998) argue that the main form of ownership structure found in the world is the one concentrated in a few majority shareholders, rather than pulverized among various owners. From this perspective, the agency problem ceases to be the conflict between managers and owners, and comes to be between minority and majority shareholders. If in a pulverized structure managers have incentives to act opportunistically, in concentrated structures the controlling shareholders are those who have a tendency in this direction, influencing the control of the company to their advantage. According to Srour (2005), the conflict of interest between the controlling shareholders of the firm and the other minority shareholders is based on the various ways in which the former can divert corporate resources, i.e., capturing the benefits of the firm to their sole benefit. The most common opportunism in this case are: transfer of resources from the firm in transactions with other companies, in the interest of the controller; preparation of fraudulent contracts; exclusive access to insider information; granting of loan guarantees for personal business dealings; expropriation of business opportunities; and others. Johnson et al. (2000) show, from 206 famous examples of firms from Italy, France and Belgium, how it is possible for the expropriation of private benefits by the controller can be done within the limits of the local law. Given the practices of expropriation sometimes used by majority shareholders, corporate governance rules are establishing themselves as key mechanisms for the protection of minority shareholders. As the expropriation of minority rights can occur in many ways, the legal framework assumes a key role in curbing these distortions (Srour, 2005). The main rights covered by a genuine policy of protection of minority shareholders are: the receipt of dividends on predetermined dates; participation in meetings and new subscriptions; the right to elect representatives to the Board of Directors; obtaining the same rights on the sale of the company as its controllers (tag-along) and others. According to Coffee (1999), the perception that stronger legislation would promote clear benefits for market efficiency has led to a convergence of the right of minority shareholders worldwide. Countries with little tradition in the protection of minority shareholders are beginning to make adjustments in the laws that would promote their participation in the affairs of the firm and impose, normatively, more rigid practices of corporate governance. V.5.10. The Impartiality Construct Despite the fact that the key executive positions are usually held by family members, most family-owned firms also hire nonfamliy managers. Therefore, choosing qualified nonfamliy employees and nurturing a shared vision and commitment are important factors for a firm to accomplish its goals. According to Barnett and Kellermanns (2006, p. 838) this becomes ‘more difficult if they do not perceive that decision outcomes, decision processes, and decision makers are fair or just.’ Lind (2001, cited in Barnett and Kellermanns, 2006) presents his fairness heuristic theory suggesting that whenever hierarchical relationships prevail, employees experience lack of self-confidence about their value and job status. This is quite understandable as nonfamliy employees are part of the business but not of the 207 family system (Mitchell et al., 2003). Therefore, they tend to compare the treatment they receive in situations such as performance appraisal, error handling, promotion and compensation with that dispensed to family members. Sir Adrian Cadbury stresses that ‘for both family and nonfamliy members, it is essential that rewards, whether financial or non-financial, are distributed fairly and transparently and accounted for in a clear and precise way’ (Cadbury, 2000, p.18) What is known is that family-business interactions may produce an environment that encourages favouritism which may lead to perceptions of unfair treatment among nonfamliy employees (Lubatkin et al., 2005; Schulze et al. 2003b). Baldridge and Schulze (1999) also see the risk of appearance of unfairness when the limits between family and business units overlap, resulting in agency conflicts. This is proved through Cropanzano and Greenberg’s (1997 cited in Barnett and Kellermanns, 2006) research which reveals that the inadequate use of authority by family members may cause agency conflicts as a result of nepotism, moral hazard, free-riding and adverse selection. This allows the proposal of the following assumption: Hypothesis 11: The more impartiality in the treatment of family and nonfamliy managers, the lower the agency conflict prevailing. The Impartiality construct comprises the subsequent categories showed in Figure 5.16. Figure 5.16: Categories of the Impartiality construct Source: Research data 208 Error handling According to Greenberg (1990) the perceptions of fairness always involve a comparison of any input or output with a reference. Thus, within the company, an employee tends always to compare the treatment he/she receives with that dispensed to another colleague (Cropanzano, 1993). The reality found, in most family businesses, is that preference is given to family members in decision-making processes (Dyer, 1986). The studies of Shaw et al. (2003) show that whenever consistent and unbiased practices are used in familyowned firms, nonfamliy employees’ perceptions of fairness are likely to be beneficially affected. Thus, the reverse is also true, leading to the emergence of inherent agency conflicts. Among the practices that lead to a greater difference in treatment between family and nonfamliy employees, there is the handling of errors. The most plausible explanation for this finding is the presence of altruism in the existing family relationships that strongly impacts the family’s ability to judge another family member, especially between parents and their children. This reality will be discussed in the analysis of the specific construct. Another factor that may explain differential treatment in the existence of an error is the phenomenon of social psychology called the actor-observer bias. He refers to the tendency that people have, in the same situation, to assign his own reaction to external causes, while attributing other persons' behaviour to internal causes. Basically, people tend to make different attributions depending upon whether they are the actor or the observer in a situation. The actor-observer bias tends to be more evident in situations whose outcomes are negative. This means, in a situation where a person experiences something negative, as an actor, he/she blames the situation or circumstances. When the same negative thing happens to another person, he/she, as an observer, often blames the individual for his/her personal choices, behaviour and actions. Taylor and Koivumaki (1976) have found that this kind of bias is usually caused by a lack of information about the needs, motivations and thoughts of these individuals. This makes it understandable that actor-observer bias is less observed in 209 close relationships, such as with family members, where people tend to use the same excuses that they would for themselves (Fiedler et al., 1995). Performance appraisal It will be discussed later how altruism affects the capacity of owning families to effectively evaluate family members (Dyer, 2006) and dismiss them (Gomez-Mejia et al., 2001). This is dealt with by Lansberg (1983) who points out how difficult it is for founders to evaluate the performance of a close relative who works in the firm, particularly their own offspring. For this author the very concept of appraisal (that is, objective assessment of an individual's contribution and worth) in the context of a family system seems a preposterous idea. In a family system individuals are, by definition, seen as ends in themselves. The standing of an individual in a family is determined more by who the individual ‘is’ than by what the individual ‘does’. Applying a set of objectively derived criteria to evaluate a family member's performance goes against the very principles that regulate and define social behaviour in the family (Lansberg, 1983, p. 43). This phenomenon can be found even in a family company which decides to adopt governance mechanisms, such as a Board of Directors. In family-owned firms, board members show less disposition in monitoring a family CEO, on a more rigorous basis, assuming that a non-conducive environment is the real cause for any failure, instead of presuming possible incompetence of the CEO (Gomez-Mejia et al., 2003). On the other hand, several authors highlight how congruent a performance appraisal process is with the objects of economic productivity of companies (AmbaRao et. al, 2000; Barrett & Kernan, 1987; Eberlin & Tatum, 2005). Lansberg (1983) proposes that seeing individuals more as ‘means’ than as ‘ends’, allows firms to recognise those who contribute most to the attainment of organizational objectives. According to him, in the family system this perspective is completely different, causing a stressful situation for a founder if he has to conduct an appraisal of managerial competence of his/her own sons/daughters. As a consequence, founder-managed firms adopting formal performance appraisal processes is rarely found (Astrachan & Kolenko, 1994). They usually prefer a centralized assessment system which identifies individuals for membership of the 210 family rather than for their contribution to the achievement of results (Dyer & Handler, 1994). The problem, however, is that such procedures lead to the perception of unfairness on the part of those who really make a difference. Barnett and Kellermanns (2006) confirm the idea that fair processes of performance appraisal, promotion and compensation can be important influences on the employees’ perceptions of justice. Absence of nepotism Nepotism is a result of altruism, common behaviour in family-owned firms, referring to the practice of hiring and promoting relatives usually in an illegitimate or unfair manner. Kruger (2003) defines it as a ‘kin selection’: the tendency to display bias in favour of family over nonfamliy members. For Kets de Vries (1993) nepotism contributes to the selection of family members to key positions, who are not necessarily competent, encouraging the adoption of unfair reward systems and making it difficult to attract professional managers. Gersick et al. (1997) comment on the existence of studies which show that in the presence of nepotism, firms have difficulty attracting and retaining qualified employees. Ward (1987) explains that since key positions are often reserved for family members, nonfamliy members may feel that their promotional opportunities are limited and prefer to quit the company as soon as this become possible. Lansberg (1983) in his studies points out the negative impacts of nepotism on companies’ management and results. On the other hand, some authors see benefits in the presence of nepotism. Nelton (1998), for example, agrees that nepotism can harm a company, but thinks that it can also help a firm´s development: all depends on how one manages it. He quotes Leon A. Danco, who in his book Beyond Survival wrote: Continuity of the family company is the justification for nepotism. I do not mean blind, untrained, irresponsible, immature nepotism. I mean planned, trained, and responsible nepotism. Whoever the nepot is - family or nonfamliy - this takes planning, and faith, and guts (Nelton, 1998, p.72). 211 V.2.11. The Altruism Construct The term altruism was coined by Auguste Comte in the nineteenth century, and its maximum - to live for others - become the basis of the moral doctrine of positivism8 (Alves et al., 2004). There are different definitions of altruism, as the science used as a basis for analysis. For theology, for example, it is a moral value that motivates individuals to act on behalf of others without any expectation of reward (Batson, 1990). According to this current, altruism is a gift from God and therefore an exogenous expression to individuals who may be acquired by means of moral education. From the standpoint of sociology, altruism presents itself as characteristic endogenous to an individual, based on feelings, instincts and emotions (Piliavin & Charng, 1990). Within the economic concept, altruism is the utility function that positively links the welfare of a person with the other (Schulze et al., 2003a; Bergstrom, 1989). In the family environment, altruism leads parents to transfer resources, tangible or intangible (e.g. love and security), to their children, acting with generosity to them. This feeling helps promote and maintain family links, creating incentives for the offspring to develop reciprocity, love, respect and trust, and encourages family members to show consideration for each other (Ling et al., 2001). Eshel et al. (1998) confirm that altruism encourages parents to care for their children, giving a sense of value to being a member of a family. Hunter (2007) indicates the presence of altruism when parents transfer family wealth to their children, hoping that the dynasty is benefited. Becker (1981), in turn, argues that it is common for an older generation to sacrifice itself spending less to invest in the education of its children, hoping that the family will benefit long-term. Carney (2005) transposes the idea of transfer of benefits between generations, 8 Positivism is a philosophical movement founded by Auguste Comte, characterized by an unshakable confidence in the sciences. For all the positivists regard to human knowledge can be systematized according to the principles adopted as a criterion of truth in the sciences and biology. Positivism compares society to a biological organism, in which no party has an independent existence. In a positive stage, close to perfection, there would be no place for individualism, only for the development of solidarity and altruism of each one, in favor of the community. 212 promoted by altruism, to demonstrate the willingness of family members suffering hardship in the short term, to benefit the company's survival in the long run. For Ling et al. (2001) and Van den Berghe and Carchon (2003) altruism produces this level of solidarity in family groups, which leads their members to accept the goals of the family unit as their own, without having to take the agency costs. The authors point out that altruistic behaviour increases a sense of community among family members, reducing asymmetric information between family members, which in itself already reduces agency costs. Therefore, altruism appears to encourage the family bond, raising loyalty, interdependence and commitment to the family's long term objectives (Ward, 1987). Thus, altruism is an element of reducing conflict in family businesses (Kellermanns & Eddleston, 2004). For another school of thought, however, altruism is a cause for agency costs. Schulze et al. (2003a) are examples, pointing as evidence, the biased perception of parents about their child-managers, even when they show a lack of competence or commitment to the company's future. To Lubatkin et al. (2005) this approach inhibits the ability to monitor them, creating a distinction as to the other employees, and how they have their performance evaluated, how disciplined they are and what criteria are used for their promotion. In the same direction, Gomez-Mejia et al. (2001) show that altruism leads to a preferential treatment of family members which, in its turn, leads to increased agency conflicts. This leads to propose the following hypothesis: Hypothesis 12: The higher the altruism existing in a family business, the lower the presence of impartiality in the treatment of family and nonfamliy managers. Buchanan (1975) also reinforces the negative side of family altruism showing how this can lead children to be selfish individuals, a phenomenon that he identifies as the 'Samaritan's dilemma' - the parents continue to be generous with their children, even if they do not show gratitude or see what they get only as something they are entitled. For the author, the Samaritans live this dilemma because they seek 213 the instant gratification that comes from the act of sharing with their children. In this case, the immediate need for love makes parents tend to spoil them, feeling that it's too painful to deny forgiveness to children misbehaving (Ling et al., 2001). Buchanan (1975) calls these kids 'spoiled', Becker (1981) of 'rotten' and Bergstrom (1989) of 'prodigal sons'. Thus, altruistic behaviour can lead to total dependence on the generosity of the children from their parents, causing them to squander the wealth in the hope that they will be redeemed when needed (Bergstrom, 1989). You can also triggering new conflicts in the business environment from the moment that the 'prodigal son' commits future assets of other family members who do not have the same opportunistic behaviour (Lubatkin et al, 2005). It can be concluded, therefore, that altruism may induce family opportunism, leading to the postulation of a new hypothesis: Hypothesis 13: The higher the altruism existing in a family business, the greater the opportunism identified in the behaviour of the family and/or the nonfamliy managers. Thus, altruism here is assessed from three different categories, as shown in the figure below: Figure 5.17: Categories of the Altruism construct Source: Research data 214 Firm main purpose The structure of incentives in family businesses is more complex than it needs to be in public companies since it has to look beyond the objective of generating wealth for the owner's family, considering also their intangible goals. In his dissertation, Zellweger (2006) points out that in addition to economic aspects, other goals may be present that are not easily perceived by nonfamliy managers. The author cites the example of hiring an outsourcing service from an external supplier which can bring positive financial results, but on the other hand, can compromise the family's social status within the local community. The family business literature discusses the need to broaden the understanding of the objectives of the owner families that usually transcend purely economic ones. Chrisman et al. (2004), for example, call attention to actions that are deemed responsible for the generation of agency costs in nonfamliy businesses, but do not correspond exactly to the family business reality that does not have financial goals as the main target. The controllers can see the family business as place to house the family members, generate some kind of family cohesion, or even act guided by philanthropy or environmental preservation (Chrisman et al., 2005), even if it means taking financial performance to a level not maximized (Chrisman et al., 2004). So if a family decides to hire its less capable members, even with future consequences in the loss of value of the company, this would be considered agency conflict in nonfamliy businesses, but not in a family business. Thus, for a correct analysis of agency conflicts in family businesses, it is necessary to understand the interests of its owners, seeking to understand their decisions and actions taken and identify the activities, incentives, policies and structures that they develop to prevent their interests are challenged (Chrisman et al., 2005). Altruistic selection We identify altruistic selection as an inducer of agency conflict that occurs at the time of hiring, where the principal believes that when selecting a family member as an agent, he/she will undertake all necessary actions to produce the desired results. But, once chosen, there is no guarantee that the agent will have the ability, 215 the commitment or the ethics expected to perform the job for which he is being hired (Moe, 1984; Chrisman et al., 2004). In fact, family-owned firms present agency conflicts related to altruistic selection, both induced by the appointment of relatives to key positions, which end up seeing the company a natural choice for the development of their careers, and by their lower ability to attract outside talent when compared to nonfamliy businesses. As for the difficulty of attracting outside professionals, Lubatkin et al. (2005) identified two basic reasons: first, the family-controlled companies are not willing to dilute the control actions such as offering bonuses to its executives, which makes them less attractive, and secondly, the lack of impartiality, with opportunities of promotions being offered to family members, to the detriment of those who have a better profile and skills for the job. Moral hazard Moral hazard arises when individuals or institutions do not assume fully the consequences of their actions and therefore has a tendency to act less carefully than they should, leaving the other party to assume the responsibility of the effect of its actions. According to Kiser (1999) in the principal-agent relationship, the agents always have better information about their actions and intentions than the principal, a view ratified by Lubatkin et al (2005) who adds that this fact may lead to an opportunistic attitude. For Bergstrom (1989), when sons and daughters belonging to the owning family are working in the family company, agency problems may arise because children can manipulate parents by working too little or spending too much. This author poses that this manipulation creates information asymmetries that hinder the action of parents to positively influence the conduct of their children. In fact, such behaviour may be the result of a permissive education, where even if the intentions were the best possible, parents may have encouraged their children to free-rider (e.g., leave an assigned household chore for a parent to complete or be a spendthrift with their parent’s money), shirk (e.g. misrepresent their actions), and demonstrate other forms of self-centred behaviours (Lubatkin et al., 2006). All of this misbehaviour 216 is expressions of moral hazard, leading to doubt how trustworthy are the family managers in ‘doing what they say they are going to do’. (Alchian & Woodward, 1988, p. 68). V.2.12. The Comprehension of Social Roles Construct A family company can be seen as a unique combination of rules and expectations characterized by a hybrid context, which gathers together at the same time market issues and family demands (Flemons & Cole, 1992). Leone (2008) points out that the family and the company have different purposes. According to her, while the family is a community of life based on bonds of affection and strongly rooted values, the company has a pecuniary objective, and submits to economic and rational principles. Because of this, the delicate challenge of distinguishing family roles from the professional responsibilities in the daily routine of these organizations arises. However, in accordance with Flemons and Cole (1992), there is no clear frontier objectively separating family from business. This intimate relationship between the two spheres can, without doubt, produce benefits, increasing trust and the level of commitment of the family managers to the company. However, this overlap can also feed problems and distortions, bringing with it disputes caused by conflicting interests which tend to overflow from the family into the business, in this manner worsening agency conflict (Miller & Rice, 1988; Kets de Vries, 1993). Examples of these tensions occur when parents are unable to distinguish between their roles as family members and as employees of the family business, adopting opportunistic behaviour by working less than expected, attitudes of shirking, and thinking that nothing can happen to them because of the protection afforded by kinship ties. They may also seek privileges, requiring compensation inconsistent with their skills and which they would never receive elsewhere in the market, or they may misappropriate the company's resources for use in situations of a private character. 217 Parents who act guided by altruism and, because of this, fail to realize the limitations of their children relative to the functions assigned to them, or who tend to justify their opportunistic behaviour, being totally partial in the treatment of error, performance appraisal and promotion to key positions, actually cannot separate out the feelings imposed by family relationships from the impartiality required of a manager. This discussion leads to the following formulations: Hypothesis 14: The lesser the degree of comprehension of their social role, by owners, the greater the risk of opportunistic behaviour. Hypothesis 15: The lesser the degree of comprehension of their social role, by owners, the greater the presence of altruism within the company. Hypothesis 16: The lesser the degree of comprehension of their social role, by owners, the lesser the impartial treatment afforded to family and nonfamliy managers. The understanding of the social roles will be approached from four directions. The first discusses the level of separation between the demands of the controlling family and business matters. Then the role fulfilled by the person succeeded faced with the succession process is dealt with. Next, the importance of evaluating the performance of top management is discussed and the level of agreement existing on the topic among the family managers. Finally, the new responsibilities of the shareholders who have withdrawn from management is analysed, discussing the functions and limits of their roles. Figure 5.18, below, presents the four perspectives cited. 218 Figure 5.18: Categories of Comprehension of Social Roles construct Source: Research data The distinction between family and company The perception of the family company as an overlapping system of different spheres is portrayed through the traditional model proposed by Tagiuri and Davis (1996), which has been already mentioned. This model has the merit of demonstrating the complicated overlapping of roles of individuals that are frequently linked to more than one sphere simultaneously. This complex web of relations, translated by the interaction of the different roles, interests and expectations, can lead to several conflicts, either between the different generations of the family, between separate groups of shareholders, or also between family and nonfamliy managers. It is, however, important to reflect on the usefulness and the limits of the model described. If, on the one hand, the three circles model makes possible a clear visualization of the insertion of the various subjects in the principal subsystems of the organization, on the other it runs the risk of interpreting the reality of family organizations in a static and overly simplified manner (Pereira, 2010). In relation to the distortions that can arise from this conjunction of roles, authors such as Gonçalves (2000a) point out the question of confusing company with family matters. Thus, such practices, given as a characteristic feature in numerous family organizations, would be triggered by the encounter of the private sphere (the family) with the public sphere (the company). This attitude allows the family to administer personal patrimony and that of the company as if they were one and the same private thing. The lack of defined limits between family property and company assets results in many family leaders considering the assets of the company, 219 including the fixed assets, as personal property, while the liabilities are understood as company obligations (Gonçalves, 2000a). This attitude can generate enormous losses for the organization, and may, in extreme cases, result in extinction. Another fundamental point, relates to the interaction between the family members of different generations. In this case, the challenge is to succeed in separating the family roles from the hierarchical obligations inherent in the roles of board member or CEO, who frequently have family members as subordinates. The risk is that of straying from professional and formal behaviour, in favour of a paternalistic posture in relation to the family members, which can result in either protectionist attitudes or criticism and censure. Bernhoeft and Castanheira (1995) state that company management is not strengthened by conflicts, but, above all, by the ability of the controlling family to administer them properly. On this point, it is argued that the development of a governance structure for a family company does not seek to eliminate clashes, but rather mediate them, and establish the clear delimitation of family roles by means of the creation of mechanisms and instances. Change of role post-succession For Davis and Harveston (1999) the founders are the essential actors in a family business since the technical skills they bring to the development of the company are unique attributes that make them distinct in the market (Carroll, 1984). Moreover, in constituting a company, they impress on it values that are the foundation for defining its corporate identity, guiding the formulation of the mission, goals, strategy and structure, and also establishing its operational standards. But Fenn (1996) draws attention to the founders' importance that goes beyond its managerial effects, to influence also the company's main social aspects. Owner-founders are portrayed as having high levels of commitment (Cyert & March, 1963), entrepreneurial skills (Davis, 1968), and strong personal ties with employees (Schein, 1983) that make them difficult to replace […] For these reasons, the transition between the founder and the next generation of leadership is often seen as the most critical and tumultuous (Davis & Harveston, 1999, p. 313). 220 Kets de Vries (1993) sees the founder-owner playing a special role in promoting peace and harmony in the family business by creating a common purpose among the family and employees and establishing the prevailing attitudes, norms, and values. Obviously, in the absence of the founder, the level of conflict is expected to be increased. At this time of transition, the founder-owner can help minimize conflicts arising from his/her withdrawn, or even expand them. Leone (2008) points his/her resistance as one of the situations that may raise insecurity, when he/she does not easily accept giving up power and having only a seat on the Board of Directors or leaving the company once and for all. If the predecessor remains highly engaged in the day-to-day activities of the family firm, giving the impression that he/she still maintains a "controlling" interest in the family business, ready to exercise the power to veto any decision taken by the successor, he/she will weaken the successor's credibility and leadership in relation to other employees (Harvey & Evans, 1995). Gersick et al. (1997) recommend that the predecessors should develop and train their potential successors for management and set up a process for the selection of the ablest leaders. They would therefore have to overcome any resistance to giving up power and help the new leaders to establish their authority in relation to the stakeholders of the company. However, the recommendation of the abovementioned authors should be seen as a problem faced with the need to ascribe a new significance to the role of the predecessor, which is usually responsible for this process being full of difficulties and resistance. The difficulty of the predecessor in understanding what his/her new role shall be with the family and the business, in accordance with Leone (2008), is overladen with strong emotional characteristics. He or she frequently feels that it is still too early to touch on the question of his or her removal, putting it off indefinitely, while the successors, through filial piety, avoid mentioning it. To this can be added the fear of seeing his/her successors destroy a dream constructed with much sweat and toil (Leone, 2008). Thus, according to the author, giving up one´s "life´s work" to the successors can be interpreted by the person being succeeded as a loss of identity, 221 as a loss of his or her usefulness, of the power of decision and the fear of no longer being listened to and respected. Oliveira Jr. (2009) points out that there also exists the fear of being pensioned off. To understand this fear it is necessary to understand that the person´s role in the company has been historically constructed in relation to the business, translating into the essence of value in the company. That is, for many founders who have been succeeded, work and company represent the meaning of life. For Bowman-Upton (1991), the intense involvement of the entrepreneur with the company increases the importance of the work in the constitution of social identity. Thus, ceasing work could be seen as synonymous with the loss of a part of oneself, having traditionally proportioned, status, respect, personal realization and financial return. Oliveira Jr. (2009) argues that, during the period of succession, there exists a greater propensity for conflicts to occur and different types of resistance to appear, involving both family members and nonfamliy managers that work in the company. In this context, it is indispensable that the predecessor is personally persuaded of the importance of participating directly in the process. Gonçalves (2000a) complements this, affirming that the play of the family unconscious tends not to prejudice the company while it is under the command of a patriarch, who uses this role to resolve conflicts or discipline them. For Bernhoeft (1996) the start of the discussions and analyses of the theme should be done by the actual person being succeeded. So that if the process is started when this person is still living, there exist good possibilities of success, demanding much renunciation and the ability to understand that his/her work and ideals should outlive the person concerned. In this sense, the awareness of the leader that he/she is not eternal and that the succession is part of the company trajectory would be the first step in overcoming his/her own resistance. After facing up to these psychological barriers, seeking a solution, the predecessor should define the future of the company, through a planned and organized succession process (Leone, 2008). For Oliveira Jr. (2009) it is fundamental that leaders of family-owned companies are aware that, in planning and organizing their succession, at the same time, the risks and the costs of the process will be minimized. Thus, a planned succession process, in which the predecessor 222 assumes a central role of one who facilitates and makes the process more dynamic raises considerably the chances of a smooth, less costly transition and one that guarantees long term efficiency and organizational competitiveness. Another important point of discussion is whether the retired CEO should assume the role of chairman as part of the succession transition. For Lane et al. (2006) the former CEO can stay on as chairman without undermining the authority of the new leader, only taking care not to feel compelled to demonstrate the possibility of his/her return to "rescue" the organization in case of an emerging crisis. Evaluation of performance of the CEO One of the managerial challenges for contemporary companies is to create systems of evaluation of performance at all hierarchical levels, to be utilized as criteria for evaluating remuneration, promotion and profit sharing and also as measures of efficiency analysis and operational and administrative effectiveness. In accordance with that envisaged in the Brazilian Code of Best Corporate Governance Practices, reissued by the IBGC (2004), it is expected that there exists a formal evaluation of the CEO, guided essentially by criteria of competence. In the United States, in accordance with Epstein and Roy (2005), the SEC Securities and Exchange Commission and the NYSE - New York Stock Exchange have specified requirements for the listed companies, in the attempt to define criteria to be used in the evaluation of CEO performance. According to Oliva and Albuquerque (2007) such indicators should be founded on observable behaviour such as results, degree of transparency and level of effectiveness of corporate governance practices. As, however, the practice of evaluation can run up against cultural questions, there may be the concern that this process could be interpreted as a professional or personal offence on the part of the person evaluated. Considering more the reality of family-owned companies, it can be seen that the question of the evaluation of the performance of top executives becomes even more complex if the fact that the CEO frequently belongs to the family is considered. In part, this question can be clarified by the patriarchal tradition, outstanding in the 223 culture of various countries such as Brazil, in which the head of the family has historically filled the social role of provider and manager of the family space (Freyre, 2003). In this context, the patriarch occupied a central position and the other components of the family possessed very well defined roles. Thus, the patrimony was carefully preserved, as the inheritance appeared as a guarantee of the perpetuation of the patriarch’s authority and the quality of the family relationships was measured by the respect shown to him (Freyre, 2003). Considering the strong reflexes of this social standard in family-owned companies, it is not surprising that a CEO, member of the controlling family, should reject having his/her performance evaluated formally and that, out of respect or fear, he is not confronted by the other family members. In accordance with that affirmed by Schein (1985) and Leach (1990), the founder usually concentrates the power and authority in the organization, in addition to representing the moral reference within the ambit of the family. In this sense, for the authors, this agglutinating force of the family leader tends to decrease with each generation that assumes the command of the company. Thus, companies whose second or third generation is in power would have greater probability of successfully structuring performance evaluation systems. The best way of evaluating the CEOs and the boards, among other actions, is to hear the stakeholders, evaluate the level of fulfilment of strategic directives, check voting histories, check the minutes of the meetings, check how the decision-making process has been conducted and what actions are foreseen to improve performance and increase the transparency of the organization (Oliva & Albuquerque, 2007). However, in accordance with Steinberg (2003) and Santos (2000), the evaluation of CEOs and of boards in Brazil has not occurred in practice or, when it does, is not formalized. Role of the family members outside management Family members exercise considerable influence on the decision-making process in family-owned firms and this interference should be analysed in relation to the distance that these members are from the management (Dyer, 1986; Astrachan, 1988). Those owners outside management should develop trust in the managers, 224 family or otherwise, and delegate to them the accountability for the company operations, creating appropriate supervisory channels without interfering with day-today business. (Lane et al., 2006). However, the task of delimiting roles and disciplining the family members of the organization is quite challenging. In accordance with the warning of Bernhoeft (2004) this difficulty can be aggravated when the business represents the only means of sustenance for the family. Of course in the start-up stage of the company, the family plays an important role in supporting the entrepreneur, that which can remain relevant in subsequent years, expanding to a level where it becomes a source of professional advice and ideas. Anderson et al. (2005) in their observations, found that family members outside the business continued to act as key network contacts and provided a range of assistance to entrepreneurs […] For example, retired parents continued to provide hands-on assistance, helping with day-to-day business tasks in pressing times. (Anderson et al., 2005, p. 164). These authors list the benefits from the support provided by family members who are not part of the management of family businesses: the high quality of the help provided; the different resources and viewpoints made available; the rapidity of the service provided and its low or non-existent cost. Despite these advantages, one must observe the risks of the entrepreneur being limited to seeking support only in the family, due to the likelihood of homogeneity, which in turn dilutes the richness of information and other available resources. It must also be noted that there are two groups who although not part of the business operations, should be considered, as they can exert pressure within the family that affects the business. The first consists of retired relatives and siblings, whose major concern is usually the income provided by the business: thus, anything that threatens their security may cause conflict. The second group consists of children and in-laws, which require attention that competes with the one required to grow the business profitably and that tend to complain that their own interests and desires are not met (Bowman-Upton, 1991). 225 V.6. Development of the Questionnaire Considering the characteristics of the present study, the survey was the next step, one appropriate for the data collection and for testing the proposed hypotheses and the theoretical model. And, since it involves the need of a large sample size in a large geographical area, such a continental country like Brazil, the questionnaire is the natural tool to be chosen. The questionnaire design is a very important step since its quality is fundamental for the success of a research study. Following Bagozzi’s (1994) perspective, five steps are considered necessary for the successful development of questionnaires: (1) problem definition; (2) preparation of a questionnaire draft; (3) execution of a pre-test and evaluation of the results; (4) final revision; and (5) questionnaire’s application. V.6.1. Problem definition Bagozzi (1994) states that the first step in research is to define well the problem one intends to solve or the theory one wishes to test. Therefore, the first step described in this chapter was the conceptualization of the model and the development of the research hypotheses, which specify what information was needed in order to achieve the objectives of this study. V.6.2. Preparation of a questionnaire draft To prepare a draft of a questionnaire it was necessary to define how to measure each conceptual construct proposed in the theoretical model. Next, the problems relating to the question wording and sequence are discussed. Finally, the design of the questionnaire is described. The constructs’ operationalization At this stage, the forms of measurement of the constructs developed in the conceptual model were defined. Constructs are operationalized by selecting the measuring scales for each item and the scale types (Hair Jr. et al., 2010). 226 The measuring scale were selected in accordance with the quotations extracted from the case studies in a grounded process of theory development and were confronted with the theory previously developed by leading scholars of the subject. Regarding the scale type, the preference has been given to the Likert scale with seven points so as to include a neutral midpoint (De Vellis, 2003) and giving preference to the use of, whenever possible, the same pair of polarities that would facilitate completion of the questionnaire: Totally disagree/Totally agree. The next step was to determine the format for measurement. The vast majority of constructs under investigation were operationalized as multidimensional in nature, and therefore multiple indicators were used for their measurement. Question wording The next step in the preparation of the questionnaire is the question wording. The writing of each question is a critical task, because inadequate phrasing of a question can cause respondents to refuse to answer it or to answer incorrectly, either on purpose or because of misunderstanding. This becomes even more important with e-mail questionnaires, where the researcher is not present to clarify doubts. Thus, great care was taken in writing the questions as simply and directly as possible, trying to use only words and expressions that respondents were familiar with. Care was also taken to avoid questions that suggested the “correct” answer, and polarities were sometimes inverted, to avoid bias at the time of completion. Question sequence Having defined the types of measuring scale and the form of wording of each question, the preparation of the questionnaire was initiated. Kinnear and Taylor (1991) call attention that the sequence in which the questions are arranged is essential to the success of the research, as it may influence the nature of the respondent's answers and cause error in the survey findings. 227 As a general guideline, I decided to leave the construct agency conflict to be measured in the final part of the questionnaire. As its questions could be perceived as sensitive, they should be placed late in the sequence and after a relationship has been established and the respondent involved in the answering process. Also, those questions with similar content were grouped together to facilitate to use of memory in answering the questions, and to maintain the focus of the respondent (Bagozzi 1994; Hair Jr. et al. 2010). Questionnaire design The aesthetics of the questionnaire should also be the object of attention: an attractive questionnaire can increase the chances of respondent cooperation (Bagozzi, 1994; Hair Jr. et al. 2010). I decided to use a combination of printed questionnaires with an electronic version sent by email. To ensure the good quality of the questionnaire in both versions, the services of a professional graphic designer were used (not included in this material, as it was prepared in Portuguese). To increase credibility with the respondents, the front page included the logo of the University of Birmingham, as well as that of the Fundação Dom Cabral, as a sponsor of the research, as this institution is of high repute in the Brazilian business market. Finally, an expression of thanks and appreciation to respondents for participating in the study was also included at the end of the questionnaire. V.6.3. Pilot test According to Oppenheim (1972, p. 47) ‘questionnaires do not emerge fully fledged; they have to be created or adapted, fashioned and developed to maturity’. Thus, a pre-test of the survey instrument has to be made, as recommended by Cooper and Schindler (2011) and Hair Jr. et al. (2010). As these authors suggest, the understanding of all questions, instructions and scales were verified, as well as 228 whether the instrument as a whole was able to capture the information of the respondents effectively. Questionnaires were applied to 50 respondents, and among these, 49 were considered valid. From the questionnaires analysed, a large body of missing data was found (20.27% of the base) which prompted concern because these levels indicate great difficulty in treating subsequent multivariate data (Hair Jr. et al., 2010). I observed that some variables that refer to technical terms and other highly restrictive names caused major loss of data. To get around this, I sought to adapt questions so that the technical terms were replaced by statements that express situations / contexts / objectives that match the organizational reality. The concentration responses were evaluated by standard deviation and those with variability above the 95% percentile (2.21) or less than the 5% percentile (1.40) were analysed for possible ambiguities or difficulties in understanding. Similar analyses were carried out on the averages of the variables, and based on these considerations an attempt was made to reduce the size of the questions to make them less technical and more easily understandable, as suggested by Cooper and Schindler (2011) and Hair Jr. et al. (2005). Finally, the degree of consistency and redundancy of the responses were evaluated by calculating both the Cronbach's α coefficient and the correlation coefficient. Those items with correlation greater than 0.900 in absolute terms (Kline, 1998) were modified to reduce their similarity, and regarding the consistency, indicators with item-total correlation less than 0.400 were evaluated, seeking greater communality and convergence to the central concept of the constructs. Generally, no measure showed Coefficient α much lower than minimum threshold, i.e., 0.600 (note that this did not occur strictly in the final application of research). After these analyses, the necessary conditions to continue with the study were considered present. V.6.4. Final revision Based on the pilot test necessary adjustments were made to the questionnaire without however, changing its structure fundamentally. 229 The final sources of measurement of the constructs are illustrated in Table 5.2: Table 5.2: Source of Measurement of Constructs 230 231 232 233 The Appendix A presents the final model of the questionnaire that was applied. 234 V.6.5. Questionnaire application Once the questionnaire had been revised, the next step in the research process was to clearly define the population from which the information will be collected. Initially an FDC database of around 1,000 family-owned companies (usually founders, Chairmen or CEOs) was used. The questionnaires returned were around 6% of those sent out. The printed questionnaires were distributed during events at FDC involving family-owned firms, as well as events promoted by the FBN – Family Business Network Brazil. In these cases, participants were requested to fill them in at an appropriate time during the event itself, avoiding the possibility of not sending them later. 235 CHAPTER VI QUANTITATIVE RESEARCH METHODOLOGY VI.1. Data analysis This section deals with the analysis of data and concepts required to cleanse the data, evaluate the quality of measurements and test the effect of the hypothetical model of research. The descriptive analysis of the sample and of the scales of the study was the starting point and, subsequently, the evaluation of statistical assumptions and requirements for quality of the measurement was made. Finally, the test of the proposed theoretical model was conducted. Throughout the development of this chapter, the limitations of, and the care to be taken in the interpretation of the results, and their implications for the research model will be dealt with. The software packages used were SPSS 13, SmartPLS 2.0 and Microsoft Excel. SPSS is a tool that provides a wide variety of statistical methods for analysing data, while SmartPLS is a software application for graphical path modelling with latent variables, using the partial least squares (PLS) method. SPSS was used for the initial analysis and SmartPLS, when building, running, and validating the theoretical model. VI.1.1. Exploratory Analysis Prior to data analysis, the characteristics of the respondents and variables were examined, with emphasis on verification of assumption violations of the techniques used in the study (Tabachnick & Fidell, 2001). Therefore, the purpose of this subsection is to present information about the variables and respondents. Description of the sample In order to describe the profile of research participants and their respective companies, a descriptive analysis of them was carried out. The following charts characterize the 152 respondents who form the basis of the valid questionnaires used: 236 o Profile of the respondent’s companies Figure 6.1: Range of annual revenues of the company 1% 2% less than R$250 thousand 13% between R$250 thousand and R$2.5 million 34% between R$2.5 million and R$60 million 24% between R$60 million and R$170 million between R$170 million and R$500 million 26% above R$500 million Figure 6.2: Level of economic and financial performance of the respondent company 16% below average for the segment to which it belongs 29% in the midrange of the segment to which it belongs above average for the segment to which it belongs 55% Figure 6.3: Number of permanent employees maintained by the company 11% Less than 100 employees Between 100 and 500 employees 47% 42% 237 Above 500 employees Figure 6.4: Regarding the constitution of the capital, the company is: 5% Publicly traded corporation Privately held corporation 34% Limited 61% Figure 6.5: Lifetime of the companies surveyed 1% 4% 4% Less than 10 years 21% Between 10 and 25 years Between 25 and 50 years 29% Between 50 and 75 years Between 75 and 100 years More than 100 years 41% Figure 6.6: % of respondents from companies with formal board of directors 39% Without Board With Board 61% 238 Figure 6.7: % of respondents from companies with independent directors of those who adopt a board 42% With Without 58% Figure 6.8: % of respondents from companies with Chairman from the controllling family 7% Family Chairman Non-family Chairman 93% Figure 6.9: Family, nonfamily CEO? 9% Family executive Non-family executive 91% 239 Figure 6.10: To which generation does the family CEO belongs 2% 13% 1st. 2nd. 43% 3rd. 4th. 42% Figure 6.11: % of respondents from companies with Family Council 27% With Without 73% o Profile of the respondents Figure 6.12: % of respondents members of the owner family 17% Family member Non-family member 83% 240 Figure 6.13: % of respondents belonging to each generation 1 st. 14% 32% 2 nd. 3 rd. 4 th. 39% 3% Not applicable / not answered 12% Figure 6.14: % of respondents working in the company 11% Employees Non-employees 89% Figure 6.15: % of respondents who are shareholders of the company 30% Shareholder Non-shareholder 70% 241 Figure 6.16: % of respondents members of the board 29% Board member Non-board member 71% Missing data A total of 168 questionnaires were received. Eight of them were previously discarded as they were not totally completed. Even so, the other 160 had a substantial amount of data missing: 1,181 missing data were detected, which represents 10.4% of the total data base (1,181/[160*71]=1,181/11,360). This is above the acceptable boundary of 5% and moreover, was concentrated only in a few variables (Hair Jr. et al., 2010). Considering that the presence or absence of the board of directors influenced the completion of the questionnaire, each block presenting respectively a larger or smaller number of questions answered, due to their nature, an assessment of missing data was conducted separately for each situation. This strategy was adopted to preserve the specificities of the companies, avoiding the distortion of responses from each group (Tabachnick & Fidell, 2001). However, when doing the Missing Completely At Random test (MCAR; p> 0.05), I noted that the missing data were not meaningful. Nevertheless, it has been decided to exclude eight respondents with more than 17 missing answers (2 standard deviations above the average) and expunge those variables with more than 50% of missing data: variables XI.4, XI.5, XI.6, XI.7, XI.8, XI.9 (this occurs precisely due to the inexistence of several proposed actors in those firms without Board of Directors). It is important to note that there was a concentration of missing data in section XI of the questionnaire (41 % of instances of missing data). In all the other cases, the 242 variables presented missing data below 20% which were replaced using a multiple regression procedure. This was necessary because of their large dispersion, the absence of identifiable patterns and the need to keep a reasonable sample size to test the model. Descriptive analysis This subsection deals with the preliminary assessment of the variables and, as suggested by Malhotra (2001), it is a fundamental step to anticipate problems and identify solutions to the model test. Table 6.1 describes the results of basic statistics applied in the sample. Table 6.1: Data descriptive analysis Observed variables N MIN MAX MEAN S I.1 I.2 I.3 II.1 II.2 III.1 III.2 III.3 III.4 III.5 III.6 III.7 IV.1 IV.2 IV.3 IV.4 IV.5 IV.6 V.1 V.2 V.3 V.4 V.5 V.6 V.7 V.8 V.9 158 157 157 151 151 154 156 153 156 152 149 136 131 152 156 154 155 154 159 157 156 160 140 150 159 152 150 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 7 7 7 7 7 7 7 7 7 7 7 7 2 7 7 7 7 7 7 7 7 7 7 7 7 7 7 4.77 3.81 3.86 4.60 3.80 4.08 4.87 5.06 4.71 4.80 4.38 4.82 1.73 2.99 2.85 5.64 4.62 3.32 3.36 2.13 5.69 1.88 5.51 3.43 5.01 4.01 5.08 1.640 1.942 1.972 1.740 2.157 1.892 1.814 1.752 1.843 1.856 1.915 2.000 0.448 2.012 1.937 1.507 1.821 1.850 2.085 1.814 2.005 1.593 1.962 2.361 1.971 1.915 1.770 243 V.10 VI.1 VI.2 VI.3 VI.4 VI.5 VI.6 VI.7 VII.1 VII.2 VII.3 VII.4 VII.5 VIII.1 VIII.2 VIII.3 VIII.4 VIII.5 VIII.6 VIII.7 VIII.8 IX.1 IX.1_A IX.2 IX.2_A IX.3 IX.4 IX.5 IX.6 IX.7 X.1 X.2 X.3 X.4 XI.1 XI.2 XI.3 XI.4 XI.5 XI.6 XI.7 XI.8 XI.9 XI.10 Valid N (listwise) 144 156 144 160 155 152 156 153 144 155 158 157 152 147 143 157 158 157 160 158 154 159 48 63 131 158 156 157 156 155 157 142 151 133 135 153 139 59 77 115 75 88 55 71 7 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 2 14 2 6 7 7 7 7 7 7 7 7 7 7 7 7 7 6 7 7 7 7 7 3.37 4.47 3.98 5.30 5.54 4.49 4.87 5.19 3.61 5.18 4.75 4.94 4.13 3.85 4.11 4.93 5.08 3.52 4.27 2.80 3.90 1.61 6.00 1.41 2.72 3.22 5.14 5.30 4.25 3.14 4.66 4.81 2.33 5.13 2.86 3.51 3.25 2.83 2.66 2.24 2.39 2.94 2.40 2.61 1.789 1.889 1.890 1.545 1.921 1.820 2.119 1.772 2.168 1.925 1.875 1.780 2.045 1.987 2.166 2.013 2.000 2.117 2.037 1.894 1.989 0.489 2.379 0.496 1.233 2.046 2.008 1.781 1.943 1.952 1.960 1.856 1.624 1.852 1.741 1.882 1.938 1.886 1.643 1.537 1.684 2.036 1.717 1.801 PS: N is the number of valid answers; MIN is the smallest number found for this variable; MAX is the higher number found for this variable; MEAN is the arithmetical mean of the variable; S is the standard deviation of the variable. 244 Normality Test Determining the distribution of a response data set is an important and common first step in many statistical analysis undertakings. Distribution analysis is performed in order to understand the spread of the population investigated, but is also important since several other statistical methods are dependent on the type of distribution represented by the data. Normal distribution is the most common and well known and several other statistical methods rely on an assumption of normality. Therefore, the behaviour of variables is studied in order to verify the correspondence with the normal distribution which is the fundamental basis for inference in this field. A distribution would be normal if the data concentrate around the mean, median and mode, and with just a few data distant from this central tendency. Normal is used to describe a symmetrical, bell-shaped curve, which has the greatest frequency of scores in the middle, with smaller frequencies towards the extremes (Gravetter & Wallnau, 2004). It is desirable to achieve the presupposition of normality, because according to Tabachnick and Fidell (2001), it makes it more probable to obtain more reliable results, even using non-parametric techniques (those which do not presuppose normality). This follows from the fact that asymmetric distributions or excess kurtosis tend to underestimate the sample correlations. Such statement shows the importance of identifying the degree of compliance of univariate and multivariate normal distributions. By using graphical methods such as histograms with normal distribution curves and normal quantile-quantile plots (Q-Q plot), one can get a good indication if the distribution can be assumed to be normal. In the normal Q-Q plot diagram the observed values of a single variable are plotted against the values expected if the sample were from a normal distribution. Should the points cluster around a reasonably straight line, the sample can be assumed to be from a normal distribution. But, this is not the case of this sample. Graphical analyses were carried out (histograms and Q-Q diagrams) which revealed deviations from normality for all variables. Significant deviations from normal patterns by degrees of skewness and kurtosis of the data were also revealed, exceeding the suggested limits of ± 1 245 (Muthén & Kaplan, 1985) for several variables. Likewise, the formal tests of normal parameters (Z test) and normality (Komogorov Smirnov) revealed significant deviations for the two distinct subgroups analysed (with and without the board) and for the full sample. These conditions imply deviations from normality that limit the application of parametric analysis techniques, especially given the sample size of less than 200 cases (Tabachnick & Fidell, 2001). Linearity Analysis Linear correlation between variables is a prior condition for the use of factorial and regression analyses and it is obtained through the Pearson coefficient test. In order to verify the overall quality of the measurement, scatter plots for each variable were drawn and searched for possible discrepancies in linearity. An increase of 5% in the R2 in the cubic relation compared to the R2 of the linear relation suggests, given the size of the sample, a significant deviation from linearity. But no such case was found in the 50 dispersion diagrams analysed, revealing linearity between the variables tested. Multicolinearity Analysis This analysis aims to identify those variables with multicolinearity, which means with high redundancy in the proposed model. The redundancy level analysis of the variables was carried out by means of the tolerance and variance inflation factor (Kline, 1998). The resulting correlation matrix demonstrated the existence of nine correlations outside the boundaries of ± 0.7. However, this was considered a marginal problem located basically on some items of agency conflict. It is concluded that this poses no threat to subsequent proceeding analyses, since these correlations are found in the indexes of the same construct and with estimates very close to the suggested levels. 246 VI.2. Analysis of the quality of the measurement Before proceeding to hypothesis testing it is necessary to evaluate the quality of the scales of measurement. When testing a model without evaluating beforehand the quality of the measures employed in the study, one incurs the risk of assuming that a relationship which is not confirmed in the sample may occur due to problems in the measuring instrument. Moreover, it might be alleged erroneously that certain indicators measure a construct, when in fact there is little empirical evidence for this. Thus, a fundamental step is assessing the congruence between the expected results of measurement and their empirical operationalization, looking for evidence of reliability and validity of measurement (Netemeyer et al., 2003). The following analysis is to assess the psychometric properties of the scales, covering the assessment of dimensionality, reliability and validity of measurements. VI.2.1. Analysis of the dimensionality At this stage, the data dimensionality was verified; checking if the number of suggested dimensions for each variable was effectively observed (Netemeyer et al., 2003). For this purpose, the exploratory factor analysis (EFA) was employed. As suggested by Hair Jr. et al. (2010) and Tabachnick and Fidell (2001), the EFA allows the factors of the studied variables to be identified, and the indicators’ polarity to be analysed, whether the reverse items possess negative loads with their respective constructs. For each constituent construct of the model, the principal components extraction procedure was applied with direct oblimin rotation (oblique rotation), to see if more than one dimension was revealed for each construct (Malhotra, 2001). When the intercorrelations between dimensions are small, it is preferred to use the varimax orthogonal rotation procedure since it meets completely the requirements of EFA and the results are almost the same. In fact, during data analysis it is noticeable that almost all dimensions founded in multidimensional scales are nearly orthogonal, and therefore varimax rotation is preferred. 247 Considering that 152 valid questionnaires were left following the checks just mentioned and that the construct with the higher number of related items, in the worst case, was agency conflict construct with 10 questions, this results in around 15.2 answers for each construct. Thus, as the most conservative boundaries suggest at least 10 answers per item to allow the use of Factor Analysis (Hair Jr. et al., 2010), it is possible to assert that the sample used in this research was of adequate size. In order to decide how many factors should be extracted from each construct, the following procedures have been used: i) analysis of the scree plot, considering as factors those with eigenvalues over 1; ii) analysis of the percentage of extracted variance (if higher than 50 or 60%); iii) checking of consistency and adhesion to the underlying theory. Also, I have identified those indicators that diverge from the others because they had been loaded in extraneous factors, as well as those with low association with the original factors (communalities below 0.400). Defining which loadings should be considered, those items with factor loadings higher than 0.5 were selected (Hair Jr. et al, 2010). The KMO measure of sampling adequacy and the percentage of variance explained of the constructs were used as a measure of overall quality of factor solutions. Whenever items load in a factor that presents more than one indicator, I decided to retain the item even if it does not show good measurement properties (communalities lower than 0.400 and loadings lower than 0.5) since its convergent validity is a preferable approach for identifying no convergence of items. At the EFA stage the analysis focuses only on identifying underlying dimensions with no warnings relative to the reliability and validity of the scales or items, since these concern subsequent analysis stages. Table 6.2 presents the results of applying these tests, including the factors extracted from each of the constructs originally proposed, and the loadings, communalities, variances extracted and KMO of each construct. Note that most constructs present just one dimension. In such cases when an indicator was reverted into two dimensions (marked with *), never more than two indicators were loaded into a second factor. 248 Table 6.2: Factor Analysis ORIGINAL CONSTRUCTS FACTORS EXTRACTED Strategic alignment Alignment Social alignment Family power within the Board Board Professionalism Strategic role of the board Family Council importance Family and Business Interaction (a) Management Control Trust (b) Relationship between family members and among them and the firm Management Control Trust between owners and managers Embeddedness ITEMS LOADING COMMUNALITIES III.2 III.3 III.4 III.5 III.6 III.1(i)* III.7* IX.3(i)* IX.6(i)* IX.4 IX.5 IX.7(i) IV.1** IV.2(i)* IV.3 (i) IV.4 IV.5 IV.6(i) VII.1 VII.2 VII.3 VII.4 VII.5(i) V.6(i) V.7 V.10(i) V.8* V.9* 0,737 0,774 0,765 0,808 0,709 -0,792 0,743 0,739 0,784 0,747 0,769 - 0,607 N.A 0,948 -0,580 0,832 0,795 -0,682 0,752 0,803 0,857 0,718 -0,421 0,635 -0,760 0,666 0,800 0,748 0,654 0,638 0,622 0,657 0,547 0,633 0,599 0,549 0,604 0,583 0,593 0,624 N.A 0,900 0,593 0,698 0,639 0,465 0,565 0,646 0,735 0,515 0,177 0,406 0,594 0,445 0,671 0,619 248 VARIANCE EXTRACTED KMO 47,529 0,809 14,624 25,869 0,537 33,195 N.A 20,475 45,439 55,740 N.A 0,720 0,766 31,097 0,562 23,590 Transparency and Disclosure Transparency Successor legitimacy Legitimacy Legitimacy of family managers Competence Managers’ competence Shirking and misappropriation Opportunism Abuse of privileges Imparciality Imparciality Altruistic selection Altruism Moral hazard Comprehension/acceptance Comprehension of social roles of social roles VI.4 VI.5 VI.6 A(i) VI.7 I.1(i) I.2(i) I.3(i) II.1* II.2* VI.1 VI.2 (i) VI.3 V.1 V.2 V.4 V.3(i)* V.5(i)* VIII.1 VIII.2 VIII.3 VIII.5* VIII.6* VIII.4(i) VIII.7 VIII.8 X.1 X.2 X.3(i) X.4 0,838 0,843 -0,521 0,724 0,744 0,772 0,739 0,831 0,842 0,851 -0,590 0,646 0,755 0,779 0,569 0,909 0,758 0,894 0,913 0,810 0,791 0,921 -0,781 0,721 0,752 0,808 0,813 -0,209 0,738 0,702 0,710 0,271 0,524 0,556 0,598 0,555 0,701 0,710 0,724 0,348 0,417 0,581 0,581 0,345 0,780 0,687 0,800 0,833 0,657 0,764 0,856 0,627 0,611 0,590 0,653 0,661 0,044 0,544 55,170 0,711 35,029 0,584 27,393 49,631 0,474 38,628 0,638 20,842 0,763 0,688 25,320 0,616 43,610 47,550 0,641 * shows that this indicator was loaded into a second factor. ** shows that this indicator was excluded in a previous step. The values in bold represent those below the recommended levels of 0.5 for the factorial loadings and 0.4 for the communalities. Despite low convergence these items were retained in analysis since factor analysis was used to understand the dimensionality of the constructs and not the validity of items. Items IX.2 and IV.1(i) were deleted in previous analysis steps. 249 (a) It is decided not to proceed with the first dimension of Family and Business Interaction since it was formed by just one item, preventing any attempt to verify the reliability and validity of the measure. Thus, this name was preserved for the second factor. (b) In the case of trust, I chosen to exclude the V.8 question for two reasons. First, if this indicator was retained, the second dimension of trust will have only one item, which does not allow analysis of the reliability and validity of the constructs ‘measurements. Second, looking at the non-rotated component matrix, it is noticeable that this item has the greatest loading on the second factor, meaning that this item is the one that forces the appearance of the second factor. In that case with the exclusion of V.8 the remaining items show a reasonable fit within a one factor solution. When V.9 was excluded, a marginal increment in the reliability indexes were found (=0.48; CR= 0.73). These indexes were lower when V.9 was included =0.45; CR= 0.70), but not so much to ensure the retaining of face validity of this item in order to get a more theoretically sound measure. On the other hand, the exclusion of V.9 increases the Average Variance Extracted to the suggested limits (AVE=0.49 against AVE=0.38 with V9). So the decision was to exclude V9. Agency conflict construct The analysis of the agency conflict construct required an EFA considering just the indicators XI.1, XI.2 and XI.3, as the complete sample using the other related indicators had high percentage of missing data (such characters simply do not exist in most firms). The results present good quality levels (KMO=0.650 and 68.29% of variance extracted) in a one-dimensional factorial solution. Table 6.3: Descriptive Statistics of the Agency Conflict Construct XI.1 Mean 2,84 Std. Deviation 1,676 Analysis N 152 Missing N 0 XI.2 3,51 1,884 152 0 XI.3 3,23 1,903 152 0 Source: Extracted data Table 6.4: Correlation Matrix(a) of the Agency Conflict Construct Correlation XI.1 XI.1 1,000 XI.2 ,554 XI.3 ,404 XI.2 ,554 1,000 ,623 XI.3 ,404 ,623 1,000 Source: Extracted data a Determinant = ,421 250 Table 6.5: KMO and Bartlett's Test of the Agency Conflict Construct Kaiser-Meyer-Olkin Measure of Sampling Adequacy. ,650 Bartlett's Test of Sphericity Approx. Chi-Square 129,149 DF 3 Sig. ,000 Source: Extracted data Table 6.6: Communalities of the Agency Conflict Construct XI.1 Initial 1,000 Extraction ,602 XI.2 1,000 ,787 XI.3 1,000 ,669 Extraction Method: Principal Component Analysis Table 6.7: Total Variance Explained of the Agency Conflict Construct Component Initial Eigenvalues Extraction Sums of Squared Loadings 1 Total 2,059 % of Variance 68,629 Cumulative % 68,629 2 ,601 20,042 88,672 3 ,340 11,328 100,000 Total 2,059 % of Variance 68,629 Extraction Method: Principal Component Analysis. Figure 6.17: Scree Plot of the Agency Conflict Construct Extraction Method: Principal Component Analysis 251 Cumulative % 68,629 Table 6.8: Component Matrix (a) of the Agency Conflict Construct Component 1 XI.1 ,776 XI.2 ,887 XI.3 ,818 Extraction Method: Principal Component Analysis. A 1 component extracted. VI.2.2. Reliability Analysis and Overall Measurement Quality According to Raykov (1997a) most psychological measures (e. g., inventories, questionnaires, scales, and self-reports) are assembled from a set of items. The measures produce scores of a construct that are sums of the scores on their components. Thus, it is necessary to verify the overall quality of the measurement of each construct. This is achieved through the reliability test, which defines how each block of indicators relates to its latent construct (Chin, 1998). In research, the term reliability means ‘repeatability’ or ‘consistency’, leading one the idea that a measure is considered reliable if it would give the same result over and over again (Trochim, 2006). The use of data with low reliability may result in: (a) a high probability of adopting a non-valid measure – the result can tell us nothing about the construct which it was intended to measure. And, even if the measure was valid, (b) the error would be high, since the observed variability affects the outcomes of any statistical test, increasing the probability of non-significant results (Maroco & Garcia-Marques, 2006). In technical terms, a measure is valid when it leads to a proper and correct conclusion about the population, using the sample as a valid basis for analysis. Any measure to be valid as part as a construct, has to be necessarily reliable. This means that reliability is a necessary but not sufficient condition to achieve validity. Therefore, measure reliability is the first step to discover its validity. But after guaranteeing its reliability, it is also necessary to prove its validity (Maroco & Garcia-Marques, 2006). In order to estimate the reliability of each scale constituents of the questionnaire administered to the 152 respondents, an internal consistency reliability 252 test should be applied. The most commonly used one, in the majority areas of social sciences as well as in many other disciplines, is Cronbach’s Alpha Coefficient, a reliability estimator for composite measures containing multiple components. It is considered by most researchers, with the exception of those who dedicated themselves to psychometric studies, to be the universal recommended reliability coefficient for a metric scale analysis, whatever its characteristics (Maroco & GarciaMarques, 2006). This ubiquitous use occurs due to its flexibility (Osburn, 2000) and due to the fact that it does not require more than one test administration (as does the test-retest method), or more than one parallel test form (as does the parallel-forms method), or the splitting of a test into two parallel halves (as do the split-half methods) (Huysamen, 2006). On the other hand, several investigators have noted that an internal consistency reliability coefficient, like Coefficient α, based on a single administration of a test, may overestimate reliability because such coefficients assign transient error due to differences in test administration, temporary changes in the examinee, and so on to true-score variance (Guion, 1965; Schmidt & Hunter, 1996). In addition, if the errors made in responding to test items are positively correlated, Coefficient α may be inflated (Komaroff, 1997). Although it is true that Coefficient α may sometimes be inflated because of the conditions mentioned above, a potentially more serious problem is its tendency to underestimate the true reliability when items do not measure the construct with equal strengths. This means that unless the measures have the same factor loadings and measurement errors, a critical assumption when Alpha’s formula is derived, Coefficient α renders a lower bound estimate (Maroco & Garcia-Marques, 2006; Osburn, 2000). The assumption underlying the fact that Coefficient α underestimates the true reliability is identified as tau equivalence. And tau equivalence is an important consideration when examining reliability as most generalization studies ‘assume that all items measure the same latent trait on the same scale, with the only variance unique to an item being comprised wholly of error’ (Graham, 2006, p.935). The tau-equivalence follows the same unidimensionality principle of the parallel model, requiring that all test items measuring a single latent variable have to be exactly equivalent to one another. All items must measure the same latent variable, on the same scale, with the same degree of precision. The only difference between both models is that while in the latter all error scores are equal across items, 253 in the first there is a possibility of different amounts of error (Raykov, 1997a). Another possible model of reliability estimation is the essential tau-equivalent model which is, as its name implies, essentially the same as the tau-equivalent model. Essential tauequivalence assumes that each item measures the same latent variable, on the same scale (similar variances), but with possibly different degrees of precision (different means) (Raykov, 1997b). The difference between item precision and scale is an important distinction to make. Whereas tau-equivalence assumes that item true scores are equal across items, the essentially tau-equivalent model allows each item true score to differ by an additive constant to each pair of variables (Raykov, 1997b). The inclusion of an additive constant affects only an item’s mean, not its variance or covariance with other items. As reliability is a variance-accounted-for statistic, it is unaffected by differing means. Therefore, for the purposes of estimating reliability, tau equivalence and essential tau equivalence would be considered identical, implying that the correlations between true scores are all unity and that the variances and covariances of the true scores on the components of the measure are all equal. Therefore, I refer, from now on, to essential tau equivalence simply as tau equivalence (Osburn, 2000). Finally, a least restrictive, most general model of use for reliability estimation is the congeneric model, which assumes that each individual item measures the same latent variable, with possibly different scales, with possibly different degrees of precision, and with possibly different amounts of error (Raykov, 1997a). That means that if the components of a measure are congeneric equivalent, the correlations between the true scores are unity but variances of the true scores may vary (Jöreskog & Sörbom. 1990). Congeneric equivalence implies that the measure is unidimensional, but the factor loadings and error variances may vary. The conclusion is that tau-equivalence hedges the use of measures which represents different aspects of the latent constructs. Thus, if prior research indicates a priori multidimensionality of the data, tau-equivalent methods of estimating reliability, like Cronbach’s α, should be avoided. The simulation performed by Osburn (2000), for instance, computing different data sets, from congeneric equivalent or multidimensional to essentially tau-equivalent measures, shows that as heterogeneity increased the Coefficient α underestimation became considerably worse. His conclusion is that the use of alpha coefficient is inappropriate in situations in which 254 the assumption of essential tau-equivalence is clearly violated, indicating that the test items measure the same latent variable in different scales. Just the presence of even a single item that is not tau-equivalent to the other items can have a dramatic impact on the accuracy of Coefficient α (Graham, 2006). It is important to notice that tests with a greater number of items are less vulnerable to underestimation when tau-equivalence is violated than tests with only a small number of items (Raykov, 1997b), which is the case in most constructs of this study. This is due to the fact that, when a single item violates tau-equivalence, the proportion of true score variance that is congeneric to the other item true scores is smaller when one has a greater number of items than when one has fewer. Although Coefficient α is widely used in practical work, a significant body of literature has developed in which several less familiar internal consistency reliability coefficients have been proposed. All of these coefficients have in common an attempt to reduce the underestimation of reliability when the components of a measure are not equivalent. Thus, in this study, the decision was to adopt more than one reliability test, as suggested by Maroco and Garcia-Marques (2006): besides the Cronbach's α evaluation, I also run the Composite Reliability (CR) and the Average Variance Extracted (AVE) tests. Among all of them, the preference is for the use of the Composite Reliability, considered a better coherent approach to reliability of the items used in this study. The composite reliability coefficient (the sum of the loadings squared, divided by the sum of the loadings squared plus error) is superior to Cronbach’s α because it does not assume that all indicators are equally weighted (Martensen et al., 2007). A third measure that has been proposed is the Average Variance Extracted (AVE), created by Fornell and Larcker (1981). It is a statistic involving the percentage error variance in a measure, representing on average how much the constructs explain the variance of the indicators and then, giving an idea of the general degree of association between the constructs and the indicators. In different terms, AVE is a measure of the error-free variance of a set of items. The higher AVE value, the more representative are the indicators of the construct on which they load. The next table presents these indexes: 255 Table 6.9: Quality of the measurement of the scales Scales AVE Abuse of privilegie Agency Conflict Altruistic selection Comprehension/acceptance of corporate roles Familiy and business interaction Family power on the Board Imparcialitty Legitimacy of family managers Management Control Managers’ competence Moral hazard Shirking and misappropriation Social alignment Strategic alignment Strategic role of the board Successor’s legitimacy Transparency Trust between owners and managers 0.70 0.69 0.71 0.63 0.55 0.63 0.76 0.70 0.53 0.50 0.59 0.51 0.63 0.61 0.48 0.55 0.55 0.48 Composite Reliability 0.82 0.87 0.83 0.83 0.83 0.77 0.91 0.83 0.84 0.74 0.81 0.76 0.77 0.89 0.72 0.78 0.83 0.73 R Square 0.21 0.38 0.00 0.00 0.00 0.00 0.53 0.31 0.00 0.00 0.42 0.36 0.11 0.59 0.05 0.04 0.41 0.39 Cronbach’s Alpha 0.60 0.77 0.71 0.70 0.73 0.43 0.84 0.58 0.76 0.48 0.65 0.52 0.41 0.84 0.54 0.62 0.72 0.48 PS: the numbers in bold represent values below the limits considered for each indicator. According to Cronbach’s α, numbers above 80% would be considered an ideal reliability criterion (Netemeyer et al., 2003). However, in cases where the scale is especially developed for a study, numbers around 0.70 are acceptable and, even a minimum of 0.60 (Malhotra, 2001). The acceptable cut-off for composite reliability would be the same as the researcher sets for Cronbach’s α since both attempt to measure true reliability. In an adequate model, composite reliabilities should be greater than 0.70 (Nunnally, 1978). Notice that some Cronbach's α coefficients are below the expected limit, but in all cases the Composite Reliability is above 0.70. This reveals the violation of the tau equivalence assumption, indicating that Composite Reliability should be applied as the adequate reliability estimator. To use a latent variable, its AVE should be greater than 0.50 (Chin, 1998; Fornell & Larcker, 1981). Following this criterion, AVE values reported in Table 6.9 demonstrate internal consistency for most of the indicators of the latent variables, except managers’ competence, strategic role of the board and trust between owners and managers. Considering the importance of these constructs for the proposed model, as well as the fact that the numbers are very close to the borderline, and even 256 that other authors, like Bollen (1989), accept the lower limit of 0.40, all constructs were considered reliable. VI.2.3. Convergent and discriminant validity After finding an adequate level of quality of the overall measurements, it is time to realize the evaluation of validity of the scales (Churchill & Iacobucci, 2002). This analysis consists in testing the congruence between the research tool and the theoretical constructs (Netemeyer et al., 2003). For this purpose, it is necessary to apply structural equation modelling techniques which are able to cope with measurement errors in a multiple hierarchical relationship between constructs using a unique tool (Fornell & Lacker, 1981; Bagozzi et al., 1991; Mackenzie, 2001). Structural equation modelling is based on the analysis of covariance using procedures of maximum-likelihood (Haenlein & Kaplan, 2004) which has been popularized by Jöreskog and Sörbom (1989) and implemented in programs like LISREL and AMOS. However, due to violation of the normality assumption and the size of the sample used in this study, the PLS method is the better alternative to be used. According to Chin (1998) the PLS approach is appropriate for samples which are at least 10 times bigger than the construct with the greatest number of indicators or with the large amount of dependency relations. In this study, the construct with the larger number of items has five indicators, revealing that the sample surpass the boundaries of the required sample size. The next step consists in doing the convergent validity and discriminant validity tests which are commonly regarded as subsets of construct validity. Convergent validity tests if there are enough strong correlations between the indicators, in order to certify that such measures are reflexes of the same latent construct. Thus, the factorial loading obtained through PLS validate the construct through a one-tailed test, where the critical t is t= 1.97. 257 Table 6.10: Convergent validity tests Indicators and Constructs Relationship I.1 <- Successor’s legitimacy I.2 <- Successor’s legitimacy I.3 <- Successor’s legitimacy II.1 <- Legitimacy of family managers II.2 <- Legitimacy of family managers III.1 <- Social alignment III.2 <- Strategic alignment III.3 <- Strategic alignment III.4 <- Strategic alignment III.5 <- Strategic alignment III.6 <- Strategic alignment III.7 <- Social alignment IV.3 <- Familiy and business interaction IV.4 <- Familiy and business interaction IV.5 <- Familiy and business interaction IV.6 <- Familiy and business interaction IX.3 <- Family power on the Board IX.4 <- Strategic role of the board IX.5 <- Strategic role of the board IX.6 <- Family power on the Board IX.7 <- Strategic role of the board V.1 <- Shirking and misappropriation V.10 <- Trust between owners and managers V.2 <- Shirking and misappropriation V.3 <- Abuse of privilege V.4 <- Shirking and misappropriation V.5 <- Abuse of privilege V.6 <- Trust between owners and managers V.7 <- Trust between owners and managers VI.1 <- Managers’ competence VI.2 <- Managers’ competence VI.3 <- Managers’ competence VI.4 <- Transparency VI.5 <- Transparency VI.6 <- Transparency VI.7 <- Transparency VII.1 <- Management Control VII.2 <- Management Control VII.3 <- Management Control VII.4 <- Management Control VII.5 <- Management Control Original Sample (O) 0.579 0.775 0.844 0.854 0.823 0.759 0.811 0.797 0.791 0.762 0.731 0.825 0.717 0.849 0.772 0.618 0.871 0.444 0.630 0.711 0.922 0.790 0.754 0.696 0.734 0.648 0.929 0.489 0.802 0.863 0.503 0.701 0.795 0.845 0.577 0.719 0.748 0.786 0.864 0.706 0.459 258 Sample Mean (M) 0.544 0.733 0.817 0.851 0.819 0.755 0.811 0.796 0.792 0.763 0.730 0.824 0.718 0.850 0.773 0.616 0.808 0.447 0.625 0.647 0.906 0.786 0.750 0.686 0.726 0.652 0.927 0.499 0.798 0.854 0.493 0.702 0.792 0.843 0.580 0.721 0.748 0.783 0.865 0.701 0.463 Standard Deviation (STDEV) 0.195 0.156 0.100 0.027 0.038 0.071 0.027 0.036 0.029 0.039 0.041 0.055 0.053 0.023 0.043 0.069 0.286 0.158 0.093 0.267 0.050 0.039 0.053 0.072 0.078 0.065 0.026 0.094 0.041 0.032 0.106 0.068 0.038 0.022 0.056 0.050 0.033 0.039 0.018 0.051 0.078 T Statistics (O/STERR) 2.969 4.967 8.458 31.870 21.814 10.688 30.399 22.403 27.502 19.477 17.859 15.065 13.535 36.927 17.890 8.936 3.043 2.816 6.762 2.661 18.334 20.144 14.113 9.695 9.381 9.947 36.247 5.220 19.366 26.851 4.746 10.358 21.172 37.683 10.297 14.283 22.742 20.307 48.578 13.939 5.903 VIII.1 <- Imparcialitty VIII.2 <- Imparcialitty VIII.3 <- Imparcialitty VIII.4 <- Moral hazard VIII.5 <- Altruistic selection VIII.6 <- Altruistic selection VIII.7 <- Moral hazard VIII.8 <- Moral hazard X.1 <- Comprehension/acceptance of corporate roles X.2 <- Comprehension/acceptance of corporate roles X.4 <- Comprehension/acceptance of corporate roles XI.1 <- Agency Conflict XI.2 <- Agency Conflict XI.3 <- Agency Conflict 0.877 0.900 0.840 0.730 0.990 0.665 0.791 0.775 0.836 0.876 0.900 0.841 0.724 0.985 0.645 0.789 0.770 0.835 0.022 0.016 0.021 0.046 0.014 0.138 0.040 0.051 0.023 40.352 55.225 39.777 15.868 70.699 4.840 19.662 15.240 36.578 0.836 0.835 0.025 34.111 0.691 0.683 0.060 11.480 0.758 0.906 0.813 0.753 0.905 0.813 0.047 0.015 0.033 16.051 59.486 24.603 PS: This table presents: (O) the standard factorial loads of the sample, which means the correlation between the indicator and the latent construct: (M) the averages of the sub-samples obtained via PLS; (STERR) estimated standard error; (O/STERR) the T value which is the proportion between the loading and its standard error. Numbers above 1.97 represent an estimate of the 5% one-tailed test. We can see from Table 6.10 that the sample values are all greater than 0.4 demonstrating, according to the criteria of Hulland (1999), good validity of the items. In addition, all the T values were significant in general. The only item excluded due to lack of convergent validity was X.3 (i), since its factor loading was less than 0.400 and it was therefore, not significant. Then, the discriminant validity test was conducted, checking if the scales conceived to measure such construct are more related to it or the others constructs (Nunnaly & Bernstein, 1994). For this purpose, a method suggested by Fornell and Larcker (1981) was used, comparing the average variance extracted from the constructs with the shared variance between theoretical constructs (R2 was obtained through the correlation of the estimated scores in the PLS). Thus, when two scales conceived to measure distinct constructs share more variance between themselves than the variance shared among their indicators, absence of discriminant validity is verified. Comparing the constructs in pairs using the Fornell and Larcker (1981) criterion, all of them have discriminant validity, which allows us to assert that the scales of each construct measure distinct latent dimensions. These results are shown in the next table: 259 Table 6.11: Discriminant validity test Construct Abuse of privilege Altruistic selection Agency Conflict Compr. /acceptance of corporate roles Familiy and business interaction Family power on the Board Imparcialitty Legitimacy of family managers Management Control Managers’ competence Moral hazard Shirking and misappropriation Social alignment Strategic alignment Strategic role of the board Successor’s legitimacy Transparency Trust between owners and managers AVE CR N 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 0,70 0,01 0,12 0,21 0,17 0,00 0,08 0,07 0,10 0,08 0,12 0,10 0,08 0,11 0,07 0,09 0,09 0,12 0,70 0,82 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 0,71 0,07 0,04 0,00 0,03 0,05 0,06 0,04 0,00 0,15 0,04 0,01 0,02 0,04 0,01 0,00 0,07 0,71 0,83 0,69 0,27 0,27 0,02 0,28 0,25 0,21 0,24 0,23 0,21 0,18 0,25 0,11 0,06 0,24 0,18 0,69 0,87 0,63 0,31 0,00 0,38 0,29 0,32 0,23 0,35 0,29 0,16 0,39 0,26 0,07 0,31 0,19 0,63 0,83 0,55 0,02 0,22 0,26 0,17 0,29 0,24 0,18 0,11 0,32 0,19 0,04 0,25 0,23 0,55 0,83 0,63 0,00 0,00 0,00 0,00 0,01 0,01 0,01 0,00 0,05 0,02 0,00 0,01 0,63 0,77 0,76 0,30 0,39 0,34 0,46 0,29 0,16 0,40 0,23 0,08 0,27 0,21 0,76 0,91 0,70 0,18 0,31 0,31 0,23 0,09 0,29 0,08 0,01 0,20 0,26 0,70 0,83 0,53 0,33 0,34 0,32 0,15 0,49 0,15 0,08 0,41 0,19 0,53 0,84 0,50 0,24 0,22 0,11 0,41 0,12 0,04 0,26 0,22 0,50 0,74 0,59 0,28 0,13 0,26 0,30 0,07 0,25 0,30 0,59 0,81 0,51 0,14 0,33 0,13 0,09 0,17 0,24 0,51 0,76 0,63 0,16 0,09 0,12 0,18 0,09 0,63 0,77 0,61 0,18 0,03 0,38 0,23 0,61 0,89 0,48 0,08 0,20 0,11 0,48 0,72 0,55 0,08 0,04 0,55 0,78 0,55 0,21 0,55 0,83 0,48 0,48 0,73 (i) The values on the main diagonal indicate the average variance extracted of the constructs. The values below the diagonal show the percentage of variance shared between constructs. When the value of shared variance between constructs is greater than the values of the AVE of the row or column of reference, there is a violation of discriminant validity. (ii) Comparing the squared correlation with the Average Variance Extracted of the constructs, it is noted that the constructs had discriminant validity, since in all cases the shared correlation between the constructs is less than the variance that the constructs share with their own indicators. 260 VI.3. Nomological validity – Test of the model The last component of construct validity is nomological validity. In this procedure, the objective is to assess if the relationships between the latent variables of the model are supported by empirical data (Netemeyer et al., 2003), testing if the hypotheses are supported based on the observation carried out in the study. This is the most relevant criterion for assessing the validity, because it has been the sine qua non condition of the quality of the underlying theory (Hunt, 2002). From the results obtained in the factor analysis, when starting from the initial constructs new factors were extracted, I opted for the reconstruction of the theoretical model originally proposed, without deviating from the observations from the case studies and the necessary theoretical underpinning. Some hypotheses were confirmed and others were proposed, as is shown below: Figure 6.18: Theoretical model modified after EFA Social Alignment H3a + Family and Business Interaction H3b + Strategic Alignment H1a H2a + + - H4 - H1b H6 - H8a Trust + H8b H10a H11 H10b - + + Transparency Successor’s legitimacy - + + H9b Abuse of privileges H14a H13 Impartiality H9a Legitimacy of family managers Shirking and misappropriation H12 Family power on the Board - H5 - - H2b Management Control + H7 Agency Conflict Acceptance of the Board role + Moral hazard H17 Manager´s competence - H14b - Comprehension of social roles - H15 + Altruistic selection H16 + To test the new theoretical model, PLS was used, showing the results that can be seen in Figure 6.19: 261 Figure 6.19: Research Model tested via PLS Family and business interaction 0.114 0.338 Family power on the board Social alignment 0.299 - 0.090 - 0.194 0.380 Agency conflict - 0.175 - 0.166 0.389 - 0.265 Trust between owners and managers 0.641 0.410 0.190 Transparency 0.282 0.306 Management control 0.553 Legitimacy of family managers - 0.017 - 0.130 0.206 0.531 0.219 Strategic role of the board 0.542 0.587 Strategic alignment 0.048 Abuse of privilege 0.340 Competence - 0.194 - 0.227 0.038 Comp. of corporate roles Successor’s legitimacy - 0.352 Impartiality 0.425 0.319 Moral hazard 0.358 Shirking and misappropriation Altruistic selection (i) The values expressed within each arrow indicate the estimated standard weights. The values placed inside each oval represent the R2, i.e., the percentage of variance explained by the structural model. (ii) It should also be pointed out that the variables presented in the model with zero variance are exogenous, in other words are those which are not explained by any other variable in the model. For this reason they necessarily have a variance explained equal to zero. 262 (iii) In the model shown in Figure 6.19, the hypothetical relationships and standardized weights obtained are presented. To simplify the demonstration of relationships between constructs, I chose to keep only the design of the latent variables, omitting the observed variables. Next, in the Table 6.12, the overall results of the explanatory model of the dependent constructs are presented: Table 6.12: Results of the hypotheses of the model proposed by the dissertation HYPHOTESES INDEPENDENT DEPENDENT SAMPLE POP ERROR T VALUE SIG. H1a Social alignment Agency Conflict -0,194 -0,198 0,071 2,75 p < 0,01 H1b Strategic alignment Acceptance of the Board role Family power on the Board Familiy and business interaction Familiy and business interaction Management Control Agency Conflict -0,175 -0,165 0,084 2,075 p < 0,05 Strategic alignment -0,09 -0,093 0,049 1,822 0,070 Acceptance of the Board role 0,22 0,231 0,095 2,301 p < 0,05 Social alignment 0,338 0,353 0,076 4,443 p < 0,001 Strategic alignment 0,299 0,297 0,062 4,859 p < 0,001 Strategic alignment 0,542 0,546 0,055 9,943 p < 0,001 0,641 0,642 0,043 15,053 p < 0,001 -0,166 -0,171 0,07 2,369 p < 0,05 0,19 0,19 0,067 2,847 p < 0,01 -0,017 -0,021 0,07 0,237 0,813 0,282 0,285 0,071 3,983 p < 0,001 -0,194 -0,224 0,072 2,69 p < 0,01 0,553 0,559 0,046 11,96 p < 0,001 -0,13 -0,139 0,06 2,177 p < 0,05 -0,227 -0,225 0,076 2,969 p < 0,01 H2a H2b H3a H3b H4 H5 H11 Management Control Transparency Trust between owners and Agency Conflict managers Trust bet. owners Transparency and managers Successor’s Trust bet. owners legitimacy and managers Legitimacy of family Trust bet. owners managers and managers Managers’ Successor’s competence legitimacy Managers’ Legitimacy of family competence managers Trust bet. owners Abuse of privilege and managers Shirking and Trust bet. owners misappropriation and managers Impartiality Agency Conflict H12 Moral hazard H13 Moral hazard H6 H7 H8a H8b H9a H9b H10a H10b Comprehension / accept. of corporate roles Comprehension / H14b accept. of corporate roles Comprehension / H15 accept. of corporate roles Comprehension / H16 accept. of corporate roles H17 Altruistic selection Source: Data from the survey. H14a -0,265 -0,271 0,073 3,625 p < 0,001 Impartiality Shirking and misappropriation -0,475 -0,466 0,069 6,861 p < 0,001 0,319 0,322 0,081 3,929 p < 0,001 Abuse of privilege -0,453 -0,466 0,061 7,433 p < 0,001 Shirking and misappropriation -0,352 -0,348 0,081 4,354 p < 0,001 Moral hazard -0,535 -0,535 0,053 10,108 p < 0,001 Impartiality 0,34 0,35 0,075 4,546 p < 0,001 Moral hazard 0,284 0,292 0,055 5,191 p < 0,001 263 PS: a) SAMPLE is the standard weight obtained for the full sample; b) POP is the average weight obtained in the bootstrapping; c) ST. DEV. is the standard deviation of the estimate d) The error is the estimated error of the estimate; e) The T VALUE is the ratio of the non-standardized weight by its standard error. f) SIG is the significance level corresponding to t test probability with 148 degress of freedom and two sided tests. For one sided tests the significance levels could be considered half the presented value. In Table 6.12, the model's hypotheses tested are presented, together with their respective weights obtained in the structural model (PLS). The absolute sizes of the standardized values of the sample are sufficient to demonstrate the relationship between the constructs (whether positive or negative) by interpreting the relationship of each dependent construct with its independent constructs. VI.4. Summary of the quantitative results This chapter has presented the empirical study carried out into the hypothesized relationships among the constructs. Table 6.13 summarizes the hypotheses and results. Overall, the hypotheses were well supported in the empirical tests. Table 6.13: Results of Hypotheses Testing HYPHO DESCRIPTION OF HYPOTHESIS RESULT -TESES H1a The greater the social alignment among the family owners, the lower the Supported agency conflict. H1b The greater the strategic alignment between family owners and the Supported managers, the lower the agency conflict. H2a The more strategic is the role of the Board, the greater will be the Not strategic alignment within the company. supported H2b The greater the power of the family within the Board of Directors, the Supported greater will be its strategic role. H3a The better the interaction between family members, and between them and the company, the greater the degree of social alignment. Supported H3b The better the interaction between family members, and between them and the company, the greater the degree of strategic alignment. Supported H4 The more developed the system of management control in the family business, the higher the prevailing strategic alignment. Supported H5 The more developed the system of management control in a family business, the greater is the level of transparency. Supported H6 The higher the existing trust in the family and business spheres, the lower the agency conflict. Supported 264 H7 The higher the level of transparency in management transactions, the higher the level of trust prevailing in the family business. Supported H8a The lower the successors’ legitimacy in their entry into the company and Not at the moment of the succession, the lower the trust prevailing within the supported family firm. H8b The lower the legitimacy of family managers, the lower the trust prevailing within the family firm. Supported H9a The higher the competence, the greater the legitimacy achieved by the successor with the family members and employees. Supported H9b The higher the competence, the greater the legitimacy of family managers with the other family members and with employees. Supported H10a The higher the abuse of privileges, the lower the level of trust prevailing in the company. Supported H10b The higher the presence of shirking and misappropriation, the lower the level of trust prevailing in the company. Supported H11 The more impartiality in the treatment of family and non-family managers, the lower the agency conflict prevailing. Supported H12 The greater the moral hazard, the less impartiality in the treatment of family and non-family managers. Supported H13 The greater the moral hazard, the greater the risk of occurrence of shirking and misappropriation. Supported H14a The lesser the degree of comprehension of their social role, by owners and managers, the greater the risk of abuse of privileges. Supported H14b The lesser the degree of comprehension of their social role, by owners and managers, the greater the risk of misappropriation and shirking. Supported H15 The lesser the degree of comprehension of their social role, by owners and managers, the greater the risk of moral hazard. Supported H16 The lesser the degree of comprehension of their social role, by owners and managers, the lesser the impartial treatment afforded to family and non-family managers Supported H17 The more the altruistic selection, the greater the moral hazard. Supported Source: Research data VI.5. The Family Business Agency Conflict Model - validated version Below, the final model validated by the multivariate analysis carried out. 265 Figure 6.20: Reduced level of agency conflict (B1) Social Alignment + (C1) Family and Business Interaction + - - (B2) Strategic Alignment + (C2) Management Control - + + (A) Agency conflict - (B3) Trust + (C3) Transparency (C4) Legitimacy of family managers + (D.1)Manager´s competence - (C5) Abuse of privileges (C.6) Shirking and misappropriation + (B4) Impartiality - - (D.2) Comprehension of social roles (C.7) Moral hazard + (C.8) Altruistic selection + Source: Research data VI.6. Findings VI.6.1. Control variables for conflicts in family businesses I sought to identify if there are differences in the existence of conflicts when taking the following control variables into account: 266 Table 6.14: Summary of findings Control variables Findings Companies with versus those without Board of Directors. Through ANOVA significant differences were found, leading to the conclusion that firms with a Board suffer less from conflict. Companies with Board, which have independent board members versus those with only internal board members. No significant differences were found. Companies with versus those without Family Council. No significant differences were found. Companies whose CEOs are not from the family. It was not possible to compare because the power rates were low due to small sample and small effect size. Companies whose CEOs are from the second and third generation, compared to the first generation. No significant differences were found. Limited Liability Company in relation to Publicly Traded Corporation and Privately Held Corporation The limited liability companies had a higher level of conflict in two cases: between family and non-family managers and among family members with each other. Regarding the conflict between family managers and the family outside the company, family conflict levels are similar for the three kinds of companies Larger companies (more than R$ 60 million revenue) versus smaller companies. The analysis was carried out through the Spearman correlation which did not indicate any significant correlation, concluding that, according to this sample, there is no relationship between conflict and size. Companies that are performing below the average level for their sector. Higher levels of conflict between family and non-family members and among family members were found in companies with inferior performance. Another question that I sought to answer is if it could be argued that the fact of the CEO being family or nonfamliy had any relation to business performance. According to the Chi-Square Tests (p = 0.871), no significant differences were found. 267 VI.6.2. Analysis of the Family Business Agency Conflict Model as a whole Initially, one must assess the variance explained (R2) of the constructs, checking if the theoretical model displays adequate predictive power. Regarding the agency conflict construct, the value found of 38% is considered quite reasonable for studies in the area of applied social sciences. Cohen et al. (2003), for example, states that a correlation (R) of a size equal to 0.3 is moderate and values of 0.5 represent a high level. These values correspond, respectively, to the R2 of approximately 9% and 25% of the dependent variable and its independent constructs. According to this criterion, several constructs have a high proportion of variance explained, namely: Strategic alignment (58.7%); Impartiality (53.1%); Moral hazard (42.5%); Transparency (41.0%); Trust between owners and managers (38.9%); Shirking & misappropriation (35.8%) and Legitimacy of the family members (30.6%). But, even those constructs at levels lower than 25%, are still at moderate levels in relation to their coefficients of determination. This is the case of Social alignment (11.4%) and Abuse of privileges (20.6%). The fact of their explanatory capacity being lower, demonstrates that the model variables can explain only a relatively small part of the variability of these constructs. Identifying new endogenous constructs or trying to improve the validity/reliability indexes9 may be a way of better understanding the relationship of these variables with others in the model. The two constructs with low R2 values, the case of Strategic role of the Board (4.8%) and Successor’s legitimacy (3.8%) were withdrawn from the final model, as the significance of their correlations with the latent constructs were lower than the acceptable minimum. In fact, the great majority of the correlations were supported by the multivariate analysis, based on the criterion that a t value greater than 1.655 supports the relation in a one-tailed test with 5% of significance. The Successor’s legitimacy and Trust between owners and managers relation is the first to be refuted by a low t value: only 0.237. For the relation between Strategic role of the Board and Strategic alignment it can be seen that the t value 9 As the PLS is not as efficient in dealing with errors of measurement as the traditional methods (Maximumlikelihood, e.g. software such as LISREL) it is more susceptible to the attenuation of the correlations. This suggests that in such methods, the correlations could possibly be stronger. 268 borders significance for a one-tailed test: 1.822. But the one-tailed test may only be applied when the sign of the weight is equal to that expected under the alternative hypothesis. In this case, as there exists a weight contrary to that expected under the model hypothesis, the one-tailed test cannot be used with the critical value equal to 1.655 (one-tailed t value with 150 degrees of freedom and 5% of significance), the two-tailed test being appropriate with critical t value of 1.975. In this case, the null hypothesis cannot be rejected that the relation between these constructs is equal to 0, that is does not exist in the population. Therefore, the relation proposed by the model cannot be supported, but neither is there sufficient evidence to affirm that the relation is the contrary of that expected by the model. VI.6.3. Analysis of each construct and comparison with qualitative findings (A) Agency conflict construct According to the loadings identified via PLS, the item with the most influence in agency conflict construct is, by far, the Family owner/managers versus family managers with loading of 0.906 (XI.2). This is followed by Family owner/managers versus non-management family members (X.1) with a loading of 0.813, and finally Family owner / managers versus nonfamliy managers with loading of 0.758 (XI.3). It is worth noting that when evaluating descriptive statistics, we find that the highest average among the three questions is precisely the (XI.2). Therefore, it is clear that the most critical conflicts in family businesses are those that can be effectively identified by agency conflicts, which are those between the family members who being the principals also assume the role of agents. Also from this study, we learn that among the four latent variables related to the agency conflict construct, social alignment, strategic alignment, trust and impartiality, the latter is that which presents the greatest estimated standard weights ( = -0;265), therefore, that best explains the existence of agency conflict in the companies in the surveyed sample. 269 (B) Alignment Alignment was initially proposed as being a category arising out of the observations of the Milk & Candies case, explained by the strong fellowship existing between its two principal shareholders: Edgar Mills who exercises the role of shareholder/manager and Friedrich Haag, co-founder of the company in its present configuration. In fact, both of them also have, strong social alignment (encompassing principles, values, cohesion, integration between the families) such as strategic alignment (shared vision regarding the future), which pointed to these elements as being essential for the mitigation of agency conflicts. In reality, I initially took the proposal of alignment to be one unique factor, but the EFA indicated two distinct constructs, even if both ended by being considered indirectly related to agency conflict. The analysis of structural equations confirmed these relations, even indicating a greater weight of the Social alignment construct ( = - 0.194) than the Strategic alignment ( = -0.175) in relation to the agency conflict construct. It is important also to emphasize that via the Bental case, the importance of remuneration as a factor promoting alignment was revealed. The company has developed a management model called Network Company, which considers the staff more as associates than employees and whose essence is in the process of delegation of responsibilities, and in variable remuneration linked to the achievement of business results. The high commitment achieved made me understand that this had been one of the principal tools responsible for sustaining Bental in its turnaround at the most critical moments. During the EFA, this item was confirmed as part of the Strategic alignment construct. (B1) Social alignment The points examined in social alignment are two:  (III.1) alignment as regards the process and the moment of succession. For this item, descriptive statistics shows a neutral tendency regarding possible misalignment on such issues (4.08 on the Likert scale).  (III.7) the importance of the existence of a shareholders’ agreement or similar instrument designed to promote corporate alignment is the one with the highest load on this construct (0.825). In addition, descriptive statistics shows 270 that on average the respondents actually perceive its relevance (4.82 on the Likert scale). The variable that impacts this construct is family and business interaction with relatively weak relationship, although of some significance (= 0.338). (B2) Strategic alignment Regarding this construct, the items with greater loading are two:  (III.2) the one that identifies the sharing of vision and values between shareholders and managers, which however does not demonstrate a very significant level of agreement (4.87 on the Likert scale).  (III.4) the one related to the existence of an ordered strategic planning process. This also does not demonstrate a high level of agreement (4.71 on the Likert scale). Strategic alignment is impacted by family and business interaction ( = 0.299), but receives a greater weight from the variable management control (= 0.542). (B3) Trust The importance of trust had already made its first appearance in the first case, when the minority shareholder of Milk & Candies, John Harris, identified the essential point for supporting the Mills management, and accordingly reducing agency costs in the search for alignment between minority and majority, as being trust in the managers. And in fact, the multivariate analysis found that the greater the trust between owners and managers the less will be the agency conflict ( = - 0.166). It is worth noting that, regarding trust as a factor of selection, the clear preference is for family managers: 39% of the respondents state that they have gone through situations that led them not to unconditionally trust nonfamliy managers. On the other hand, 26% of the respondent companies have experienced situations that have led to decreased trust in family managers. 271 Concerning embeddedness 47% see that the indication of nonfamliy managers by members of their social network is the best way of guaranteeing the selection of trustworthy executives and 43% disagree with this position. It is also important to note that among the latent variables that are correlated with the construct Trust, the one that presents greatest weight is Legitimacy ( = 0.282). (B4) Impartiality In the Milk & Candies case practices of recognition & remuneration, evaluation and promotion of family and nonfamliy managers were identified apparently equitable, and there was an agreement that this was an important factor for the low presence of agency conflict in the company. However, it could be seen that with the arrival of the third generation, if the rules thought out were not implemented, the gain of a low incidence of conflict could be reversed in the future. The weight of – 0.265 sustains the relationship between Impartiality and Agency Conflict. Regarding the questions that constitute the impartiality construct, performance appraisal is the one that has the highest loading (0.900) and therefore the main item that would tend to impact positively or negatively on the perception of impartiality in family businesses. Analyzing the answers to questions related to the impartiality variable, some additional findings were made:  Question (VIII.1) shows that there is a slight majority of respondents, 48%, (sum of the options 1, 2 and 3 on the Likert scale) who disagree that family and nonfamliy managers should receive the same sort of error treatment. Another 42% advocate equal treatment (sum of the options 5, 6 and 7 on the Likert scale) and 10% are neutral (option 4 on the Likert scale).  This theme is also reflected in the perception of 45% of respondents in identifying in their organizations the existence of different criteria for the evaluation of the performance of family and nonfamliy managers. Another 45% see the performance assessment as equivalent (VIII.2). 272  However, for the hiring and promotion for and to key positions, 67% say they prioritize the level of competence of the candidate, regardless of kinship ties (VIII.3). (C1) Family and business interaction None of the four cases analysed had strained family relationships that had crossed the limits of the family and reached the company. However, the very fact of concern for the maintenance of family harmony, always emphasized in most interviews, showed us the importance of its verification in quantitative research. But in fact, descriptive statistics showed that there is no widespread disharmony in the sample studied. To question (IV.3), for example, the disagreement was relatively large regarding a possible lack of good interaction between family members (2.85 on the Likert scale). Item (IV.4) is the one that measures this construct with greatest loading (0.849). It exams to what degree family members are aware of the history and values of the company and also, it presents a high level of agreement in the descriptive statistics (5.64 on the Likert scale). It has appeared so evident in the Bental case, where we see that via the Family Council, there was concern in maintaining the trajectory of the founder and attention to the new generations knowing and preserving the foundations that led to the birth and growth of the company. The transmission of values from one generation to another was also measured and a small majority believes that family ties are strong enough to support this transition (4.62 on the Likert scale). Finally, the choice of social interactions in augmenting the ties between the controlling families and the employees (IV.6), as a category, came from the Milk & Candies case. In spite of living outside Brazil, Friedrich Haag attaches value to participation in social festivities during his trips here. This, in his view, increases the ties with family and nonfamliy managers, and has a positive effect on alignment. The descriptive statistics shows that respondents disagree, mostly regarding the existence of relatives' disinterest in maintaining social contact with the company and its employees (3.32 on the Likert scale). Although not having remained in the model after the factorial analysis it is worth investigating information collected in the survey on the Family Council. The 273 Family Business Survey conducted by PwC (2011) identifies the Family Council as one of the most common mechanisms adopted to minimize conflicts among family members, with a proportion of 44% of companies surveyed. It is interesting to note that this is the alternative most adopted in emerging countries, 70% of companies, different from the 39% of companies in mature markets, who are betting more on the adoption of a shareholders agreement (PwC, 2011). Thus, my survey found that this formal governance mechanism in family businesses is being adopted by only 27% of companies in this sample. However, 70% recognize it as a vital element in ensuring family harmony and a good relationship between the family and the company. (C2) Management Control This category emerged from observation of the Bental case, in its growth phase, when the absence of a formal control system made the process of alignment difficult, resulting in there being no shared targets and with every family member in the management conducting their responsibilities in accordance with their own business vision (or its absence). Allied to the great diversification, this was one of the principal causes of the crisis that supervened. Another point which contributed to the discussion of this category was the perception that companies whose founders are entrepreneurs, normally lack more developed controls, as because of their nature, such entrepreneurs tend to concentrate all the information in their own person and personally supervise the working of the company. The fact is that with the growth of the company and the natural increase in complexity, the need to establish management instruments which lead to the alignment of everyone becomes fundamental. In the case of Minas Agro for example, with the sudden departure of its founder from the management, such alignment was damaged profoundly, as he, personally, was the director of the organization’s destiny, without formal plans that would allow the successors to engage more easily in the business and share the same ends and objectives with the nonfamliy managers. Through the descriptive statistic, we realize that while 64% of family-owned firms claim to adopt a rigorous internal control system to monitor the operations of 274 their company, 56% of them recognize that they have already experienced situations where the managers have taken decisions harmful to the company, which were not detected in time by the shareholders, due to them having adopted a more laissez-fair oversight. Even so, 70% of respondents say the economic and financial monitoring by the shareholders and their representatives occurs on a regular and continuous basis. It is noteworthy that 71% of companies believe in negotiating the levels of autonomy in order to guarantee agility in decision making. Another point to highlight is that although 83% of companies surveyed agree that CEOs, even if members of the controlling family, should be evaluated for their performance, only 38% of the family-owned companies surveyed adopt a clear and formal process for the evaluation of the performance of the CEO. It is worth pointing out that management control has an even higher weight in relation to the construct transparency ( = 0.641) when compared to its relation with strategic alignment ( = 0.542). (C3) Transparency In relation to the transparency construct, it is noteworthy that 68% agree that the succession process is not openly discussed regarding the criteria for its implementation. Also, 74% of the respondents confirm that the executives from their firms have the habit of providing information, positive and negative, over and above what is obligatory, to all stakeholders, to facilitate the correct comprehension and evaluation of the organization. On the other hand, regarding disclosure, firms show concern in establishing the essential controls, financial and non-financial, giving unrestricted access to their stakeholders, as mentioned by 70% of respondents. There is, of course, a difference between intention and putting into practice, since only 57% of them admit to having effective channels for disseminating information. 275 (C4) Legitimacy Legitimacy was the construct most often mentioned across the cases studied. Starting with Milk & Candies, trust existed between the minority and majority shareholders, the latter at the head of management, due to the competence demonstrated in carrying out their duties. It is important to note that trust through competence was also achieved by the heirs of the 2nd generation of the Mills family, but the arrival of the 3rd generation into management brought legitimacy to the Model as an important category. This happened because, despite the existence of rules for family members’ entry in management, these were neither formalized nor followed. From Bental, we saw a well prepared process for building the legitimacy of Pedro Talma, 3rd generation, in his rise to become CEO and the establishment of rules for the entry of family members into management. From Prize, the category legitimacy also appeared in evidence. Elton had great difficulty in establishing himself with the nonfamliy managers, a team that had been formed by the founder. His father’s degenerative disease, which gradually removed him from the company’s daily activities, together with the fact of not having had any specific preparation, made this a hard task. The staff did not recognize his competence and thought that he was there only because of family ties. Also in Prize tensions were identified as regards the entry of the third generation, demanding a shareholders' agreement that established clear rules. Legitimacy also appeared as the predominating factor in the case of Minas Agro. The disregard of the building of the successors’ legitimacy, probably due to their youth and the premature death of the founder, resulted in great difficulty arising in the relations between the owners and the nonfamliy managers. The founder’s wife, in spite of being respected by all, did not have the capacity to take over the company, and the daughters, who were graduates in the area, were rejected by the managers because they were not identified with the business. Some of these managers even had to be substituted, which generated tensions and loss of accumulated knowledge in the company. When statistical analysis was carried out, the importance of legitimacy in the model construct was confirmed by the significant relationship between Legitimacy of the family members and Trust between owners and managers ( = 0.282). 276 In this construct a slight convergence in the respondents’ replies (4.6 on the Likert scale) that family managers are recognized by all employees as competent for the exercise of their current functions should be mentioned. On the other hand, the rules for the entry of family members into the company still do not appear to be clear in the majority of companies, which could be a factor making the process of legitimation difficult (3.8 on the Likert scale disagree that they are clear and/or obeyed). Although the successor legitimacy construct is not supported in the theoretical model, it is worth mentioning some other findings related to it. For example, 70% of respondents believe that the principal challenge faced by the successor is the difficulty of obtaining immediate recognition of his/her authority from family members and staff in the running of the company. As for the possible shadow that the predecessor could generate over his/her successor, there is a division of perceptions: 49% of respondents think the image of the founder is so pervading and honoured by all, that it inhibits the family successor in naturally coming into his or her own. Another 45% disagree with this position. Interestingly, 43% of these companies are still in the first generation, which could cause distinct perceptions between predecessors and successors. (C5) Abuse of privileges In no case were there explicit situations of opportunism via expropriation of minority rights by the majority, nor did expropriation through remuneration and excessive bonuses exist. But the concern precisely that such acts should not occur in any circumstances, led to the test of the hypothesis that the abuse of privileges leads to reduction in trust, a hypothesis which was verified, albeit with a degree of significance near to the borderline (t value of 2.177). And in fact, through descriptive statistics (V.3 and V.5 questions), we could see the high level of agreement of respondents regarding the absence of abuse of privileges in their companies, neither by the executives, nor on the part of the majority against the minority. 277 (C6) Shirking & Misappropriation The Milk & Candies case also discussed opportunistic behaviour through practices of misappropriation. The company was the victim of the opportunistic action of nonfamliy managers, fraudulent activity that damaged the company. The consequence was the dismissal of several managers, the solution for the agency conflict that had been established. Also in this case, the concern of the founders in trying to avoid practices that confounded personal with company property can also be considered as an element that increased trust, not only of the other blocks of owners, but of the staff itself that identified this attitude as ethical. The presence of shirking was clearly the consequence of the altruism reigning in Bental, in its first decades of existence, where several family managers were working with less than the effort needed to achieve the required results. The descriptive statistics shows that the practice of shirking does not seem to be a reality common in Brazilian family business (question V.1 with 3.36 on the Likert scale); even less the existence of attitudes of misappropriation by managers (V.2 question with 2.13 on the Likert scale). In fact, respondents show total disagreement relative to attitudes of misappropriation, even when only considering the 'use of company resources for his or her own benefit' (question V.4 with 1.88 on the Likert scale). Throughout the PLS analyses, we found a significant relationship between Trust and Shirking & Misappropriation with weight of -0.227. (C.7) Moral hazard In the case of Bental, the fact that the founder, Benedito Talma, does not see the company from a predominantly economic perspective, resulted in the presence of altruism with manifest moral hazard, which has consequently led to shirking. Multivariate analysis showed relationship of both variables: Moral hazard and Shirking & Misappropriation ( = 0.319). In the same case, it is observed that moral hazard leads to a justification of opportunistic behaviour of family members, who start to be treated in a privileged manner in the treatment of errors, which would not be accepted in the case of 278 nonfamliy managers. The multivariate analysis also demonstrates high significance in the relationship between moral hazard and impartiality ( = -0.475). Among the questions that measure this variable, that of greatest loading (0.791) is (VIII.7) that affirms that Family affection leads to unconditional acceptance of behaviour related to the firm, even if inappropriate. There is, nevertheless, strong disagreement (2.8 on the Likert scale of 1 to 7) that would lead to the conclusion that the respondents are not favourable to altruistic practices. In the same way, another altruistic perspective is denied by the respondents in indicating in question (VIII.4) that the well-being of the family members is not prioritize to the detriment of the company’s economic objectives. What in fact appears to raise doubts are the neutral replies of the respondents (3.9 on the Likert scale) to question (VIII.8), regarding the continuation of the family managers in the company even when they discharge their functions less than competently. This is typically an altruistic attitude that leads to the moral risk of accepting family members in managerial functions even when one knows beforehand that nothing will be done about inappropriate behaviour/performance. (C.8) Altruistic selection Also in the case of Bental it was clear that altruistic selection in the choice of the family for management had favoured the appearance therein of attitudes of moral hazard. And. it is sustained via multivariate analysis that altruistic selection leads to moral hazard ( = 0.284). The question (VIII.5) has high loading on this construct (0.990), and descriptive statistics shows a slight disagreement that the respondent companies have been a natural way to accommodate family members who wish to work in them (3.52 on the Likert scale). (D1) Manager´s competence Looking at the qualitative analyses, the variable was a relevant category in two cases. The Milk & Candies case presented the breakdown of trust in a nonfamliy 279 CEO, not because of opportunistic attitudes, but inability to meet the new demands and assume the risks which the shareholders wanted in the search for growth, a typical action of agency conflict due to incompetence, in accordance with that cited by Hendry (2002). In the case of Bental, in spite of the tough measures introduced by Giuseppe Mancini with the dismissal of the entire family, his attitude helped the family itself, after a while, to understand the need to establish limits for the entry of family members into the company based on criteria of competence. The variable manager's competence has a significant weight on legitimacy (= 0.553) which shows its importance in the model. Looking deeper into this variable, we found that a significant number of family businesses, 80% see their managers as the main cause in the achievement of their strategic outcomes. This percentage contrasts, however, with a not inconsiderable number of 35% of respondents who do not see the family managers as having sufficient capacity for conducting the significant necessary changes, if they require the breaking of traditions in the company (59% see to the contrary and 7% are neutral). And 51% of them identified nonfamliy managers as more ready to take risks in times of change than family managers (41% disagree with this position and 8% have neutral opinion). (D2) Comprehension of social roles The comprehension of one’s social role has various direct relationships, as supported by the model. When there is no clear distinction of the role of being a family member or an employee, this could leads to opportunistic behaviour: shirking & misappropriation ( = -0.352) and abuse of privileges ( = -0.453). This occurs also regarding altruism: the mixture of roles leads to moral hazard ( = -0.535). Thus, the reasons of the existence of a direct relationship between this construct and Impartiality ( = 0.340) is easily understandable. 280 CHAPTER VII DISCUSSION VII.1. Closing remarks At the end of the last century the literature tended to concentrate its studies more on public companies than family businesses (Litz, 1997). Lansberg et al. (1988) argued that this was happening due to the widespread belief that the distancing between capital and control was a definite tendency in general business, adopting an increasingly professional management instead of being owned and controlled by families. However, family businesses are still being not only seen as a viable form of enterprise, but are also increasing their importance in the economy of most countries. This leads to a growing body of researchers willing to investigate a wider range of issues concerning family business as noted by Craig and Salvato (2012) The debate over whether family business is a worthwhile research pursuit has been won. Whether the metric is the number of special issues in highly ranked journals, the number of minted PhDs, the list of established scholars publishing family business papers, open faculty postings that include family business as a preferred concentration, or citation impact factor standings, irrefutable evidence can now be provided that pursuing research in the family business space is not the career death knoll it arguably once was (Craig & Salvato, 2012, p.109). It is, however, important to identify which fields of research in family businesses have been given more attention from scholars. Yu et al. (2012) have evaluated 257 empirical family business studies from 1998 to 2009, identifying 327 different dependent / outcome variables. The authors categorized these variables in seven clusters: performance, strategy, social and economic impact, governance, succession, family business roles, and family dynamics. Among them, governance appears with the greater frequency of use (20.47%), showing how significant it became in the current reality of family businesses. On the other hand, conflict in family-owned firms, which is the main subject of inquiry in this thesis, represents just 3.13% of frequency of use, which confirms the importance of this research in deepening studies in one facet of governance of family-owned companies yet little explored. 281 The conflict that has been shown throughout this study to be the most important for analysis is the one between family members themselves in their familyowned firm. That is, family members in the role of agent, tend to develop agency conflict more easily in relation to the principals that they represent, than do nonfamliy managers. This can be explained because in this type of company, ‘the familial relations, conflicts escalate much more easily and can rapidly shift to the personal level’ (Frank et al., 2011). This is an important finding, where the prevalent belief that trust arising out of kinship is in itself sufficient to mitigate agency conflicts, not taking into account a number of other key elements in the enhancement of agency conflict, as addressed by this study. Recognizing that family firms are not homogeneous entities, I decided in this study to investigate the phenomenon of agency conflict in Brazilian family-owned companies, regardless of the constitution of their capital or the type of involvement of family in business: whether they were family-owned, family-governed or familymanaged. Of the four companies analysed as case studies, two were privately held corporations and two were limited liability companies. But the four were all run by family members of the controlling family. In the survey, questionnaires were distributed to family businesses in general. The final sample was represented by 61% of limited liability firms, 34% privately held and 5% publicly traded corporations, which was a good representation of the spectrum of Brazilian family businesses, medium and large. As for the participation of family members in the management, the sample shows that only 9% of companies were managed by a nonfamliy CEO (which does not mean that family members were not working in other executive functions). Therefore, we consider that this research deals with agency conflicts in family businesses where the controlling family effectively manages the company. It is important to note that 83% of the respondents were family members, and among those who reported, 57% are from the second generation, 21% from the first, 18% from the third and 4% from the fourth generation. This distribution was important since it avoids possible intergenerational tension bias. Moreover, 70% of the respondents are shareholders, 89% work in the researched firms and among the companies that have a Board, 29% are Directors. 282 This demonstrates that I had access to a privileged public, with good knowledge of the dynamics of family and business relationships and with adequate capacity for critical analysis that requires a survey of this nature, including sensitive issues such as research on the existence of conflicts. I should stress that the questionnaires were completed by manufacturing, services and agribusiness companies. Also, companies from Brazil's major economic regions have taken part: south, southeast and northeast. This is important to ensure a sample that is representative of a country with such a huge territorial extent and with significant regional cultural differences. Therefore, this sample has made possible the development of a theoretical model, identifying the principal inducers of agency conflict in family businesses, which was the main objective of this thesis. This goal was accomplished with a set of interrelated variables of the model able to explain 38% of the variance in agency conflict. VII.2. Contributions to knowledge Academic research is ultimately about theory – the lenses through which we view, interpret, and test real-world data. For Colquitt and Zapata-Phelan (2007) theory may be determined in terms of relationships between independent and dependent variables. Other scholars, on the other hand, have defined theory in terms of narratives. Following this perspective, as suggested by Eisenhardt (1989b), ‘theory is evaluated primarily by the richness of its account, the degree to which it provides a close fit to empirical data, and the degree to which it results in novel insights’ (Colquitt & Zapata-Phelan, 2007, p. 1281). The use of grounded theory via case studies has enabled the construction of accounts which after being categorized and grouped as factors have resulted in causal relationships being established between the factors, giving meaning to the model developed, as recommended by Whetten (1989): Relationships, not lists, are the domain of theory. [...] theoretical insights come from demonstrating how the addition of a new variable significantly alters our understanding of the phenomena by reorganizing our causal maps. (Whetten, 1989, p. 492). 283 The final product of this research, entitled Family Business Agency Conflict Model, is a genuine contribution to the theory and practice of corporate governance in family businesses. VII.2.1. Theoretical and empirical contributions Agency conflict is typically seen as a phenomenon of large corporations resulting from the pulverization of the share control which ends by promoting the separation of ownership and management. This separation brings with it as a consequence a new role for the shareholder who ceases to participate in the daily running and basically confines him/herself to evaluating the results and investing capital in the most profitable. Thus the owners start to procure professional people in the market for taking the decisions that maximize the value of the corporation, the wealth of the shareholders and the return on investments.It is just that once contracted and having a large part of the control of company decisions in their hands these managers can try to further their own ends, generating what is identified as agency conflict. A good part of the scholars, until recently, understood that family companies given the absence of a separation of ownership and management, would be immune to agency conflict. But, in fact, this only occurs when the company is run by the founder without any partner. But from the moment the founder acquires a partner each with his/her distinct roles, there exists potencial for agency conflict. If the owning family chooses to contract an executive to run the company, agency conflict becomes latent, and in succession processes with the entry of second and subsequent generations when a family member assumes the role of being the agent of the others, that is being the main executive acting on behalf of the other family members, agency conflict is potentialy explosive. The result of this work concerns basically three possibilities of conflict in the family company: family managers versus family managers; family managers versus non-family managers and family managers versus family members outside management. In the universe researched the confict with the most impact occurs in the first case: between family managers (loading of 0.906), which is an important empirical contribution to agency theory in family companies. 284 The question that should be asked is: how can one inhibit the appearance of such agency conflicts that cause attrition and impact business dealings? The most usual tools in publicly held companies are surveillance and incentives to align interests which end up being unsufficient for family companies because of their particular characteristics and complexity. Chami and Fullenkamp (2002), reinforce these usual elements of surveillance and incentives as the key factors in the management of the agency conflicts Two general strategies for dealing with agency problems have emerged from the academic literature and found practical application in many corporations. One strategy is to avoid the agency problem entirely by expending some effort or paying some cost to improve the principal’s ability to monitor the agent, so that the principal and agent can contract directly on the task to be performed. […] The other strategy is to align the incentives of principal and agent. The main mechanism for aligning incentives, which was suggested in the voluminous academic literature on executive compensation, is through compensation contracts that alter the agent’s incentives. For example, by making executive compensation depend on a measure of firm performance generally, the stock price - shareholders give executives the incentive to act in the shareholders’ interests (Chami & Fullenkamp, 2002, p. 1787). However, although both strategies are correct, the belief that it is possible to mitigate agency conflict only via monitoring and aligning incentives is an extremely reductionist view, since the environment of the family business is governed by more complex issues. This study represents a pioneering piece of research in the study of agency conflict in family businesses, by identifying three main alternatives for reducing conflict: promoting alignment (either strategic or social), enhancing trust and embracing impartiality. Thereby, it contributes to knowledge by proposing safe pathways for family business to escape from threats and losses arising from divergences between principals and agents.  Mitigation of agency conflict in family firms by the promotion of social alignment Strategic alignment is linked primarily to the existence of cohesion between partners, translated by some formal instrument such as a shareholders' agreement, for example. Its existence typically reduces the existence of agency conflicts. 285  Mitigation of agency conflict in family firms by the promotion of strategic alignment Strategic alignment is the convergence of strategic purposes in terms of the same vision, consensus on the path to follow, common goals, and acceptance of the same level of risk. It is achieved on the basis of shared values and principles between managers and shareholders. Thus, the presence of managers who do not share the same values as the owners will probably be a source of conflict. So that, incurring agency costs in the promotion of alignment is essential for the reduction of conflict, as pointed out by several authors (Ensley & Pearson, 2005; Haugh & Mckee, 2003; Pieper et al. 2008). Our model also reveals, given the strong significance between both constructs, the importance of adopting systems of control and monitoring of managers for the increase of strategic alignment.  Mitigation of agency conflict in family firms by the enhancement of trust. In the model developed in this study, trust is negatively impacted by opportunistic behaviour and abuse of privileges, and shirking & misappropriation. Therefore, avoiding such attitudes is essential so as not to undermine trust, which would lead to an increase in agency conflict. At the same time, implementing a culture of transparency is a factor that impacts positively on trust. But, ultimately, the legitimacy of family members’ is the key factor to attend to, if we want to increase trust in the family business. And it must be remembered that the model generated in this study, indicates that competence is the key factor for legitimacy occurrence. Thus, subsequent generations of the family should be the subject of a succession plan that gives them legitimacy before taking executive positions in the company, avoiding the emergence of conflict. 286  Mitigation of agency conflict in family firms by the embracement of impartiality. This study indicates that companies are less susceptible to the presence of agency conflicts, complying with criteria of justice and meritocracy in the evaluation of their managers, adopting fair error handling practices, not distinguishing between family and nonfamliy managers and avoiding nepotism. This finding is in line with Van der Heyden et al. (2005) who identify lack of fairness in the management decision-making process as one of the main limiting factors on commitment and trust of the individuals involved with the family business. To sum up, this thesis is an important contribution to the theoretical analysis of agency problems in family businesses, probing further than the usual variables used to reduce agency conflicts. In fact, the model developed in this thesis includes compensation and incentives as an item that promotes strategic alignment (loading of 0.731). But this is just one factor among several others. These have an even higher loading: Vision & values (loading of 0.811); Harmonic actions / attitudes (loading of 0.797); Ordered planning process (loading of 0.791) and Risk taking (loading of 0.762). Therefore, the fact of just valuing the incentives to promote alignment makes the relationship merely mercenary, neglecting other more relevant criteria, as demonstrated by this model. It should also be noted that in family businesses where the manager is a family member, the incentive via compensation has even less impact, since, as GomezMejia et al. (2003, p. 19) affirm ‘executives with family ties to owners receive lower total pay than professional managers’. Thus, the compensation factor carries less weight with family managers who require less calculating and utilitarian contracts than do nonfamliy managers, because they are more emotionally attached to the firm and because they have fewer employment options in an open labour market (Gomez-Mejia, et al., 2001; Gomez-Mejia et al., 2003). Monitoring is also part of the proposed model identified as Management Control construct. It relate both to the strategic alignment ( = 0.542), as to transparency ( = 0.641). Transparency in turn affects trust ( = 0.190) in a directly proportional way in the same manner that the legitimacy of the family managers does 287 ( = 0.282). This myriad of correlations shows the number of elements that are part of this model to explain agency conflict, confirming the level of the theoretical contribution of this study, which goes beyond the understanding of the variables ordinarily presented by the literature. Thus, this study aimed to develop a better understanding of family firm governance by addressing both the procedural aspects (such as boards, agreements, systems, procedures) that regulate the relationship between the owners and the management, and the social aspects of governance that are considered important to promote a healthy interaction between the family and the company (shared vision, trust, impartiality, altruism, competence, and so on). Therefore, this thesis submits a more comprehensive analysis, through a model that proposes and tests relationships among constructs of different nature in a family firm governance context. VII.2.2. Practical Implications Beckhard and Dyer (1983) pointed out the risks that may generate conflicts that paralyse decision-making processes in family businesses. Glasl (2002 cited in Frank et al., 2011, p. 131) therefore advocates the development of a conflict management system to detect and prevent conflicts emerging and entering into an escalation process that is difficult to deal with. As described herein agency conflicts tend to impact corporate performance, and since family-owned firms are particularly prone to conflicts, the value of the proposed model is extended. Through the analyses carried out, it is possible to return to the questions initially posed in the early development of this thesis, whose answers make an immediate practical contribution to the reality of family businesses. 1 Who are the main actors in conflict in Brazilian family-owned companies? The conflict in family businesses is mainly between family managers. This finding is obtained both in the multivariate analysis with its greater loading on the construct, and in the analysis of the means, through descriptive statistics. It is 288 observed therefore, that when a family member holds a management position, his/her chance of experiencing an agency conflict with other family members/owners who are also in management, is greater than in the case when a nonfamliy manager is appointed. In fact, authors such as Kellermanns and Eddleston (2004) recognize that, the more concentrated in the hands of the family the management is, the less will be the level of conflict related to the execution of tasks. However, the tendency for personal conflicts to occur between the family members, because of the superposition of roles, will be greater. 2 Is the type of capital structuring of the company related to the level of conflict? Agency conflict (between owners and managers, family members or otherwise) was found to be higher in limited liability companies. That is, the legal requirements and the resulting professionalization of corporations are factors reducing agency conflicts. In this sense, authors such as Songini (2006) affirm that the process of professionalization is important in attenuating the informal atmosphere of the organization, introducing more formalized systems and, consequently, increasing the systematic use of information to orientate management. That is, it is possible to infer that family-owned companies that exhibit more complex capital structures, in adopting more formal management models and more efficient control systems, will be able to minimize the level of agency conflict existing in the organization. 3 Does the size of the company influence the level of conflict? Initially, it should be pointed out that the low number of large family-owned companies is frequently attributed to the desire of the owner to restrict the growth of the company to maintain his control and property in the family (Daily & Dollinger, 1993). Other studies, however, such as those of Ward (1997), Loan-Clarke et al. (1999) and Harris et al. (2004), attribute the reduced number of large family-owned companies to a lack of efficacy and the informality of their management systems. 289 Frank et al. (2011) claim that there are different views in the literature on the causes and implications related to the size of family-owned firms in the market. The fact is that the findings of this research indicate that the size of the company does not influence the level of agency conflict. Thus, it is not growth that makes a company more susceptible to agency conflict. Rather, the way it develops and converts its management and governance in a professional manner tends to reduce such conflict. 4 Do companies with economic and financial performance better than the average of the segment to which they belong have a lower level of conflict? Does inferior performance imply a rise in it? It was shown that firms with performance below the average level of companies in the sector have increased agency conflict. Kellermanns and Eddleston (2004, 2007) postulate that the greater the level of conflict, the greater will be the tendency to poor performance in the company. Likewise, Elbanna’s (2009) studies done on Egyptian companies identified that affective conflict is a significant predictor of firm performance. He himself cites Amason (1996) who had already reported the significant and negative relationship between affective conflict and organizational outcomes. Several authors such as Schulze et al. (2003b), Dyer and Handler (1994) and Kets de Vries (1993) also call attention to the low efficiency in family-owned companies associated with psychological conflicts, such as, disputes among successors, autocratic leadership and the spreading of family conflicts into the business field. 5 Are the conflicts very different in a family firm managed by the first generation owner-manager from those in a family firm controlled by a second, third and subsequent generations and an extended family? Several authors like Davis and Harveston (1999, 2001) and Frank et al. (2011) seek to establish the relation between the level of conflict and the generation responsible for management. For these authors, the level of conflict will be greater in 290 family-owned companies run by the third generation if compared to organizations administered by the second or by the founder. On the other hand, independently of the degree of conflict, Westhead et al. (2002) found in their work that multigeneration companies, run by two or more generations, tend to be better administered than first generation companies. In the present study, I have not identified a higher level of conflict in the second and third generations, compared to that in the first. From this result it is possible to infer that the generation in power should not be taken in itself as a central category to explain the breaking out of conflicts, but should be analysed with reference to its constituent elements such as the increase in the complexity of management; the presence or otherwise of regulatory mechanisms of the relations between family and company; the manner in which power relationships form between the generations and; the degree of access of the different generations to the decision-making process in the company. 6 Do companies with nonfamliy CEOs have a higher level of conflict? Due to the small number of cases of respondents in this situation, I cannot address this question. Nevertheless, if we revert to the literature, it is possible to find several authors like Chitoor and Das (2007) who maintain that passing over management to nonfamliy managers minimizes considerably the risk of conflicts during the succession process and the day-to-day activity in the company. This view is not new, Levinson (1971, p. 98) having been one of the first to set out the various intra-family conflicts that surround succession in family-owned companies, recommending to the entrepreneurial family that it: ‘move to professional management as quickly as possible’. Dyer (1986) perceives the option of delivering over the management to nonfamliy professionals as a rational alternative to confront the nepotism and family conflicts that plague family-owned companies. The author further argues that nonfamliy CEOs lead the business to a new level of leadership and, specifically to a new strategic direction, focused on growth and expansion. Westhead and Howorth (2006), however, stress that it is common to find owners of family businesses who prefer the CEO to be a member of the owning 291 family, in order to ‘protect the store of family wealth, as well to provide jobs for current and future family members’ (p. 305). On the other hand, Dyer (1989) also warns of the difficulties inherent in choosing nonfamliy managers by indicating that, very often, a lack of understanding can be seen on the part of them in relation to the human aspects that traditionally guided the company. In addition, nonfamliy managers tend to present a short-term view focused on financial business performance, so that faults in communication can occur, and a lack of synergy between the new CEO and the members of the business family, leading to serious problems, misunderstandings and even real agency conflicts (Hall & Nordqvist, 2008). 7 What about the tools and practices of management control: are they found to be in evidence in the Brazilian family business? Authors such as Songini (2006) emphasize the importance of formal systems of control and management to guarantee the efficient running of operations and stimulating the interest of shareholders in the business. Sundaramurthy (2008) stresses the difference between the evolution of control systems in family and nonfamliy-owned companies. For him, in the case of nonfamliy-owned companies formal contracts and control systems represent essential prerequisites for starting up business. In their turn, in family businesses the opposite tendency can be seen, where initially family relationships are responsible for sustaining informally the whole organizational system and, only with the development of the business, formal complementary control systems assume significant importance in stimulating the necessary competencies and the satisfactory monitoring of the business (Sundaramurthy, 2008). In the family businesses surveyed here, there is a clear emphasis on financial control by shareholders and directors, where 70% of respondent companies indicate that they do regular monitoring of indicators of this nature. The companies also seem to be acting through rigorous internal control systems to monitor their operations, as indicated by 64% of the respondents. Thirty percent of the respondents admit that their control systems are fully adjusted to their needs. However, when questioned on the occurrence of management decisions that 292 were harmful to the company and that were not foreseen by shareholders due to slack control, 38% admit that this has happened in their company. 8 What is the real relevance of transparency and disclosure in family business and how can it be achieved? This relationship between trust, management control and transparency can be found in many studies in the literature of the field. Sundaramurthy (2008) defends the idea that transparency, the formalization of operational activities and a good system of communication can reinforce a positive cycle of trust. Thus, a family company that heeds these essentials can reinforce interpersonal trust between its members and contribute to the maintenance of organizational trust as the family evolves. In this sense, it is fundamental to make transparent the rules and traditional values which guide the entrepreneurial family, to engender trust. In their turn, authors such as Paine et al. (2005) maintain that transparency is not synonymous with total openness, but presenting with certain precision relevant information that could impact directly the financial or personal health of the members of the family organization. In this sense, the clear divulging of results and important information permits the integration of family values with those of the company within well-defined limits and helps to establish confidence in the family business system with all its stakeholders. This transparent stance may also be essential to communicate the desire of the family to maintain the business, but at the same time stresses the commitment to meritocracy and not nepotism. (Ward, 2004). 9 Are the criteria for succession negotiated and made explicit in the Brazilian family business? And also, is there convergence on the natural process and the timing of its implementation? Is competence the most considered factor for hiring and promotion, regardless of the ties of consanguinity? Regarding succession, there is a division in the perception of the existence of alignment in the succession process, as well as on the right timing for its execution: 39% who agree and 45% that did not perceive its existence. And a significant proportion of 63% point to the absence of a succession process that can be openly 293 discussed and treated with transparency as to the criteria for the choice of the successor. For Ward (2004) there is a visible relation between transparency in the succession process and agency conflict. For this author, the establishment of clear policies to guide the succession, regulating the entry of family members into the business and establishing salary standards and the criteria for promotion of family members can reduce the ambiguity of the process and minimize the risk of occurrence of conflicts. Concerning competence as a primary factor for selecting the successor, there is a clear division of viewpoints among the respondents: 47% believe that despite the fact that they consider competence as an important criterion, kinship ties are still the main consideration in choosing the successor. Another exactly 47% disagree with this position. For Dyer (1989) competence would be fundamental for the survival of the family company. In this sense, the author points out that as the company grows, the search for the abilities needed to run the business in the different areas (financial, marketing, controller) should be placed above the ties of kinship. Thus, particularly in a complex environment, it is improbable that the family holds all the talent necessary to fill the key positions in the company. In this context, in addition to the incentive to the training of family successors, the option arises of obtaining nonfamliy managers in the marketplace to guarantee the maintenance and development of the business (Dyer, 1989). 10 Is the Brazilian family-owned company basically managed by family members? Are they considered competent acting as managers? Based on the selected sample, I can say that the two main functions of command are in the hands of family members. Ninety-one percent of the familyowned companies are led by a family CEO. Thirty-nine percent of the companies said they had boards of directors, and in 93% of these the Chairman was a family member. 294 Regarding the family managers' competence, it is observed that 35% believe that a family manager is not capable enough to lead significant changes, if that requires a rupture with the traditions of the company (59% disagree with this statement and 7% are neutral). As 83% of the respondents are in fact family members, one cannot say that 35% is negligible percentage. Furthermore, 51% reported that nonfamliy managers are more prepared to take risks in times of great change, than the family managers (41% disagree with this statement and 8% are neutral). Therefore, this research shows that there is a majority view of the family members themselves involved with the business, that family managers may not be the best alternative to lead turnarounds. This idea is confirmed by Kesner and Dalton (1994) who argue that external managers are best placed to take on companies which are performing poorly, and raise them again, because they are least compromised by the past. In fact, authors such as Donckels and Fröhlich (1991) and McConaughy et al. (2001) say that family managers tend to present an attitude averse to risk, and prefer a conservative and informal management style. This view is shared by Westhead and Howorth (2002) when they provide evidence that family managers do not generally place their trust in outsiders, even those professionally qualified, to take over the management of the family patrimony. The authors point out that this reluctance to rely on outside managers and investors can retard the process of growth or even impact negatively the survival of the business (Westhead & Howorth, 2002). In their turn, nonfamliy managers usually work to a short-term time frame, orientated by profit and attracting investments, to reach the goals established by the owners. In addition contracted agents tend to set up more formal management systems, making possible a more transparent accountability process to make shareholder control more feasible (Daily & Dollinger, 1993; Kotey, 2005). 11 Has the Board of Directors been a feature of governance adopted by Brazilian family businesses? Is its presence a factor in mitigating agency conflicts? What are its level of independence and efficiency? 295 As already pointed out, 39% of the sample have Boards of Directors, demonstrating that although the majority is not required to do it by law, as most of them are not listed firms (in fact, only 5% are), they have made such a choice because they recognize their value. And indeed, it seems that one of the measurable benefits of adopting Boards of Directors, as demonstrated by my research, is that companies that have them showed lower levels of conflict. In fact, the literature indicates that a Board of Directors is particularly important in situations where the potential for agency conflicts is most critical (Baysinger & Butler, 1985; Pearce & Zahra, 1992). To ensure the independence of the Board, among the 39% who adopt such an instance of governance, 42% have outside directors. Fama and Jensen (1983a) agree with the importance of balancing the presence of inside and outside board members to have an effective and independent board. Cotter and Silvester (2003) add that firms with boards that are dominated by insiders may suffer from collusion, which certainly affects the alignment. These authors also affirm that external members may act in the arbitration of the disagreements. It was therefore expected that the relationships proposed in the theoretical model regarding the level of independence and efficiency of the board (via its constructs derived from the EFA) and the alignment influencing, ultimately, the agency conflict would be supported through the multivariate analysis, but this did not happen. One possible reason is the fact that 61% of respondents do not adopt the Board of Directors. As the questions were framed so that even companies without a Board could answer them, it appears that most of the responses have been given by the respondents replying as if their company had a board, rather than their real situation. I can also infer that one reason for the hypothesis relative to the strategic role of the board and family power on the board was not supported, may be that the power of owners was perceived by the respondents to be greater than that of boards. Indeed, over half of respondents (55%) claim that the decision-making power has to be concentrated in the family, the independent board members, therefore, not having the same level of authority. This represents an apparent contradiction, since 64% disagree that in its constitution, all family blocks should be represented, rather than guaranteeing the presence of experienced and competent board members. Similarly, 296 64% of the respondents disagree that the presence of external board members causes discomfort. Other issues related to the Board may also be mentioned. For example, 74% of the respondents agree that a Board of Directors has to have a more directive role considering it is the ultimate instance in making strategic decisions. Also, 69% say they would never accept a rubber stamp board. In addition to answering the initial questions, some other practical suggestions to be adopted by family businesses that can mitigate agency conflict are: i. Promote strategic alignment, since this is a cause of agency conflict. In addition to my findings, I turned to PwC's research (2011) which indicates that the main cause of discord in family businesses is disagreement about the future strategy of the business. ii. Family and business interaction was found to be a process central to promoting the formation of a shared vision and strategic alignment. Owner families can induce more varied social interactions by implementing various family institutions, such as family council and shareholders agreement. Also research of PwC (2011) shows that the most common mechanisms to minimize conflicts among family members are the shareholders’ agreement and the family council. The adoption of both is highly recommended to prevent the emergence of agency conflicts. iii. Mandatory disclosure is one that exists in order to produce information to meet legal requirements of the stakeholders. It is suggested that the vast majority of companies not listed, such as the sample used, adopt voluntary disclosure as a way to keep themselves always attractive to new investment opportunities, when they need or choose them, not suffering setbacks when a major change in a process of going public is requested. Such transparency, as already discussed, generates trust, which in turn minimizes agency conflict. 297 iv. The results of this research Indicated that legitimacy improves trust, which consequently diminishes the risk of the emergence of conflict. Therefore, developing the legitimacy of family managers, either establishing rules for their entry, or building their competence in the fields of knowledge, skills and attitudes is a recommendation for owners who want their family members assuming management positions in the companies. v. Understand that altruism is a natural attitude of parents toward their children and therefore this will hinder a critical view of the level of their actual competence, which is essential for one to work more professionally in the selection process, evaluation of performance and promotion. Adopting existing tools on the market is a way to avoid the mistakes of altruistic selection, which can lead to attitudes of moral hazard. Otherwise, there will be a relative risk of developing opportunistic attitudes and emergence of lack of fairness in the treatment of family and nonfamliy managers. VII.3. Limitations and directions for further research The first limitations of this study are related to Grounded Theory. The use of this methodology demands some caution, such as: the measures used are more subjective and they are more subject to the observation bias of the researcher (Suddaby, 2006); grounded theory does not analyse changes over time (Egan, 2002); a grounded theory approach has the effect of emphasizing the readily visible and observable in detriment to the commingling of the structural characteristics of social situations with activities (Layder, 1993). The use of Grounded Theory was combined with the development of multiple case studies seeking into each of the observed realities to allow inductively the emergence of the main elements. And a case studied, indeed, is an excellent alternative to be adopted, since it is grounded in on-going reality. Some limitations to the use of a case study are the fact that there is too much data for analysis and that 298 the complexity examined is difficult to represent simply (Hodkinson & Hodkinson, 2001). This is precisely minimized by the Grounded Theory methodology for distinguishing the main ideas in an orderly manner, and arriving at categories which will constitute the constructs of the model that is finally proposed. However, there is the danger that an inexperienced researcher can get lost in the various possibilities that this process offers. In the use of quantitative techniques, it is worth mentioning that I worked with a convenient sample, one in which the researcher selects the most accessible members of the population (60% of respondents are companies that are current or past clients of the FDC's database) representing a more established family business. This limits the generalization of the results, making the hypotheses tests to be evaluated with parsimony. Furthermore, the small sample size restricts the application of conventional methods of structural equation which could have been used for overall adjustment of the model, and its adhesion, to construct a more comprehensive theory. Further studies should then retrieve a larger and more representative sample (random), to assess the robustness of the hypotheses raised here. It must be stressed that four constructs presented KMO below the acceptable level of 0.600: Level of Independence and Efficiency of the Board, Trust, Legitimacy and Competence. The Kaiser-Meyer-Olkin (KMO) statistics is a measure of the quality of the factorial analysis conditions carried out, indicating whether the correlations between variables can be explained by other variables in the dataset (Mooi & Sarstedt, 2011). These low values found, however, represent a limitation on the quality of the factorial analysis. Nevertheless, given that the AVE and CR are in fact the measures of quality of the questions in themselves, and these being in accordance with the desired standards, I preferred to maintain these constructs. To obtain a better KMO, we should have a larger sample, a greater number of questions per variable, as well as the revision of those existing. It should also be noted that the moderate levels of reliability and validity of some measures, as well as the exclusion of questions and dimensions that could be relevant in future studies also suggest the possibility of refining, adapting and creating new measures for the constructs of interest and testing the model in new contexts and different samples. 299 We cannot overlook the possibility of some bias in the questionnaire responses, because the admission that conflict exists is a sensitive matter. At the pilot stage, I took care, after completion of the survey, to talk to six respondents, and one of the points discussed was how comfortable they felt filling out section (X) of the questionnaire. There was no evidence of discomfort. However, it should be noted that in the pilot test, I was with a group attending a training programme precisely on management in family businesses, which one would expect to demonstrate, a degree of maturity and openness on the part of respondents. We cannot, however, rule out the possibility that respondents to a questionnaire, would choose to minimize any conflicts, for the sake of appearances. I need to stress that the questionnaires were mostly completed by senior family managers and owners. Despite this representativeness, it would have been ideal to collect data from multiple sources. Understanding the perception of agency conflict of a nonfamliy manager would therefore be quite relevant for comparison with the vision of the owners. In fact the central debate in family businesses has concentrated on the nature and consequences of principal-agent behaviour when both actors belong to the same family (James et al., 2012). I, therefore, make recommendations for future research to consider the viewpoint of the nonfamliy manager, since most investigations of agency conflict presented in the literature are given from the viewpoint of the principal rather than the agent. Another point is that the survey is of a cross-sectional nature, thus limiting our ability to examine the dynamic interaction among the constructs studied. The use of a longitudinal approach would help to obtain better insight into what influences agency conflict in family businesses. This is an opportunity for future research. It would also be interesting to work with a meaningful basis of listed family businesses that would enable us to compare whether the regulatory elements of the capital market exert some influence on governance, and, consequently, on the agency conflicts in family businesses. A weak point which should be recognized in the model developed concerns, the importance of boards of directors for family businesses. It is a vital body of governance of any company, and according to the ANOVA test, it can be seen in the sample studied, that companies with boards of directors are less prone to agency conflict. However, after multivariate analysis, both variables related to the level of independence and efficiency of the board did not 300 remain in the model because the correlations among themselves and with the alignment construct showed no significance. It is suggested, therefore, that new researches should adopt more profound and consistent analysis of the role of boards of directors in mitigating agency conflict in family businesses. There are a number of questions that this research has not been able to address fully and which warrant further investigation:  Analyse only family-owned firms listed on the stock exchange, having boards of directors to see how the same variables included in the present study would behave in relation to agency conflict.  Since the majority of respondents in the present study are family members, it would be interesting to carry out new research exclusively with nonfamliy managers of family businesses to check their point of view, and examine which constructs would change in this model.  When the entry of a nonfamliy CEO in a family business generates more agency conflict: whether it is the 1st, 2nd , or 3rd generation which is in charge?  To examine the validity of the same model in enterprises with other forms of ownership and control: corporations, state enterprises, cooperatives? Which constructs would be retained? Which constructs would be kept? Which new ones should be added?  To investigate whether companies that are adopting corporate governance by way of evolution, seeking the benefits of the contribution from a broader view of experienced board members, promoting alignment, and developing trust and practicing fairness, are subject to less conflict than those who embrace governance only for its regulatory, procedural and surveillance nature? The fact that the used sample is focused on Brazilian firms limits the possibilities for the generalization of the results. Brazilian culture is strongly 301 relational, which clearly involves different behaviours and different influences on the constructs from that which should occur in more transaction-oriented cultures that exist in Anglo-Saxon countries. Also, the power distance index10 in Brazil is high, showing that many underlying conflicts cannot be captured in surveys of this nature, simply because managers more easily accept the command of owners and tend not to oppose their interests. Although for Mustakallio (2002, p.215) the ‘national cultures are likely not to exercise a significant influence on how different governance mechanisms function within a family firm’, applications in other countries and cultures are encouraged in order to verify and contrast the differences found here. Finally, despite the limitations just described, it is believed that this study offers an advance on previous research in the field of the corporate governance of family firms. It contributes some insights and provides empirical support for, or comment on, theoretical propositions present in the literature. It is hoped that the theoretical framework proposed and validated in this research can offer a useful basis for future studies. 10 Hofstede’s Power distance Index measures the extent to which the less powerful members of organizations and institutions (like the family) accept and expect that power is distributed unequally. It suggests that a society’s level of inequality is endorsed by the followers as much as by the leaders. 302 REFERENCES Alberti, V. (1990). História oral: a experiência do CPDOC (Oral history: the experience of CPDOC). Rio de Janeiro: Fundação Getúlio Vargas. Alchian, A. and Woodward, S. (1988). The Firm is Dead; Long Live the Firm: a review of Oliver E. Williamson's "The Economic Institutions of Capitalism". Journal of Economic Literature, 26: 65-79. Alvares, E., Giacometti, C. and Gusso, E. (2008). Governança corporativa: um modelo brasileiro (Corporate governance: a Brazilian model). Rio de Janeiro: Elsevier. Amason, A. C. (1996). Distinguishing the effects of functional and dysfunctional conflict on strategic decision making: resolving a paradox for top management teams. Academy of Management Journal, 39(1): 123-148. Amba-Rao, S. C., Petrick, J. A., Gupta, J. N. D. and Von Der Embse, T. J. (2000). Comparative performance appraisal practices and management values among foreign and domestic firms in India. The International Journal of Human Resource Management, 11(1): 60-89. Anderson, A. R., Jack, S. L. and Dodd, S. D. (2005). The Role of Family Members In Entrepreneurial Networks: Beyond the Boundaries of the Family Firm. Family Business Review, 18: 135–154. Anderson, R. C. and Reeb, D.M. (2003). Founding-family ownership and firm performance: evidence from the S&P 500. Journal of Finance, 58: 1301-1328. Andrade, A. and Rossetti, J. P. (2006). Adesão à boa governança: uma irrecusável tendência (Subscribing to good governance adhesion: an irresistible trend). Unpublished paper. Corporate Governance Program. Fundação Dom Cabral. Andrade, A. and Rossetti, J. P. (2011). Governança Corporativa - fundamentos, desenvolvimento e tendências (Corporate Governance: principles, development and trends). 5th ed. São Paulo: Atlas. Andres, C. (2006). Family Ownership as the Optimal Organizational Structure? EFA 2006. Meeting Paper, available in SSRN, Pages 25. Andrews, K. R. (1991). The concept of corporate strategy. In: Mintzberg, H. The strategy process: concepts, contexts and cases. 2nd ed. Englewood Cliffs, NJ; Prentice-Hall. Ang, J. S., Cole, R. A. and Lin, J. W. (2000). Agency costs and ownership structure. The Journal of Finance, 55(1): 81-106. 303 Aronoff, C. E. and Ward, J. L. (1994). Set policies to solve future problems. Nation’s Business, 82(7): 70–71. Aronoff, C. E. and Ward, J. L. (1996). Family business governance: maximizing family and business potential. Georgia: Family Enterprise Publisher. Arregle, J.-L., Hitt, M. A., Sirmon, D. G. and Very, P. (2007). The Development of Organizational Social Capital: Attributes of Family-owned firms. Journal of Management Studies, 44(1): 73-95. Arruñada, B. (2011). Mandatory accounting disclosure by small private companies, European Journal of Law and Economics, 32(3): 377-413. Assis, L. O. M. and Neto, M. T. R. (2011). Remuneração Variável por Desempenho no Setor Público: investigação das causas do fracasso e implicações para o estado brasileiro (Variable Pay for Performance in the Public Sector: Research the causes of failure and implications for the Brazilian state). Forum of International Research Society for Public Management. Astrachan, J. H. (1988). Family Firm and Community Culture. Family Business Review, 1(2): 165-189. Astrachan, J.H. and Kolenko, T. (1994). A neglected factor explaining family business success: human resource practices. Family Business Review, 7: 251– 262. Astrachan, J. H. and McMillan, K. S. (2003). Conflict and communication in the family business. Marietta, GA: Family Enterprise Publishers. Astrachan, J. H. and Jaskiewicz, P. (2008). Emotional Returns and Emotional Costs in Privately Held Family Businesses: Advancing Traditional Business Valuation. Family Business Review, 21(2): 139-149. Ayuso, S., Rodríguez, M. A., García-Castro, R. and Ariño, M. A. (2007). Maximizing Stakeholders’ Interests: An Empirical Analysis of the Stakeholder Approach to Corporate Governance. IESE Research Papers D/670, IESE Business School. Babic, C. V. (2003). Corporate governance problems in transition economies. Winston-Salem: Wake Forest University, Social Science Research Seminar. Bacellar, F. (2005). Contribuições para o ensino de marketing: revelando e compreendendo a perspectiva dos professores (Contributions to the teaching of marketing: revealing and understanding the perspective of teachers). PhD thesis, Faculdade de Economia, Administração e Contabilidade, Universidade de São Paulo. Bagozzi, R. P. (1994). Measuring in marketing research; basic principles of questionnaire design’ in Bagozzi, R .P., ed., Principles of Marketing Research, Cambridge MA: Blackwell Publishers. 304 Bagozzi, R. P., Yi, Y. and Phillips, L. W. (1991). Assessing construct validity in organizational research. Administrative Science Quarterly, 36(3): 421-458. Bailyn, L. (1993). Breaking the mold: Men, Women and Time in the New Corporate World. New York: The Free Press. Baldridge, D. C. and Schulze, W. S. (1999). Fairness in family-owned firms: An organizational justice perspective on agency problems. Academy of Management Best Paper Proceedings. Academy of Management, Briar Cliff Manor, NY. Bammens, Y., Vooedeckers, W. and Van Gils, A. (2009). The impact of family cohesion on behavioral board processes: an examination of family firm heterogeneity. In: 9th EURAM Conference, Liverpool. Barach, J. A., Gantisky, J. B., Carson, J. A. and Doochin, A. (1988). Entry of the next generation: Strategic challenge for family business. Journal of Small Business Management, 26(2): 49–56. Barnes, L. B. and Hershon, S. A. (1994). Transferring Power in the Family Business. Family Business Review, 7: 377–392. Barnett, T. and Kellermanns, F. W. (2006). Are we family and are we treated as family? Nonfamliy employees’ perceptions of justice in the family firm. Entrepreneurship Theory and Practice, 30(6): 837-854. Barrett, G.V. and Kernan, M. C. (1987). Performance appraisal and terminations: a review of court decisions since Brito v. Zia with implications for personnel practices. Personnel Psychology, 40(3): 489-503. Bartholomeusz, S. and Tanewski, G. A. (2006). The Relationship between Familyowned firms and Corporate Governance. Journal of Small Business Management, 44 (2): 245-267. Basco, R. and Rodríguez, M. J. P. (2009) Studying the Family Enterprise Holistically: Evidence for Integrated Family and Business Systems. Family Business Review, 22(1): 82-95. Basu, A. (2004). Entrepreneurial aspirations among family business owners: An analysis of ethnic business owners in the UK. International Journal of Entrepreneurial Behaviour & Research, 10(1/2): 12-33. Bathala, C. T. and Rao, R. P. (1995). The Determinants of Board Composition: An Agency Theory Perspective. Managerial and Decision Economics, 16: 59-69. Batson, C. D. (1990). How social an animal? The human capacity for caring. American Psychologist, 45: 336-346. Baysinger, B. D. and Butler, H. N. (1985). Corporate Governance and the Board of Directors: Performance Effects of Changes in Board Composition. Journal of Law, Economics, and Organization, 1: 101-124. 305 Baysinger, B. D. and Hoskisson, R.E. (1990). The composition of boards of directors and strategic control: effects on corporate strategy. Academy of Management Review, 15(1): 72-87. Bebchuk, L., Cohen, A. and Ferrell, A. (2009). What matters in corporate governance? Review of Financial Economics, 22: 783-827. Beccerra, M. and Gupta, A. K. (1999). Trust within the Organization: Integrating the Trust Literature with Agency Theory and Transaction Costs Economics. Public Administration Quarterly, 23(2): 177-203. Becker, G. S. (1981). Altruism in the Family and Selfishness in the Market Place. Economica, New Series, 48(189): 1-15. Beckhard, R., and Dyer, W. G., Jr. (1983). Managing Continuity in the Family-Owned Business. Organizational Dynamics, 12(1): 5-12. Berenbeim, R. (1990). How business families manage the transition from owner to professional management. Family Business Review, 3(1): 69-110. Berg, B. L. (2001). Qualitative Research Methods for the Social Sciences. Boston: Allyn and Bacon. Bergstrom, T. C. (1989). A Fresh Look at the Rotten Kid Theorem – and Other Household Mysteries. Journal of Political Economy, 97(5): 1138-1159. Berhold, M. (1971). A Theory of Linear Profit-Sharing Incentives. The Quarterly Journal of Economics, 85(3): 460-482. Bernhoeft, R. (1991). Empresa Familiar: sucessão profissionalizada ou sobrevivência comprometida (Family Business: professional succession or impaired survival). 2nd ed. São Paulo: Nobel. Bernhoeft, R. (1996). Como criar, manter e sair de uma sociedade familiar - sem brigar (How to create, maintain and exit a family company - without fighting). São Paulo: SENAC. Bernhoeft, R. and Castanheira, J. (1995). Manual de sobrevivência para sócios e herdeiros (Survival Guide for members and heirs). São Paulo: Nobel. Bernhoeft, R. and Gallo. M. (2003). Governança na Empresa Familiar (Governance in Family Business). Rio de Janeiro: Campus. Bernhoeft, R. (2004). Carta a um jovem herdeiro (Letter to a young heir). Rio de Janeiro: Alegro. Berle, A. and Means, G. (1932). The modern corporation and private property. New York: Harcourt, Brace & World. 306 Bhagat, S., Bolton, B. and Romano, R. (2008). The promise and peril of corporate governance indices. Columbia Law Review, 108: 1803-1882. Bigliardi, B. and Dormio, A. I. (2009). Successful generational change in family business. Measuring Business Excellence, 13(2): 44–50. Blair, M. M. (1995). Ownership and control: rethinking corporate governance for the twenty-first century. Washington: Brookings Institution. Bollen, K. A. (1989). Structural equations with latent variables. New York: Wiley. Bornholdt, W. (2005). Governança na empresa familiar: implementação e prática (Governance in the family owned company: implementation and practice). Porto Alegre: Bookman. Boyd, B. (1990). Corporate Linkages and Organizational Environment: A Test of the Resource Dependence Model. Strategic Management Journal, 11: 419–430. Boyd, K. B. (1991). Strategic planning and financial performance: a meta-analytic review. Journal of Management Studies, 28: 353-374. Brickley J. A. and James, C. M. (1987). The takeover market, corporate board composition, and ownership structure: The case of banking. Journal of Law and Economics, 10(1): 161-180. Bowman-Upton, N. (1991). Transferring Management in the Family-Owned Business. Texas: SBA – Small Business Administration. [online] th <http://www.sba.gov/library/pubs/eb-1.doc> [Accessed December 4 2011]. Buchanan, J. M. (1975). The Limits of Liberty: Between Anarchy and Leviathan. Chicago: University of Chicago. Cabrera-Suárez, K., Saa-Perez, P. D. and Almeida, D. G. (2001). The Succession Process from a Resource and Knowledge-Based View of the Family Firm. Family Business Review, 14(1): 37-47. Cadbury, Sir A. 2000. Family Firms and their Governance: Creating Tomorrow's. Company from Today's. Zurich: Egon Zehnder International. Carlock, R. S. and Ward, J. L. (2010). When Family Businesses Are Best: The Parallel Planning Process for Family Harmony and Business Success. Hampshire, UK: Palgrave Macmillan. Carney, M. (2005). Corporate governance and competitive advantage in family controlled firms. Entrepreneurship: Theory & Practice, 29: 249–265. Carroll, G. (1984). Dynamics of publisher succession in newspaper organizations. Administrative Science Quarterly, 29: 93-113. 307 Cassiani, S. de B., Caliri, M. H. L. and Pelá, N. T. R. (1996). A teoria fundamentada nos dados como abordagem da pesquisa interpretativa (The grounded theory approach to the interpretive research). Revista Latino-Americana de Enfermagem, 4(3): 75-88. Chami, R. and Fullenkamp, C. (2002). Trust and efficiency. Journal of Banking and Finance, 26(9): 1785–1809. Charmaz, C. (2000). Grounded Theory: Objectivist and Constructivist Methods. In Denzin, N. and Lincoln, Y. (Eds.) Handbook of Qualitative Research. 2nd ed. London: Sage. Chen, S., Chen, X. and Cheng, Q. (2008). Do Family-owned firms Provide More or Less Voluntary Disclosure? Journal of Accounting Research, 46: 499–536. Chenitz, W. C. and Swanson, J. M. (Eds.) (1986). From practice to grounded theory: Qualitative research in nursing. Menlo Park, CA: Addison-Wesley. Child, J. and Rodrigues, S. B. (2003). Corporate Governance and New Organizational Forms: Issues of Double and Multiple Agency. Journal of Management and Governance, 7: 337-360. Child J. and Rodrigues, S. B, (2004). Repairing the breach of trust in corporate governance. Corporate Governance: An International Review, 12(2): 143-152. Chin, W. W. (1998). Issues and Opinion on Structure Equation Modeling. MIS Quarterly, 22: 7-16. Chittoor, R. and Das, R. (2007). Professionalization of Management and succession Performance - A Vital Linkage. Family Business Review, 20(1): 65-79. Chrisman J. J., Chua, J. H. and Sharma, P. (1997). Strategic management of the family business: Past research and future challenges. Family Business Review, 10(1): 1-35. Chrisman J. J., Chua, J. H. and Sharma, P. (1999). Defining the family business by behavior. Entrepreneurship Theory and Practice, 23(4): 19-39. Chrisman, J. J., Chua, J. H. and Litz, R. (2004). Comparing the agency costs of family and nonfamliy-owned firms: Conceptual issues and exploratory evidence. Entrepreneurship Theory and Practice, 28(4): 335–354. Chrisman, J. J., Chua, J. H. and Sharma, P. (2005). Trends and directions in the development of a strategic management theory of the family firm. Entrepreneurship Theory and Practice, 29(5): 555-575. Chrisman, J. J., Chua, J. H. and Steier, L. (2003). An introduction to theories of family business. Journal of Business Venturing, 18: 441–448. 308 Chrisman, J. J., Chua, J. H. and Steier, L. (2008). Toward a theoretical basis for understanding the dynamics of strategic performance in family-owned firms. Entrepreneurship Theory and Practice, 32: 935–947. Chua, J. H,, Chrisman, J. J. and Bergiel, E. B. (2009). An agency theoretic analysis of the professionalized family firm. Entrepreneurship Theory and Practice, 33: 355–372. Churchill, N. C. and Hatten, K. J. (1987). Non-market-based transfer of wealth and power: A research framework for family businesses. American Journal of Small Business, 11(3): 51-64. Churchill, G. and Iacobucci, D (2002). Marketing research: Methodological foundations. 8th ed. Orlando: Harcourt College Publishers. Churchill, N. and Lewis, V. (1983). The five stages of small business growth. Harvard Business Review, 61(3): 30-50. Clarkson, M.B.E. (1995). A stakeholder framework for analyzing and evaluating corporate social performance. Academy of Management Review, 20: 65-91. Coffee, J. (1999). The future as history: The prospects for global convergence in corporate governance and its implications. Northwestern University Law Review, 93: 631–707. Cohen, J., Cohen, P., Leona S. A. and West, S. H. (2003). Applied multiple regression/correlation analysis for the behavioral sciences. Hillsdale, N.J: L. Erlbaum Associates. Coleman, J. W. (1987). Toward an integrated theory of white-collar crime. American Journal of Sociology, 93(2): 406-439. Colquitt, J. A. and Zapata-Phelan, C. P. (2007). Trends in theory building and theory testing: a five-decade study of the Academy of Management Journal. Academy of Management Journal, 50(6): 1281–1303 Cook, K. S. (2001). Trust in Society. New York: Russell Sage Foundation. Cooper, D. R. and Schindler, P. S. (2011). Business research methods. New York, McGraw-Hill Higher Education. Corbetta, G. and Salvato, C. (2004). Self-serving or self-actualizing? Models of man and agency costs in different types of family-owned firms: a commentary on ‘Comparing the agency costs of family and nonfamliy-owned firms: conceptual issues and exploratory evidence’. Entrepreneurship: Theory and Practice, 28(4): 355362. Corbetta, G. and Tomaselli, S. (1996). Boards of directors in Italian family businesses. Family Business Review, 9(4): 403-421. 309 Cotter, J. and Silvester, M. (2003). Board and Monitoring Committee Independence. Abacus, 39: 211–232. Cowling, M., Howorth, C. A. and Westhead, P. (2001). The development of familyowned companies. Family Business Review, 14(4): 369-385. Craig, J. B. and Salvato, C. (2012). The Distinctiveness, Design, and Direction of Family Business Research: Insights From Management Luminaries. Family Business Review, 25(1) 109-116. Cropanzano, R. (1993). Justice in the Workplace: Approaching Fairness in Human Resource Management. New Jersey: Lawrence Erlbaum Associates. Daily, C. M. and Dalton, D. R. (1994). Bankruptcy and corporate governance: The impact of board composition and structure. Academy of Management Journal, 37(6): 1603-1617. Daily, C. M., Dalton, D. R. and Cannella, A. A. Jr (2003). Corporate governance: decades of dialogue and data. Academy of Management Review, 28(3): 371-382. Daily, C. M. and Dollinger, M. J. (1992). An empirical examination of ownership structure in family and professionally-managed firms. Family Business Review, 5: 117-136. Daily, C. M. and Dollinger, M. J. (1993). Alternative methodologies for identifying family vs non family managed businesses. Journal of Small Business Management, 31: 70-90. Daily, C. M. and Schwenk, C. (1996). Chief executive officers, top management teams, and boards of directors: Congruent or countervailing forces? Journal of Management, 22: 185-208. Dalton, D. R., Hitt, M. A., Certo, S. T. and Dalton, C. M. (2008). The fundamental agency problem and its mitigation: Independence, equity, and the market for corporate control. In J. P. Walsh & A. P. Brief (Eds.), Annals of the Academy of Management, 1: 1-64. New York: Lawrence Erlbaum. Danes, S. M., Zuiker, V., Kean, R., Arbuthnot, J. and Kaye, K. (1999). Predictors of family business tension and goal achievement. Family Business Review, 12(3): 241–252. Das, T. K. and Teng, B. S. (2001). Trust, Control and Risk in Strategic Alliances: An Integrated Framework. Organization Studies, 22(2): 251–283. Davel, E. and Colbari, A. (2003). Pesquisas sobre organizações familiares: instigações e contribuições (Research on family-owned firms: inducements and contributions). II Colóquio Internacional – Análise de Organizações: Perspectivas Latinas, vol. PDGS, EA/UFBA. 310 Davis, J. A. and Herrera, R. M. (1998). The Social Psychology of Family Shareholder Dynamics. Family Business Review, 11: 253–260. Davis, J. A., Pitts, E. L. and Cormier, K. (2000). Challenges facing family-owned companies in the Gulf region. Family Business Review, 13: 217–237. Davis, J. H. (2003). Radiografia da relação empresa, família e propriedade (Radiography of the firm, family and property). HSM Management, 41: 96-103. Davis, J. H., Schoorman, F. D. and Donaldson, L. (1997). Toward a Stewardship Theory of Management. Academy of Management Review, 22: 20–47. Davis, P. (1983). Realizing the potential of the family business. Organizational Dynamics, 12(3): 47-56. Davis, P. S. and Harveston, P. D. (1998). The influence of family business succession process: A multi-generational perspective. Entrepreneurship Theory and Practice, 22(3): 31-53. Davis, P. S. and Harveston, P. D. (1999). In the Founder’s Shadow: Conflict in the Family Firm. Family Business Review, 12(4): 311-323. Davis, P. S. and Harveston, P.D. (2001). The phenomenon of substantive conflict in the family firm: a cross-generational study. Journal of Small Business Management, 39(1): 14-30. Davis, P. S. and Stern, D. (1988). Adaptation, Survival, and Growth of the Family Business. Family Business Review, 1(1): 69-85. Debicki B. J., Matherne C. F., Kellermanns, F. W. and Chrisman, J. J. (2009). Family business research in the new millennium. Family Business Review, 22: 151–166. Demsetz, H. (1983). The structure of ownership and the theory of the firm. Journal of Law and Economics, 26: 375-390. De Vellis, R. F. (2003). Scale Development: Theory and Practice, Applied Social Research Methods Series, 26, London: Sage Publications. Distelberg, B. and Sorenson R. L. (2009). Updating Systems Concepts in Family Businesses: A Focus on Values, Resource Flows, and Adaptability. Family Business Review, 22(1): 65-81. Donaldson, L. (1990). The Ethereal Hand: Organizational Economics and Management Theory. Academic Management Review, 15(3): 369-381. Donaldson, L. and Davis, J. H. (1991). Stewardship theory or agency theory: CEO governance and shareholder returns. Australian Journal of Management, 16: 4964. 311 Donaldson, T. and Preston, L. (1995). A Stakeholder Theory of the Corporation: Concepts, Evidence, and Implications. Academy of Management Review, 20(1): 65-91. Donckels, R., and Fröhlich, E. (1991). Are family businesses really different? European experiences from STRATOS. Family Business Review, 4(2): 149-160. Donnelley, R. G. (1988). The Family Business. Family Business Review, 1(4): 427445. Dooley, R. S. and Fryxell, G. E. (1999). Attaining decision quality and commitment from dissent: The moderating effects of loyalty and competence in strategic decisionmaking teams. Academy of Management Journal, 42(4): 389-402. Drucker, P. (1977). Introdução à administração (Introduction to management). São Paulo: Pioneira. Dunn, B. (1996). Family enterprises in the UK–a special sector? Family Business Review, 9: 139–155. Dupas, G., Lank, A. G., Aires, A., Vries, M. K. and Costa, R. T. (2003). Governando a Empresa Familiar (Governing the Family Business). Álvares, E. (coord.). Rio de Janeiro: Qualitymark. Dyer, J. H. (1997). Effective interfirm collaboration: How firms minimize transaction costs and maximize transaction value. Strategic Management Journal, 18(7): 535– 556. Dyer, J. H. and Singh, H. (1998). The relational view: Cooperative strategy and sources of inter-organizational competitive strategy. Academy of Management Review, 23(4): 660-679. Dyer, W. G., Jr. (1986). Cultural Change in Family-owned firms: Anticipating and Managing Business and Family Transitions. San Francisco: Jossey-Bass. Dyer, W. G., Jr. (1988). Culture and Continuity in Family-owned firms. Family Business Review, 1: 37–50. Dyer, W. G., Jr. (1989). Integrating professional management into a family owned business, Family Business Review, 2(3): 221–235. Dyer, W. G., Jr. (1992). The Entrepreneurial experience: Confronting career dilemmas of the start-up executive. San Francisco, CA: Jossey-Bass. Dyer, W.G., Jr. (1994). Potential contributions of organizational behavior to the study of family-owned businesses. Family Business Review, 7(2): 109-131. Dyer, W. G. Jr (2006). Examining the ‘Family Effect’ on Firm Performance. Family Business Review, 19: 253-273. 312 Dyer, W. G. Jr. and Handler, W. (1994). Entrepreneurship and family business: exploring the connections. Entrepreneurship: Theory and Practice, 19(1): 71-83. Eberlin, R. and Tatum, B. C. (2005). Organizational justice and decision making: When good intentions are not enough. Management Decision, 43(7/8): 1040-1048. Eccles, R. and Mavrinac, S. (1995). Improving the Corporate Disclosure Process. Sloan Management Review, 36(4): 11-25. Eddleston, K. A., Kellermans, F. W. and Sarathy, R. (2008). Resource configuration in family-owned firms: Linking resources, strategic planning and technological opportunities to performance. Journal of Management Studies, 45(1): 26-50. Egan, T. M. (2002). Building theory in HRD from a grounded theory approach. Advances in Developing Human Resources, 4(3): 277-295. Eisenhardt, K. M. (1989a). Agency theory: An assessment and review. Academy of Management Review, 14(1): 57-74. Eisenhardt, K. M. (1989b). Building theories from case study research. Academy of Management Review, 14(4): 532-550. Eisenhardt, K. M. (1991). Better stories and better constructs: The case for rigor and comparative logic. Academy of Management Review, 16: 620–627. Elbanna, S. (2009). The impact of affective conflict on firm performance. Management Research News, 32(9): 789-803. Ensley, M. D. and Pearson, A. W. (2005). An exploratory comparison of the behavioral dynamics of top management teams in family and nonfamliy new ventures cohesion, conflict, potency and consensus. Entrepreneurship Theory and Practice, 29(3): 267-284. Epstein, M. J. and Roy, M. (2005). Evaluating and monitoring CEO performance: evidence from US compensation committee reports. Corporate Governance, 5(4): 75-88. Eshel I., Samuelson, L. and Shaked A. (1998). Altruists, Egoists, and Hooligans in a Local Interaction Model. The American Economic Review, 88(1): 157-179. Fahed-Sreih, J. and Djoundourian, S. (2006) Determinants of Longevity and Success in Lebanese Family Businesses: An Exploratory Study. Family Business Review, 19(3): 225-234. Falk, A. and Kosfeld. M. (2004). Distrust - the hidden cost of control. IZA. Working Paper 1203. Fama, E. F. and Jensen, M. C. (1983a). Separation of ownership and control. Journal of Law and Economics, 26(2): 301-325. 313 Fama, E. F. and Jensen, M. C. (1983b). Agency Problems and Residual Claims. Journal of Law and Economics, 26(2): 327-349. Flemons, D. G. and Cole, P. M. (1992). Connecting and Separating Family and Business: A Relational Approach to Consultation. Family Business Review, 5(3): 257–269. Fletcher, D. E. (2002). Understanding the Small Family Businesses. London and New York: Routledge. Fiedler, K., Semin, G. R., Finkenauer, C. and Berkel, I. (1995). Actor-observer bias in close relationships: The role of self-knowledge and self-related language. Personality and Social Psychology Bulletin, 21: 525-538. Filbeck, G. and Lee, S. (2000). Financial Management Techniques in Family Businesses. Family Business Review, 13: 201–216. Finkelstein, S. and D. Hambrick (1996). Strategic Leadership: Top Executives and Their Effects on Organizations. Minneapolis–St. Paul: West. Fogel, E. M. and Geier, A. M. (2007). Strangers in the house: Rethinking SarbanesOxley and the independent board of directors. Delaware Journal of Corporate Law, 32: 33-72. Forbes, D. P. and Milliken, F. J. (1999). Cognition and corporate governance: Understanding boards of directors as strategic decision-making groups. Academy of Management Review, 24(3): 489-505. Ford, R. H. (1988). Outside directors and the privately-owned firm: Are they necessary? Entrepreneurship Theory and Practice, 13(1): 49–57. Fornell, C. and Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 48: 39–50. Foster, A. (1995). Developing leadership in the successor generation. Family Business Review, 8(4): 201-209. Frank, H., Kessler, A., Nosé, L. and Suchy, D. (2011). Conflicts in family-owned firms: state of the art and perspectives for future research. Journal of Family Business Management, 1(2): 130-153. Freeman, R. E. (1984). Strategic Management: A Stakeholder Approach. Boston: Pitnam. Friedman, M. and Friedman, S. (1994). How to run a family business. Cincinnati, OH: Betterway Books. Freyre, G. (2003). Casa Grande e Senzala (The Masters and the Slaves). 22nd ed. Rio de Janeiro: Livraria José Olympio Editora. 314 Furuta, F. and Santos, A. (2010). Comitê de Auditoria versus Conselho Fiscal Adaptado: a visão dos analistas de mercado e dos executivos das empresas que possuem ADRs. Revista Contabilidade & Finanças, 21(53): 1-23. Gallo, M. A. (1993). Organos de Gobierno de La Empresa Familiar (Governing Bodies of Family Business). Nota técnica de la División de Investigación del IESE. Gallo, M. A. (1998). La sucesión en la empresa familiar (The succession in family business). Barcelona: La Caixa. Gallo, M. A. and Sveen, J. (1991). Internationalizing the family business: Facilitating and restraining factors. Family Business Review, 4(2), 181-190. García-Castro, R., Ariño, M. A. and Canela, M. A. (2011) Over the Long-Run? ShortRun Impact and Long-Run Consequences of Stakeholder Management. Business & Society, 50(3): 428-455. Ge W. and McVay, S. (2005). The Disclosure of Material Weaknesses in Internal Control after the Sarbanes-Oxley Act. Accounting Horizons, 19(3): 137-158. George, W. G. (2002). Restoring governance to our corporations: crisis in the corporate world. Vital Speeches of the Day, 8(24): 791-676. Gersick, K. E., Davis, J. A., McCollom Hampton, M. M., and Lansberg, I. (1997). Generation to generation: Life cycles of the family business. Boston: Harvard Business School Press. Gil, A. C. (1999). Métodos e técnicas de pesquisa social (Methods and techniques of social research). São Paulo: Atlas. Gill, J. and Johnson, P. (1997). Research Methods for Managers. 2nd ed. London: Paul Chapman. Ghoshal, S. and Moran, P. (1996). Bad for practice: A critique of the transaction cost theory. Academy of Management Review, 21(1): 13-47. Glaser, B. G. and Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine Publishing Company. Goldberg, S. D. and Wooldridge, B. (1993). Self-confidence and managerial autonomy: Successor characteristics critical to succession in family-owned firms. Family Business Review, 6(1): 55-73. Goldenberg, M. (2002). A arte de pesquisar: como fazer pesquisa qualitativa em ciências sociais (The art of research: how to do qualitative research in social sciences). Rio de Janeiro: Record. 315 Gomez-Mejia, L., Larraza-Kintana, M. and Makri, M. (2003). The determinants of executive compensation in family controlled public corporations. Academy of Management Journal, 46: 226–237. Gomez-Mejia, L., Nunez-Nickel, M. and Gutierrez, I. (2001). The role of family ties in agency contracts. Academy of Management Journal, 44: 81-95. Gomez-Mejia, L. R, Welbourne, T. and Wiseman, R. M. (2000). The role of risk sharing and risk taking under gainsharing. Academy of Management Review, 25(3): 492-507. Gonçalves, S. C. (2000a). Patrimônio, Família e Empresa – Um estudo sobre as transformações no mundo da economia empresarial (Patrimony, Family and Company - A study of the transformations in the world of business economics) 1 st ed. São Paulo: Negócio Editora. Gonçalves, S. C. (2000b). As Empresas Familiares no Brasil (Family Business in Brazil). Revista de Administração de Empresas, RAE Light, 7(1): 7-12. Gordon, J. N. (2007). The rise of independent directors in the United States, 19502005: Of shareholder value and stock market prices. Standard Law Review, 59: 1465-1568. Graham, J. M. (2006). Congeneric and (Essentially) Tau-Equivalent Estimates of Score Reliability. What They Are and How to Use Them. Educational and Psychological Measurement, 66(6): 930-944. Granata, D. and Chirico, F. (2010). Measures of Value in Acquisitions: Family versus Nonfamliy-owned firms. Family Business Review, 23(4): 341-354. Granovetter, M. (1985). Economic action and social structure: the problem of embeddedness. American Journal of Sociology, 91(3): 481-510. Grant, R. M. (2003). Strategic Planning in a Turbulent Environment: Evidence from the Oil Majors. Strategic Management Journal, 24(6): 491-517. Gravetter, F. J. and Wallnau, L. B. (2004). Statistics for the behavioral sciences. 6th ed. Belmont, CA: Wadsworth. Greenberg, J. (1990). Organizational justice: Yesterday, today and tomorrow. Journal of Management, 16: 399-432. Gregory, H. and Simmelkjaer, R. T. (2002). Comparative study of corporate governance codes relevant to the European Union and its member states on behalf of the European Commission. New York: Weil, Gotshal & Manges, LLP. Grzybovski, D. (2007). Plataforma de conhecimentos e aprendizagem transgeracional em empresas familiares (Platform of knowledge and transgenerational learning in family businesses). Unpublished doctoral thesis, Universidade Federal de Lavras. 316 Guion, R. M. (1965). Personnel testing. New York: McGraw-Hill. Haberman, H. and Danes, S. M. (2007) Father-Daughter and Father-Son Family Business Management Transfer Comparison: Family FIRO Model Application. Family Business Review, 20(2): 163-184. Hackman, J. R. and Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 15: 250-279. Haenlein, M. and Kaplan, A.M. (2004). A beginner's guide to partial least squares (PLS) analysis. Understanding statistics, 3(4), 283-297. Hair Jr, J. F., Black, W. C., Babin, B. J., and Anderson, R. E. (2010). Multivariate data analysis. 7th ed. Upper Saddle River, NJ: Prentice Hall. Hall, A. and Nordqvist, M. (2008). Professional management in family businesses: toward an extended understanding. Family Business Review, 21(1): 51-69. Handler, W. C. (1989a). Managing the family succession process: The next generation family member’s experience. Unpublished doctoral dissertation, School of Management, Boston University, USA. Handler, W. C. (1989b). Methodological issues and considerations in studying Family Businesses. Family Business Review, 2(3): 257-276. Handler, W. C. (1990). Succession in family-owned firms: A mutual role adjustment between entrepreneur and next-generation family members. Entrepreneurship Theory and Practice, 15(1): 37-51. Harris, R. I. D., Reid, R. S. and McAdam, R. (2004). Employee involvement in family and nonfamliy owned businesses in Great Britain. International Journal of Entrepreneurial Behaviour & Research, 10(1/2): 49-58. Harvey, J. S. Jr. (1999a). What can the family contribute to business? Examining contractual relationships. Family Business Review, 12(1): 61-71. Harvey, J. S. Jr. (1999b). Owner as manager: extended horizons and the family firm. International Journal of the Economics of Business, 6(1): 41-55. Harvey M. and Evans R. (1994). Family Business and Multiple Levels of Conflict. Family Business Review, 7(4); 331-348. Harvey M. and Evans R. (1995). Life After Succession in the Family Business: Is It Really the End of Problems? Family Business Review, 8(3): 3-15. Haugh H.M. and Mckee L. (2003). It’s just like a family - shared values in the family firm. Community, Work & Family, 6(2): 141-158. 317 Heidrick, G. W. (1998). Selecting Outside Directors. Family Business Review, 1: 271-277. Hendry, J. (2002). The principal's other problems: Honest incompetence and the specification of objectives. Academy of Management Review, 27: 98-113. Hermalin, B. E. and Weisbach, M. S. (2003). Boards of Directors as an Endogenously Determined Institution: A Survey of the Economic Literature, Economic Policy Review. Federal Reserve Bank of New York. 9: 7–26. Herrero, I. (2011). Agency Costs, Family Ties, and Firm Efficiency. Journal of Management, 37(3): 887-904. Hill, C. W. L and Snell, S. A. (1989). Effects of ownership structure and control on corporate productivity. Academy of Management Journal, 32(1): 25-46. Hodkinson, P. and Hodkinson, H. (2001). The Strengths and Limitations of Case Study Research. Paper presented to the Learning and Skills Development Agency conference. [online]. http://education.exeter.ac.uk/tlc/publications.htm [Accessed May 22nd 2012]. Holland, J. (1998). Private voluntary disclosure, financial intermediation and market efficiency. Journal of Business Finance and Accounting, 25(1): 29-68. Hollander, B. S. and Elman, N. S. (1988). Family-owned businesses: An emerging field of inquiry. Family Business Review, 1(2): 145-164. Hood, C. (2006). Gaming in the target world: the targets approach to managing British public services. Public Administration Review, 66(4): 515-521. Howorth, C., Rose, M., Hamilton, E. and Westhead, P. (2010). Family firm diversity and development: an introduction. International Small Business Journal, 28: 437– 451. Howorth, C., Westhead, P. and Wright, M. (2004). Information asymmetry in management buyouts of family-owned firms. Journal of Business Venturing, 19(4): 509–534. Hoy, F. and Verser, G. (1994). Emerging business, emerging field: Entrepreneurship and the family firm. Entrepreneurship Theory and Practice, 19(1): 9-23. HSM Management (2003). Retrato de família (Family Portrait). HSM Management, 6(41): 84-91. Hulland, J. (1999). Use or Partial Least Squares (PLS) in strategic management research: a review of four recent studies. Strategic Management Journal, 20(2): 195-204. Hunt, S. D. (2002). Foundations of Marketing Theory: Toward a General Theory of Marketing. Armonk, NY: M.E. Sharpe, Inc. 318 Hunter, I. (2007). Entrepreneurship and pure altruism: Reconciling the irreconcilable? Evidence from a New Zealand mail-order firm. Management & Organizational History, 2(1): 63-82. Hunter, J. E. and Schmidt, J. L. (1996). Cumulative research knowledge and social policy formulation: The critical role of meta-analysis. Psychology, Public Policy, and Law, 2: 324-347. Huysamen, G. K. (2006). Coefficient alpha: Unnecessarily ambiguous; unduly ubiquitous. South-African Journal of Industrial Psychology, 32(4): 34-40. Ibrahim, A. B., McGuire, J., Soufani, K. and Poutziouris, P. (2004). Patterns in Strategy Formation in a Family Firm. International Journal of Entrepreneurial Behaviour & Research, 10(1/2): 127-140. Ibrahim, A. B., Soufani, K. and Lam, J. (2001). A study of succession in a family firm. Family Business Review, 14 (3): 245–258. IBGC (2004). Código das melhores práticas de governança corporativa (Code of corporate governance best practices). 3rd ed., São Paulo: IBGC. IBGC (2009). O que são empresas familiares? (What are family businesses?) [online]. www.ibgc.org.br/PerguntasFrequentes.aspx [Accessed August 25th 2009]. IBRI (2007). Guia IBRI-BOVESPA de relações com investidores (IBRI-BOVESPA Guide to Investor Relations) [online] http://www.ibri.com.br/2C83497F-CA16-4F6583D3-3B6105B66ECC/FinalDownload/DownloadId204B791FF6187497778654C0DB3AFF7A/2C83497F-CA16-4F65-83D33B6105B66ECC/download/publicacoes/Guia_de_RI.pdf [Accessed November 18th 2011]. James, A. E., Jennings, J. E. and Breitkreuz, R. S. (2012). Worlds Apart? Rebridging the Distance Between Family Science and Family Business Research. Family Business Review, 25(87): 87-108. Janjuha-Jivraj, S. and Spence, L. J. (2009). The Nature of Reciprocity in Family Firm Succession. International Small Business Journal, 27(6): 702–719. Jehn, K. A. (1997). A quantitative analysis of conflict types and dimensions in organizational groups. Administrative Science Quarterly, 42(3): 530–558. Jehn, K. A. and Mannix, E. A. (2001). The dynamic nature of conflict: A longitudinal study of intragroup conflict and group performance. Academy of Management Journal, 44(2): 238–251. Jensen, M. C. (1993). The Modern Industrial Revolution: Exit, and the Failure of Internal Control Systems. Journal of Finance, 48(3): 831-880. 319 Jensen, M. C. (1994). Self-interest, altruism, incentives and agency theory. Journal of Applied Corporate Finance, 7: 40–45. Jensen, M. C. (2000). Value Maximization, Stakeholder Theory and the Corporate Objective Function. Working Paper 00-0058. Boston: Harvard Business School. Jensen, M. C. and Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3: 305– 360. Jimenez, R. M. (1999). Research on Women in Family-owned firms: Current Status and Future Directions. Family Business Review, 22(1): 53-64. Johnson, J. L., Daily, C. M. and Ellstrand, A. E. (1996). Board of directors: A review and research agenda. Journal of Management, 22(3): .409-438. Johnson, S., Boone, P., Breach, A. and Friedman, E. (2000). Corporate governance in the Asian financial crisis. Journal of Financial Economics, 58: 141–186. Jonovic, D. J. (1989). Outside review in a wider context: An alternative to the classic board. Family Business Review, 2(2): 125–141. Jöreskog, K. G. and Sörbom, D. (1989). LISREL 7: A guide to the program and applications. 2nd ed. Chicago: SPSS Inc. Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoek, J. D. and Rosenthal, R. A. (1964). Organizational stress: Studies in role conflict and ambiguity. New York: John Wiley. Karasek, R.A. (1979). Job demands, job decision latitude, and mental strain: Implications for job redesign. Administrative Science Quarterly, 24: 285–308. Karasek, R.A. and Theorell, T. (1990). Healthy work: Stress, productivity, and the reconstruction of working life. New York: Basic Books. Karataş-Özkan, M., Erdoğan, A. and Nicolopoulou, K. (2011). Women in Turkish family businesses: Drivers, contributions and challenges. International Journal of Cross Cultural Management, 11(2): 203-219. Kellermanns, F. W. and Eddleston, K. (2004) Feuding families: When conflict does a family firm good. Entrepreneurship Theory and Practice, 28(3): 209-228. Kelsey, K. (2003) Grounded Theory Designs www.okstate.edu/ag/agedcm4h/academic/aged5980/power/598314.ppt May 9th 2006]. [online]. [Accessed Kellya, E. L., Kossekb, E. E., Hammerc, L. B., Durhamd, M., Braye, J., Chermackf, K., Murphyg, L. A. and Kaskubarh, D. (2008). Getting There from Here: Research on 320 the Effects of Work–Family Initiatives on Work–Family Conflict and Business Outcomes. The Academy of Management Annals, 2(1): 305-349. Kerlinger, F.N. (1980). Metodologia da pesquisa em ciências (Methodology of social science research). São Paulo: EPU/EDUSP. sociais Kesner, I. F. and Dalton, D. R. (1994). Top management turnover and CEO succession: An investigation of the effects of turnover on performance. Journal of Management Studies, 31(5): 701–713. Kesner, I. F. and Johnson, R. B. (1990). An Investigation of the Relationship Between Board Composition and Stockholder Suits. Strategic Management Journal, 11: 327-336. Kets de Vries, M. F. R. (1993). The dynamics of family controlled firms: The good and the bad news. Organizational Dynamics, 21(3): 59-71. Kinnear, T. C. and Taylor, J. R. (1991). Marketing Research: An Applied Approach, New York: Mc Graw Hill. Kiser, E. (1999). Comparing Varieties of Agency Theory in Economics, Political Science, and Sociology: An Illustration from State Policy Implementation. Sociological Theory, 17(2): 146-170. Klein, R. B. (1983). Contracting costs and residual claims: the separation of ownership and control. Journal of Laws and Economic, 26: 367-373. Kline, R. B. (1998). Principles and practice of structural equation modeling. New York: The Guilford Press. Komaroff, E. (1997). Effect of simultaneous violations of essential tau-equivalence and uncorrelated error on coefficient alpha. Applied Psychological Measurement, 21: 337-348. Kotey, B. (2005). Goals, management practices, and performance of family SMEs. International Journal of Entrepreneurial Behaviour & Research, 11(1): 3-24. Kruger, D. J. (2003). Evolution and altruism: combining psychological mediators with naturally selected tendencies. Evolution and Human Behavior, 24, 118–125. Kvale, S. (1996). Interviews: An introduction interviewing. Thousand Oaks, CA: Sage Publications. to qualitative research Kwan, H. K., Lau, V. P. and Au, K. (2011). Effects of Family-to-Work Conflict on Business Owners: The Role of Family Business. Family Business Review, 20(10): 1–13. Lado, A. A., Dant, R. R. and Tekleab, A. G. (2008). Trust-opportunism paradox, relationalism, and performance in interfirm relationships: evidence from the retail industry. Strategic Management Journal, 29: 401–423. 321 Lam, W. (2011). Dancing to two tunes: Multi-entity roles in the family business succession process. International Small Business Journal, 29(5): 508-533. Lambrecht, J. (2005). Multigenerational Transition in Family Businesses. A New Explanatory Model. Family Business Review, 18(4): 267-282. Lane, S., Astrachan, J., Keyt, A. and McMillan, K. (2006). Guidelines for family business boards of directors. Family Business Review, 19: 147-167. Langley, A. (1999). Strategies for theorizing from process data. Academy of Management Review, 24: 691-710. Lansberg, S. (1983). Managing human resources in family-owned firms: the problem of institutional overlap. Organizational Dynamics, 12: 71-80. Lansberg, I. (1988). The succession conspiracy. Family Business Review, 1(2): 119–143. Lansberg, I. (1999). Succeeding generations: Realizing the dream of families in business. Boston: Harvard Business School Press. Lansberg, I., Perrow, E. L. and Rogolsky, S. (1988). Family Business as an emerging field. Family Business Review, 1(1): 1-8. Laplume, A. O., Sonpar, K. and Litz, R. A. (2008). Stakeholder Theory: Reviewing a Theory That Moves Us. Journal of Management, 34(6): 1152-1189. La Porta, R., Lopez-de-Silanes, F. and Shleifer, A. (1997). Legal determinants of external finance. Journal of Finance, 52: 1131–1150. La Porta, R., Lopez-de-Silanes, F. and Shleifer, A. (1998). Corporate ownership around the world. Journal of Finance, 54(2): 717-738. La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R.W. (1999). Corporate ownership around the world. Journal of Finance, 54: 471-517. LaShapelle, K., and Barnes, L. B. (1998). The trust catalyst in family-owned businesses. Family Business Review, 11(1): 1-17. Lauterbach B. and Vaninsky, A. (1999). Ownership Structure and Firm Performance: Evidence from Israel. Journal of Management and Governance, 3(2): 189-201. Lawler, E. E. and Hall, D. T. (1970). Relationship of job characteristics to job involvement, satisfaction, and intrinsic motivation. Journal of Applied Psychology, 54: 305-312. Layder, D. (1993). New Strategies in Social Research: an Introduction and Guide. Cambridge: Polity Press. 322 Leana, C. R. and Van Buren, H. J. III. (1999). Organizational social capital and employment practices. Academy of Management Review, 24(3): 538-555. Leach, P. (1990). Managing the Family Business in the U.K. London: Stoy Hayward. Lebas, M. and Wiegenstein, J. (1986). Management control: The roles of rules, markets and culture. Journal of Management Studies, 23: 259-272. Lee, P. G. and O’Neil, H. M. (2003). Ownership structures and R&D investments of U.S and Japanese firms: agency and stewardship perspectives. Academy of Management Journal, 46(2): 212-225. Leone, N. M. (2005). Sucessão na empresa familiar: preparando as mudanças para garantir sobrevivência no mercado globalizado (Succession in family business: preparing changes to ensure survival in the global market). São Paulo. Ed. Atlas. Leone, N. M. (2008). Os Duelos (Duetos) Identificados por Ocasião do Processo Sucessório (The Duel (Duets) identified through the succession process). Estratégia e Negócios, 1(1): 105-122. Levinson, H. (1971). Conflicts that plague the family business. Harvard Business Review, 49(2): 90-98. Li, J. (1994). Ownership Structure and Board Composition: A Multi-country Test of Agency Theory Predictions. Managerial and Decision Economics, 15: 359-368. Linck, J. S., Netter, J. M. and Yang, T. (2009). The effects and unintended consequences of the Sarbanes-Oxley Act on corporate boards. Review of Financial Studies, 22: 3287-3328. Lima, A. P. (2000). Is blood thicker than economic interest in familial enterprises? In: Schweitzer, P. Dividends of kinship: meaning and uses of social relatedness. London and New York: Routledge. Ling, Y., Lubatkin, M. and Schulze, W. (2001). Altruism, utility functions and agency problems at family-owned firms. In: Craig S. Galbraith (ed.) Strategies and Organizations in Transition (International Research in the Business Disciplines, Volume 3). Lings, B. and Lundell, B. (2005). On the adaptation of Grounded Theory procedures: insights from the evolution of the 2G. Information Technology & People, 8(3): 196211. Lipton, M., and Lorsch, J. (1992). A Modest Proposal for Improved Corporate Governance. Business Lawyer, 48(1): 59-77. Litz, R. A. (1995). The family business: toward definitional clarity. Family Business Review, 8(2): 71-81. 323 Litz, R. A. (1997). The family firm’s exclusion from business school research: Explaining the void, addressing the opportunity. Entrepreneurship Theory and Practice, 21: 55-71. Loan-Clarke, J., Boocock, G., Smith, A. and Whittaker, J. (1999). Investment in management training and development by small business. Employee Relations, 21(3): 296-310. Lodi, J.B. (1998). A Empresa familiar (The family company). 5th ed. São Paulo: Ed. Pioneiras. Lomnitz, L. A. and Pérez-Lizaur, M. (1987). A Mexican elite family, 1820-1980: Kinship, class, and culture. Princeton, NJ: Princeton University Press. Lopes, F. T. (2008). Fotografia de família: contribuições da abordagem psicanalítica para o estudo sobre organizações familiares (Family portrait: contributions of the psychoanalytic approach to the study of family-owned firms). XXXII Encontro da ANPAD. In: Encontro Científico de Administração, Rio de Janeiro. Low, M. B. and MacMillan, I. C. (1988). Entrepreneurship: past research and future challenges. Journal of Management, 14(2), 139–161. Lubatkin, M. H., Schulze, W. S., Ling, Y. and Dino, R. (2005). The effects of parental altruism on the governance of family-managed firms. Journal of Organizational Behavior, 26: 313–330. Lubatkin, M. H., Simsek, Z., Ling, Y. and Veiga, J. F. (2006). Ambidexterity and performance in small to medium-sized firms: The pivotal role of top management team behavioural integration. Journal of Management, 32(5): 646–672. Lyman, A. R. (1991). Customer service: does family ownership make a difference? Family Business Review, 4: 303–324. Lyons, B. and Mehta, J. (1997). Contracts, opportunism and trust: self-interest and social orientation. Cambridge Journal of Economics, 21(2): 239-257. Machado, R. M. C. (2006). A experiência de sucessão para herdeiras de empresas familiares do Rio de Janeiro: estudo de casos. (The experience of succession by heiresses of family owned companies in Rio Janeiro: case studies). Unpublished MA thesis, Instituto Coppead de Administração, Universidade Federal do Rio de Janeiro. Mace, M. (1972). The President and the Board of Directors. Harvard Business Review, 50(2): 37–49. Mackenzie, S. B. (2001). Opportunities for Improving Consumer Research through Latent Variable Structural Equation Modelling. Journal of Consumer Research, 28 (1): 159‐166. 324 Malhotra, N. K. (2001). Pesquisa de marketing: uma orientação aplicada (Marketing research: an applied orientation). 3rd ed. Porto Alegre: Bookman. Mallin, C. A. (2002). The Relationship between Corporate Governance, Transparency and Financial Disclosure. Corporate Governance: An International Review, 10(4): 253–255. Mallin, C. A. (2004). Corporate Governance. New York: Oxford University Press. Mallin, C. and Melis, A. (2012). Shareholder rights, shareholder voting, and corporate performance. Journal of Management and Governance, 16(2):171–176. Maroco, J. and Garcia-Marques, T. (2006). Qual a fiabilidade do alfa de Cronbach? Questões antigas e soluções modernas (What is the reliability of the Cronbach's alpha? Old issues old and modern solutions). Laboratório Psicologia, 4: 65-90. Martensen, A., Gronholdt, L., Bendtsen, L. and Jensen, M. J. (2007). Application of a model for the effectiveness of event marketing. Journal of Advertising Research, 47(3): 283–301. Martínez, J. I.; Stöhr, B. S. and Quiroga, B. F. (2007). Family Ownership and Firm Performance: Evidence From Public Companies in Chile. Family Business Review, 20(2): 83-94. MassMutual Financial Group and Raymond Family Business Institute (2003). American Family Business Survey. [online] www.caregiverslibrary.org/portals/0/CGM.Family%20Businesses%20Changing%20H ands%20AFBSresults.pdf [Accessed October 24th 2011] Matthews, C. H., Vasudevan, D. P., Barton, S. L. and Apana, R., (1994). Capital Structure Decision Making in Privately Held Firms: Beyond the Finance Paradigm. Family Business Review, 7(4): 349-367. Mayer, R., Davis, J. and Schoorman, F. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3): 709–734. Mayer, R. C. and Schoorman, F. D. (1992). Predicting Participation and Production Outcomes through a Two-Dimensional Model of Organizational Commitment. Academy of Management Journal, 35: 671-684. McCahery, J. A. and Vermeulen, E. P. M. (2006). Corporate Governance and Innovation - Venture Capital, Joint Ventures, and Family Businesses. ECGI Law Working Paper No. 65/2006. [online] http://ssrn.com/abstract=894785. [Accessed October 27th 2011]. McConaughy, D. L. and Phillips, G. M. (1999). Founders versus descendants: The profitability, efficiency, growth characteristics and financing in large, public, foundingfamily-controlled firms. Family Business Review, 12(2): 123–131. 325 McConaughy, D. L. (2000). Family CEOs vs. Nonfamliy CEOs in the FamilyControlled Firm: An Examination of the Level and Sensitivity of Pay to Performance. Family Business Review, 13(2): 121–131. McConaughy, D. L., Matthews, C. H. and Fialko, A. S. (2001). Founding-family controlled firms: performance risk and value. Journal of Small Business Management, 39(1): 31-49. McConaughy, D. L., Walker, M., Henderson, G. and Mishra, C. (1998). Founding family controlled firms: efficiency and value. Review of Financial Economics, 7: 119. Mckinsey & Company (2002) Global Investor Opinion Survey: Key Finds. [Unpublished Paper]. Corporate Governance Program. Fundação Dom Cabral. Mckinsey & Company and Korn Ferry International (2001) Panorama da Governança Corporativa no Brasil (Panorama of Corporate Governance in Brazil) [online]. www.ibgc.org.br/ibConteudo.asp?IDArea=61&IDp=2 [Accessed May 9th 2006). McLaren, D. (2004). Global Stakeholders: corporate accountability and investor engagement. Corporate Governance: An International Review, 12(2): 191-201. Means, G. C. (1931). The separation of control in American industry. Quartely Journal of Economics. 46(1): 68-100. Meyer, J. P. and Allen, N.J. (1991). A three-component conceptualization of organizational commitment. Human Resource Management Review, 1(1): 61-89. Miller, D. and Le Breton-Miller, I. (2006). Family Governance and Firm Performance: Agency, Stewardship, and Capabilities. Family Business Review, 19(1): 73-87. Miller, E. J. and Rice, A. K. (1967). Systems of Organization: Task and Sentient Groups and their Boundary Control. London: Tavistock Publications. Miller, E. J. and Rice, A. K. (1988). The Family Business in Contemporary Society. Family Business Review, 1(2): 193-210. Milton, L. P. (2008). Unleashing the relational power of family-owned firms: Identity confirmation as a catalyst for performance. Entrepreneurship Theory and Practice, 32: 1063–1081. Mitchell, J. R., Hart, T. A., Valcea, S. and Townsend D. M. (2009). Becoming the Boss: Discretion and post-succession success in family-owned firms. Entrepreneurship: Theory and Practice, 33(6): 1201–1218. Mitchell, R. K., Morse, E. A. and Sharma, P. (2003). The transacting cognitions of nonfamliy employees in the family businesses setting. Journal of Business Venturing, 18(4): 533–551. 326 Mitnick, B. M. (1973). Fiduciary rationality and public policy: The theory of agency and some consequences. Paper presented at the Annual Meeting of the American Political Science Association, New Orleans, LA. In Proceedings of the APSA, 1973 (formerly available from Xerox University Microfilms and, later, UMI Serials). Mitnick, B. M. (1982). Regulation and the Theory of Agency. Review of Policy Research, 1(3): 442-453. Mitnick, B. M. (2006). Origin of the Theory of Agency: An Account by One of the Theory’s Originators. [online]. www.pitt.edu/~mitnick/agencytheory/agencytheoryoriginrev11806r.htm [Accessed December 12th 2009]. Moe, T. M. (1984). The New Economics of Organisation. American Journal of Political Science, 28: 739-777. Mooi, E., and Sarstedt, M. (2011). A concise guide to market research: The process, data, and methods using IBM SPSS statistics. Berlin: Springer. Molly, V., Laveren, E. and Deloof, M. (2010). Family Business Succession and Its Impact on Financial Structure and Performance. Family Business Review, 23(2): 131-147. Monks, R. A. G. and Minow, N. (2004). Corporate Governance. 3rd ed. Massachusetts: Blackwell Publishing. Moore, M. T. (2002). Corporate governance: An experienced model. Director’s Monthly, 26(3): 1–9. Morck, R. Shleifer, A. and Vishny R. W. (1988). Management ownership and market valuation: an empirical analysis. Journal of Financial Economics, 20: 293-316. Morck, R. and Yeung, B. (2004). Family control and the rent-seeking society. Entrepreneurship Theory and Practice, 28: 391-409. Moreira Junior, A. L. and Altheman, E. (2004). Empresa familiar: uma outra perspectiva, profissionalização e sucessão (The family company: another perspective, professionalization and succession). Revista da ESPM, 11(5): 20-24. Morgan, R. M. and Hunt, S. D. (1994). The Commitment-Trust Theory of Relationship Marketing. Journal of Marketing, 58: 20-38. Morris, M., Williams, R., Allen, J. and Avila, R. (1997). Correlates of success in family business transitions. Journal of Business Venturing, 12(5): 385–401. Munhall, P. L. (2007). Nursing research: A qualitative perspective. 4th ed. Sudbury, MA: Jones & Bartlett. 327 Mustakallio, M.A. (2002). Contractual and relational governance in family-owned firms: effects on strategic decision-making quality and firm performance. Doctoral dissertation, Helsink University of Technology. Mustakallio M., Autio E., and Zahra S. A. (2002). Relational and Contractual Governance in Family-owned firms: Effects on Strategic Decision Making. Family Business Review, 15: 205-222. Muthén, B. and Kaplan D. (1985). A comparison of some methodologies for the factor analysis of non-normal Likert variables. British Journal of Mathematical and Statistical Psychology, 38, 171-189. Nadler, D. A. (2004). Building Better Boards. Harvard Business Review. 82(5): 1-9. Nahapiet, J. and Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. Academy of Management Review, 23(2): 242-266. Nash, J.M. (1995). Boards of privately held companies: A guide. National Association of Corporate Directors Governance Series: Private Company Boards. Nelton, S. (1998). The bright side of nepotism. Nation’s Business, 86(5): 72. Netemeyer, R. G.; Bearden, W. O. and Sharma, S. (2003). Scaling procedures: Issues and Applications. Thousand Oaks, CA: SAGE. Neubauer, F., and Lank, A. (1998). The family business: Its governance for sustainability. New York: Routledge. Newell, R. and Wilson, G. (2002). A premium for good governance. McKinsey Quarterly, 3: 20–23. Nooteboom, B. (1996). Trust, Opportunism and Governance: A Process and Control Model. Organization Studies, 17(6): 985-1010. Nordqvist, M. (2012). Understanding strategy processes in family-owned firms: Exploring the roles of actors and arenas. International Small Business Journal, 30(1): 24-40. Nunnally, J. C. (1978). Psychometric theory. 2nd ed. New York: McGraw-Hill. Nunnaly, J. C. and Bernstein, I. H. (1994). Psychometric Theory. 3rd ed. New York: McGrawHill. Oliva, E. C. and Albuquerque, L. G. (2007). Sistema de Remuneração de Executivos e Conselheiros como Suporte à Estrutura de Governança Corporativa (Executive and Board Members Compensation System to Support Corporate Governance). BASE - Revista de Administração e Contabilidade da Unisinos, 4(1): 61-73. 328 Oliveira, D. de P. R. de. (1999). Empresa familiar – Como fortalecer o empreendimento e otimizar o processo sucessório (Family business - How to strengthen the enterprise and optimize the process of succession). São Paulo: Ed. Atlas. Oliveira Jr., L. A. (2009). Ações Facilitadoras ao Processo de Sucessão e o Êxito de Empresas Familiares na Terceira Geração em Diante (Actions Facilitating the Succession Process and Family Business Success of the Third Generation forward). Unpublished master’s thesis. Rio de Janeiro: Fundação Getúlio Vargas. Olson D. H. and Gorall D. M. (2003). Circumplex model of marital and family systems. In F. Walsh (Ed.), Normal Family Processes 3rd ed. New York: Guilford. Olson, P. D., Zuiker,V. S., Danes, S. M., Stafford, K., Heck, R. K. Z. and Duncan, K. A. (2003). Impact of family and business on family business sustainability. Journal of Business Venturing, 18(5): 639–666. Oppenheim, A. (1972). Questionnaire design and attitude measurement, London: Heinemann. O’Reilly, C. (1989). Corporations, Culture, and Commitment: Motivation and Social Control in Organizations. California Management Review, 31(4): 9-25. Osburn, H. G. (2000). Coefficient alpha and related internal consistency reliability coefficients. Psychological Methods, 5: 343–355. Osnes, Gry. (2011). Succession and authority: A case study of an African family business and a clan chief. International Journal of Cross Cultural Management, 11(2): 185-201. Ouchi, W. G. (1980). Markets, bureaucracies, and clans. Administrative Science Quarterly, 25(1): 129-141. Pádua, A. D. (2004). Empresa familiar: profissionalização, desenvolvimento e sucesso (The family company: professionalization, development and success). Porto Alegre: Sebrae/RS. Paine, L., Deshpande, R., Margolis, J., and Bettcher, K. (2005). Does your company’s conduct meet world class standards? Harvard Business Review, 83(12): 122–133. Pandit, N. R. (1996). The Creation of Theory: A Recent Application of the Grounded Theory Method. The Qualitative Report, 2(4): 1-21. Parker, T. (2004). A Multi-Level Family Business Choice Model: A Dichotomous Approach. The Coastal Business Journal, 3(1): 56-60. Patel, S., Balic, A. and Bwakira, L. (2002). Measuring Transparency and Disclosure at Firm-level in Emerging Markets. Emerging Markets Review, 3: 325–337. 329 Pearce, J. A and Zahra, S. A. (1989). Boards of directors and corporate financial performance: A review and integrative model. Journal of Management, 15(2): 291334. Pearce, J. A. and Zahra, S. A. (1992). Board Composition from a Strategic Contingency Perspective. Journal of Management Studies, 29(4): 411-438. Pereira, R. D. (2010). Sucessão, Profissionalização e Governança: Um Estudo Multicasos em Empresas Familiares Mineiras (Succession, Professionalization and Governance: A multicase study in Family Business of Minas Gerais) Unpublished master’s thesis. Universidade Federal de Minas Gerais. Perman, S. (2006). Taking the Pulse of Family Business www.businessweek.com/smallbiz/ [Accessed February 13th 2006). [online]. Perrow, C. (1986). Complex organizations: A critical view. 3rd ed. New York: Random House. Pieper, T. M., Klein, S. B. and Jaskiewicz, P. (2008). The Impact of Goal Alignment on Board Existence and Top Management Team Composition: Evidence from Family-Influenced Businesses. Journal of Small Business Management, 46(3): 372-394. Piliavin, J. A. and Charng, H. (1990). Altruism: A Review of Recent Theory and Research. Annual Review of Sociology, 16: 27-65. Pollak, R. A. A (1985). Transaction Cost Approach to Families and Households. Journal of Economic Literature, 23: 581-608. Popa, A. and Peres, I, (2008). Aspects Regarding Corporate Mandatory and Voluntary Disclosure. Annals of Faculty of Economics, University of Oradea, 3(1): 1407-1411. Potts, S. D. and Matuszewski, I. L. (2004). Ethics and corporate governance. Corporate Governance International Review, 12(2): 177-179. Prencipe, A., Markarian, G. and Pozza, L. (2008). Earnings Management in Familyowned firms: Evidence from R&D Cost Capitalization in Italy. Family Business Review, 21(1): 71-88. PwC (2011). Kin in the game: PwC Family Business Survey 2010/11. [online] www.pwc.com/gx/en/pwc-family-business-survey/download.jhtml [Accessed January 7th 2012] Quinn, D. and T. Jones (1995). An Agent Morality View of Business Policy. Academy of Management Review, 20(1): 22–42. Quinn, J. B. (1991). Strategies for change. In: Mintzberg, H. The strategy process: concepts, contexts and cases. 2nd ed. Englewood Cliffs, NJ: Prentice-Hall. 330 Raskas, D.F. (1998). Familiarity breeds trust as well as contempt. What about familiality? An examination of familial involvement and trust in family-owned firms. Unpublished doctoral dissertation, Columbia University, USA. Raykov, T. (1997a). Estimation of composite reliability for congeneric measures. Applied Psychological Measurement, 21, 173-184. Raykov, T. (1997b). Scale reliability, Cronbach’s coefficient alpha, and violations of essential tau-equivalence with fixed congeneric components. Multivariate Behavioral Research, 32(4): 329, 353 Riketta, M. (2002). Attitudinal organizational commitment and job performance: A meta-analysis. Journal of Organizational Behavior, 23(3): 257-266. Ring, P. S. and Van de Ven, A. H. (1992). Structuring cooperative relationships between organizations. Strategic Management Journal, 13(17): 483-498. Ring, P. S. and Van de Ven, A. H. (1994). Developmental processes of cooperative inter-organizational relationships. Academy of Management Review, 19(1): 90-118. Roberts, J. (2001). Trust and control in Anglo-American systems of corporate governance: The individualizing and socializing effects of processes of accountability. Human Relations, 54: 1547-1572. Rogoff E. G. and Heck, R. K. Z. (2003). Evolving research in entrepreneurship and family business: recognizing family as the oxygen that feeds the fire of entrepreneurship. Journal of Business Venturing, 18: 559–566. Rosenblatt, P. C., de Mik, L., Anderson, R. M. and Johnson, P. A. (1985). The family in business. San Francisco: Jossey-Bass. Ross, S. (1973). The economic theory of agency: The principal's problem. American Economic Review, 63(2): 134-139. Rossato Neto, F. J. and Cavedon, N. R. (2004). Empresas Familiares: desfilando seus processos sucessórios (Family Business: parading their succession processes). Cadernos EBAPE, 2(3): 1-16. Royer, S., Simons, R., Boyd, B. and Rafferty, A. (2008). Promoting family: A contingency model of family business succession. Family Business Review, 21(1): 15–30. Santiago, A. L. (2000). Succession experiences in Philippine family businesses. Family Business Review, 13(1): 15-40. Santos, H. M. (2000). Conselhos de administração: um estudo do funcionamento da governança corporativa no contexto brasileiro (Boards of directors: a study of the functioning of corporate governance in the Brazilian context). São Paulo, SP. Unpublished master’s thesis. Universidade de São Paulo – FEAUSP. 331 Schein, E. (1985). How culture forms, develops and changes. IN: Gaining Control of the Corporate Culture. Ed. R. Kilmann, M. Saxton, and R. Serpa. San Fransico: Jossey Bass. Scholes, L., Wright, M., Westhead, P. and Bruining, H. (2010). Strategic changes in family-owned firms post management buyout: Ownership and governance issues. International Small Business Journal, 28(5): 505-521. Schulze, W. S., Lubatkin, M. H., Dino, R. N. and Buchholtz, A. K. (2001). Agency relationships in family-owned firms: Theory and evidence. Organization Science, 12: 99-116. Schulze, W. S., Lubatkin, M. H. and Dino, R. N. (2003a). Toward a theory of agency and altruism in family-owned firms. Journal of Business Venturing, 18(4): 473–490. Schulze, W. S., Lubatkin, M. H. and Dino, R. N. (2003b). Exploring the agency consequences of ownership dispersion among inside directors at family-owned firms. Academy of Management Journal, 46(2): 179–194. Sciascia, S. and Mazzola, P. (2008). Family Involvement in Ownership and Management: Exploring Nonlinear Effects on Performance. Family Business Review, 21(4): 331-345. Shankman, N. A. (1999). Reframing the Debate between Agency and Stakeholder Theories of the Firm. Journal of Business Ethics, 19: 319-334. Shapiro, S. P. (1987). The Social Control of Impersonal Trust. American Journal of Sociology, 93(3): 623-658. Shapiro, S. P. (1990). Collaring the crime, not the criminal: reconsidering the concept of whitecollar crime. American Sociological Review, 55: 346–365. Shapiro, S. P. (2005). Agency Theory. Annual Review of Sociology, 31: 263–284. Sharma, P. (1997). Determinants of the satisfaction if the primary stakeholders with the succession process in family-owned firms. Unpublished doctoral thesis, University of Calgary, Canada. Sharma, P., Chrisman, J. J., Pablo, A. L. and Chua, J. H. (2001). Determinants of initial satisfaction with the succession process in family-owned firms: A conceptual model. Entrepreneurship Theory and Practice, 25(3): 17-35. Shaw. J. C., Wild, E. and Colquitt, J. A. (2003). To justify or excuse? A meta-analytic review of the effects of explanations. Journal of Applied Psychology, 88: 444-458. Shepherd, D. and Haynie, J. M. (2009). Family Business, Identity Conflict, and an Expedited Entrepreneurial Process: A Process of Resolving Identity Conflict. Entrepreneurship Theory and Practice, 33(6): 1245–1264. 332 Shleifer, A. and Vishny, R. W. (1997). A survey of corporate governance. Journal of Finance, 52: 737–783. Siggelkow, N. (2007). Persuasion with Case Studies. Academy of Management Journal, 50(1): 20–24. Silveira, A. Di M. (2006). Governança Corporativa e Estrutura de Propriedade (Corporate Governance and Ownership Structure). São Paulo: Ed. Saint Paul Institute of Finance. Slagmulder, R. (1997). Using management control systems to achieve alignment between strategic investment decisions and strategy. Management Accounting Research, 8: 103-139. Smyrnios, K. X., Romano, C. A., Tanewski, G. A., Karofsky, R. M., Millen, R. and Yilmaz, M, R. (2003). Work-family conflict: a study of American and Australian family businesses. Family Business Review, 16(1): 35-51. Smyrnios, K. X. and Walker, R.H. (2003). Australian Family and Private Business Survey. The Boyd Partners and RMIT, Australia. Songini, L. (2006). The professionalization of family-owned firms: Theory and practice. In P. Poutziouris, K. Smyrnios, & S. Klein (Eds.), Handbook of research on family businesses (pp. 269–297). Cheltenham, UK/Brookfield, US: Edward Elgar. Sorenson R. L. (1999). Conflict Strategies Used by Successful Family Businesses. Family Business Review, 12(4): 325–339. Sorenson, R. L., Goodpaster, K. E., Hedberg, P. R. and Yu, A. (2009). The Family Point of View, Family Social Capital, and Firm Performance: An Exploratory Test. Family Business Review, 22(3): 239-253. Srour, G. (2005). Práticas Diferenciadas de Governança Corporativa: Um Estudo sobre a Conduta e a Performance das Firmas Brasileiras (Corporate Governance Practices: A Study of the Conduct and Performance of Brazilian Firms). Revista Brasileira de Economia, 59(4): 635-674. Steier, L. P., Chua, J. H. and Chrisman, J. J. (2009). Embeddedness perspectives of economic action within family-owned firms. Entrepreneurship Theory and Practice, 33: 1157–1167. Steinberg, H. A (2003). Dimensão Humana da Governança Corporativa: pessoas criam as melhores e as piores práticas (Human Dimension of Corporate Governance: People create the best and worst practices). São Paulo: Editora Gente. Stern, P. N. (1980). Grounded Theory Methodology: Its Uses and Processes. Journal of Nursing Scholarship, 12: 20–23 Sternberg, E. (1999). The Stakeholders Concept: A Mistake Doctrine. Issue paper n.4. Leeds: Foundation for Business Responsibilities. 333 Stewart, A. (2003). Help one another, use one another: Toward an anthropology of family business. Entrepreneurship Theory and Practice, 27: 383-396. Stewart, A. and Hitt, M. A. (2012). Why Can’t a Family Business Be More Like a Nonfamliy Business? Modes of Professionalization in Family-owned firms. Family Business Review, 25(1): 58-86. Strader, J. K. (2002). Understanding white collar crimes. Library of Congress Cataloging-in-Publication Data, (Supplement). Strauss, A. and Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage Publications, Inc. Strauss, A. and Corbin, J. (1994). Grounded Theory methodology: An overview. In: Handbook of Qualitative Research (Denzin, N., K. and Lincoln, Y.,S., Eds.). Sage Publications, London. Suddaby, R. (2006). From the editors: what grounded theory is not. Academy of Management Journal, 49(4): 633-642. Sundaramurthy, C. (2008). Sustaining Trust within Family Businesses. Family Business Review, 21(1): 89-102. Sundaramurthy, C. and Lewis, M. (2003). Control and collaboration: Paradoxes of governance. Academy of Management Review, 28: 397-415. Sutton-Brady, C. (2000). Towards Developing a Construct of Relationship Atmosphere. Paper presented at the 16th Annual IMP Conference, Bath. Tabachnick, B. G. and Fidell, L. S. (2001). Using Multivariate Statistics. 3rd ed. New York: HarperCollins. Taber, K. S. (2000). Case studies and generalizability: Grounded theory and research in science education. International Journal of Science Education, 22(5): 469-487. Tagiuri, R. and Davis, J. (1996). Bivalent attributes of the family firm. Family Business Review, 9(2): 199-208. Tàpies, J. and M. Fernández (2010). Values and longevity in family business: Evidence from a cross-cultural analysis. Working paper no. 866, IESE Business School, University of Navarra, Barcelona, Spain. Taylor, S. E. and Koivumaki, J. H. (1976). The perception of self and others: Acquaintanceship, affect, and actor-observer differences. Journal of Personality and Social Psychology, 33: 403-408. Thiollent, M. (1988). Metodologia Methodology). 4th ed. São Paulo : Cortez. da 334 pesquisa-ação (Research-action Tricker, B. (2000). Editorial: Corporate Governance – the subject whose time has come. Corporate Governance: An International Review, 8(4): 289-296. Triviños, A. R. S. (1987). Introdução a Pesquisa em Ciências Sociais: Pesquisa Qualitativa em Educação (Introduction to Social Science Research: Qualitative Research in Education). São Paulo: Atlas. Trochim, W. M. (2006). The Research Methods Knowledge Base. 2nd ed. [online] http://ifmr.ac.in/cmf/mrap/wp-content/uploads/2009/04/qualitative-measures.pdf [Accessed July 19th 2011]. Tsoukas, H. (1996). The firm as a distributed knowledge system: A constructionist approach. Strategic Management Journal, 17: 11-26. UNCTAD (2006). Guidance on Good Practices in Corporate Governance Disclosure. [online] www.unctad.org/en/docs/iteteb20063_en.pdf [Accessed November 8th 2011]. Uzzi, B. (1996). The sources and consequences of Embeddedness for economic performance of organizations: The network effect. American Sociological Review, 61: 674-698. Van den Berghe, L. A. A. and Carchon, S. (2003). Agency relations within the family system: An exploratory approach. Corporate Governance: An International Review, 11(3): 171-180. Van der Heyden, L., Blondell, C. and Carlock, R. S. (2005). Striving for Justice in Family Business. Family Business Review, 18(1): 1-21. Van der Merwe, S. P. (2009). An investigation of the determinants of estate and retirement planning in intergenerational family businesses. South African Journal of Business Management, 40(3): 51–63. Veliyath, R. and Ramaswamy, K. (2000). Social Embeddedness, Overt and Covert Power, and Their Effects on CEO Pay: An Empirical Examination among Family Businesses in India. Family Business Review, 13: 293–312. von Lossberg, A. (1990). The Role of the Nonfamliy Administrator in Family Foundations. Family Business Review, 3: 375-382. Venter, E., Boshoff, C. and Maas, G. (2005). The influence of successor-related factors on the succession process in small and medium-sized family businesses. Family Business Review, 18(4): 283-303. Voordeckers, W., Van Gils, A. and Van den Heuvel, J. (2007). Board composition in small and medium-sized family-owned firms. Journal of Small Business Management, 45(1), 137-156. 335 Voss, C., Tsikriktsis, N. and Frohlich, M. (2002). Case research in operations management. International Journal of Operations & Production Management, 22(2): 195-219. Wacker, J. (1998). A definition of theory: Research guidelines for different theorybuilding research methods in operations management. Journal of Operations Management, 16: 361-385. Walsh, F. (1994). Healthy family functioning: Conceptual developments. Family Business Review, 7(2): 175-198. and research Wang, Y., Ahmed, P. and Farquhar, S. (2007). Founders versus Descendants: The Profitability, Growth and Efficiency Characteristics Comparison in the UK Small and Medium Sized Family Businesses. Journal of Entrepreneurship, 16(2): 173-195. Warburton, W.I. (2005). What are grounded theories made of? In, University of Southamtpon LASS Faculty Post-graduate Research Conference, Southampton, UK. Ward, J. L. (1987). Keeping the family business healthy: How to plan for continuing, growth, profitability, and family leadership. San Francisco, CA: Jossey-Bass. Ward, J. L. (1988). The special role of strategic planning for family businesses. Family Business Review, 1(2): 105-118. Ward, J. L. (1991). Creating effective boards for private enterprises. San Francisco, CA: Jossey-Bass Publishers. Ward, J. L. (1997). Growing the family business: special challenges and best practice. Family Business Review, 10(4): 323-337. Ward, J. L. (2004). Perpetuating the Family Business: 50 Lessons Learned from Long-Lasting Successful Families in Business, Basingstoke: Palgrave Macmillan. Ward, J. L. and Handy, J. (1988). A survey of board practice. Family Business Review, 1(3): 289-308. Weber, M. (1971). Ensaios de Sociologia (Essays in Sociology). 2nd ed., Rio de Janeiro: Zahar. Weber, M. (1999). Economia e sociedade: fundamentos de sociologia compreensiva (Economy and society: fundamentals of comprehensive sociology). Brasília: UnB. Weber, R. P. (1990). Basic Content Analysis. Newbury Park, CA: Sage Publications. Weinstein, A. G. (1999). Family business in the United States – Research and observations. Presented at the 44th World Conference of the International Council for Small Business. Naples, Italy. 336 Weisbach, S. M. (1988). Outside Directors and CEO Turnover. Journal of Financial Economics. 20: 431-460. Westhead, P. and Howorth, C. A. (2006). Ownership and Management Issues Associated With Family Firm Performance and Company Objectives. Family Business Review, 19(4): 301-316. Westhead, P., Howorth, C. A. and Cowling, M. (2002). Ownership and management issues in first and multi-generation family-owned firms. Entrepreneurship and Regional Development, 14(3): 247-269. Westphal, J. D. (1999). Collaboration in the board room: Behavioural and performance consequences of CEO-board social ties. Academy of Management Journal, 42(1): 7-24. Whetten, D. A. (1989). What Constitutes a Theoretical Contribution? Academy of Management Review, 14(4): 490-495. Whyte, M. K. (1996). The Chinese family and economic development: Obstacle or engine? Economic Development and Cultural Change, 44: 1-30. Williamson, O. (1975). Markets and hierarchies. New York: Free Press. Williamson, O. (1985). The Economic Institution of Capitalism. New York: Free Press. Wilkins, A. L., and W. G. Ouchi (1983). Efficient Cultures: Exploring the Relationship between Culture and Organizational Performance. Administrative Science Quarterly 28: 468–481. Wood Jr., T. and Picarelli Filho, V. (2004). Remuneração e carreira por habilidades e competências (Remuneration and career by skills and competencies). São Paulo: Atlas. Yin, R. K. (1994). Case study research: Design and methods. 2nd ed. Beverly Hills, CA: Sage Publishing. Yin, R. K. (2005). Introducing the world of education. A case study reader. Thousand Oaks: Sage Publications. Young, M. N., Peng, M. W., Ahlstrom, D., Bruton, G. D. and Jiang, Y. (2008). Corporate Governance in Emerging Economies: A Review of the Principal–Principal Perspective. Journal of Management Studies, 45: 196–220. Yu, A., Lumpkin, G. T., L. Sorenson, R. L. and Brigham, K. H. (2012). The Landscape of Family Business Outcomes: A Summary and Numerical Taxonomy of Dependent Variables. Family Business Review, 25(1) 33-57. 337 Yupitun, M. A. (2008). Agency Trade-offs in Family-owned firms: Theoretical Model, Empirical Testing and Implications. Unpublished doctoral thesis, Bond University, Australia. Zahra, S. A., Hayton, J. C. and Salvato, C. (2004). Entrepreneurship in family vs. nonfamliy-owned firms: A resource based analysis of the effect of organizational culture. Entrepreneurship Theory and Practice, 28(4): 363–381. Zahra, S. A., Hayton, J. C., Neubaum, D. O., Dibrell, C. and Craig, J. (2008). Culture of Family Commitment and Strategic Flexibility: The Moderating Effect of Stewardship, Entrepreneurship Theory and Practice, 32: 1035–1054. Zellweger, T. M. (2006). Risk, Return and Value in the Family Firm. Unpublished doctoral thesis, University of St. Gallen, Switzerland. Zellweger, T. M. and Astrachan, J. H. (2008). On the Emotional Value of Owning a Firm. Family Business Review, 21(4): 347-363. Zellweger, T. M. and Nason, R. S. (2008). A Stakeholder Perspective on Family Firm Performance. Family Business Review, 21(3): 203-216. Zhang, Y. and Wildemuth, B. M. (2009). Qualitative analysis of content, in Applications of social research methods to questions in information and library science, Westport, Connecticut: Libraries Unlimited. 338 APPENDIX A Corporate Governance in Family-owned Firms This research is part of the development of the doctoral thesis of Dalton Penedo Sardenberg at the University of Birmingham, UK, sponsored by Fundação Dom Cabral. PhD Student Dalton Penedo Sardenberg ‘This survey takes approximately 25 minutes to be completed’. 339 A. Your perception of your company (or corporate group) In this section, most questions are presented as statements. You should choose always using a number from 1 to 7, according to the following scale: Totally disagree 1 2 3 4 5 6 7 Totally agree 1: means you disagree completely with the statement 2: means you largely disagree with the statement 3: means you somewhat disagree with the statement 4: means your opinion is indifferent about the statement 5: means you somewhat agree with the statement 6: means you largely agree with the statement 7: means you agree completely with the statement NA: means this statement does not apply to your situation or you do not have enough information to assess it.  PLEASE CIRCLE ONLY ONE NUMBER PER QUESTION OR THE OPTION ‘NA’.  DO NOT LEAVE ANY QUESTION IN BLANK.  DO NOT EVALUATE BASED ON WHAT YOU IMAGINE TO BE THE IDEAL STAGE TO BE REACHED BY YOUR COMPANY IN THE FUTURE. REPLY ACCORDING TO THE PRESENT SITUATION. SHOULD SOME IMPORTANT CHANGE IN THE COMPANY’S GOVERNANCE STRUCTURE HAVE OCCURRED, THE RISE OF A NEW GENERATION TO ASSUME CONTROL OR A CHANGE IN THE SHARE PARTICIPATION, YOUR ASSESSMENT SHOULD TAKE INTO ACCOUNT THIS NEW SITUATION.  CONFIDENTIALITY: I REALIZE THAT SOME QUESTIONS ASK FOR POTENTIALLY SENSITIVE INFORMATION AND I GUARANTEE TOTAL CONFIDENTIALITY IN RELATION TO SUCH INFORMATION IN YOUR ANSWERS. NOBODY, BESIDES MYSELF WILL ACCESS THIS QUESTIONNAIRE. THEREFORE, I ASK FOR TOTAL SINCERITY IN YOUR ANSWERS. 340 Company (group) evaluated: _________________________________________________ Section I: In relation to the successor Totally disagree Totally agree I.1 The principal challenge faced by the successor is the difficulty of obtaining immediate recognition of his/her authority from family members and staff in the running of the company. 1 2 3 4 5 6 7 NA I.2 Despite the fact that we consider competence in the job an important criterion, kinship ties are still the principal one in choosing the successor. 1 2 3 4 5 6 7 NA I.3 Within our company, the image of the founder is so intense and honoured by all, that it inhibits any family successor from coming naturally into his/her own. 1 2 3 4 5 6 7 NA Section II: Family managers in the firm Totally disagree Totally agree II.1 The employees recognize that our family managers are competent in the functions they exercise. 1 2 3 4 5 6 7 NA II.2 In our company the rules for the entry of family members into the management are clearly negotiated and followed, without exception. 1 2 3 4 5 6 7 NA Section III: Alignment Totally disagree Totally agree III.1. The various members of the controlling family have a distinct notion of the process and the best moment for the succession. 1 2 3 4 5 6 7 NA III.2. In our company, shareholders and managers share a common vision and the same values in relation to the running of the business. 1 2 3 4 5 6 7 NA 341 III.3. In the decisions of great strategic impact in the recent history of the company, shareholders and managers have shown harmony in their thinking and acting. 1 2 3 4 5 6 7 NA III.4. We adopt an ordered planning process that promotes consensus as to the long-term strategic objectives of the organization. 1 2 3 4 5 6 7 NA III.5. Shareholders are in total agreement regarding the standard of risk assumed by the executives in the search for results. 1 2 3 4 5 6 7 NA III.6. The remuneration and incentives practised by our company have constituted an instrument capable of producing alignment of the interests between the executives and shareholders. 1 2 3 4 5 6 7 NA III.7. The existence of an instrument regulating interests (shareholders' agreement, social contract or statute) has shown itself to be fundamental for the promotion of cohesion among the various partners. 1 2 3 4 5 6 7 NA Section IV: Family interaction IV.1 Is there a Family Council whose object is to represent the family as a whole and prepare the formal meetings of its members? ❑ Yes ❑ No Totally disagree Totally agree IV.2 We do not think that is vital the constitution of a Family Council to guarantee family harmony and a good relationship of the family with the company. 1 2 3 4 5 6 7 NA IV.3 Relationships between the members of our family have been extremely strained, hindering the possibility of a good family interaction. 1 2 3 4 5 6 7 NA IV.4 The members of our family know the history of our company and are interested in the values that guide us in managing it. 1 2 3 4 5 6 7 NA 342 IV.5 Our family ties are sufficiently strong to guarantee the transition of the founders’ values to future generations. 1 2 3 4 5 6 7 NA IV.6 The members of the owner-family (ies) show lack of interest in participating in events and celebrations of the company to improve their closer relationships with employees and nonfamliy managers. 1 2 3 4 5 6 7 NA Section V: Trust, Integrity and Risk Perception Totally disagree Totally agree V.1 In our company it is common to find managers (family or nonfamliy) whose work shows shortcomings and whose attitudes can typically be classified as ‘shirking’. 1 2 3 4 5 6 7 NA V.2 In our recent history there have been several situations involving the improper use or misappropriation of funds that could be characterized as ‘white collar crimes’. 1 2 3 4 5 6 7 NA V.3 The managers of our company are not in the habit of establishing excessive levels of remuneration or unjustifiable dividends or bonuses in their own benefit. 1 2 3 4 5 6 7 NA V.4 It is acceptable that an individual, being part of the owner family, makes use of company resources for his or her own benefit. 1 2 3 4 5 6 7 NA V.5 In spite of having effective control of the company, the majority shareholders have never used this power to create privileges at the expense of the rights of the minority. 1 2 3 4 5 6 7 NA V.6 We have already faced situations that have led us to no longer trust, unconditionally, nonfamliy members. 1 2 3 4 5 6 7 NA V.7 All family members that work, or have worked in the company, have always demonstrated high professional integrity, possessing similar beliefs and values to those of the founder(s). 1 2 3 4 5 6 7 NA V.8 The indication of nonfamliy managers by members of our social network is the best way of guaranteeing the selection of trustworthy executives. 343 1 2 3 4 5 6 7 NA V.9 The shareholders perceive that the actions taken by the managers always have the final objective of protecting their interests, or at least not to run counter to them. 1 2 3 4 5 6 7 NA V.10 Managers feel that they should not openly reveal their difficulties to the shareholders (or board members, if they exist) for fear of this being used against them in the future. 1 2 3 4 5 6 7 NA Section VI: Competence and disclosure Totally disagree Totally agree VI.1 The family managers have always demonstrated that they are totally capable of conducting significant necessary changes, even if these have required breaking with tradition. 1 2 3 4 5 6 7 NA VI.2 In times of change, family managers tend to demonstrate less disposition in running risks than nonfamliy managers would if they were in the same position. 1 2 3 4 5 6 7 NA VI.3 The competence of our managers has shown itself to be the principal cause of reaching the results that we have. 1 2 3 4 5 6 7 NA VI.4 All shareholders have at their command unrestricted access to the financial and nonfinancial information which guide business action in the attempt to create value. 1 2 3 4 5 6 7 NA VI.5 Efficient channels exist for the dissemination of information aimed at the public of interest, whether they are internal or external to the company. 1 2 3 4 5 6 7 NA VI.6 Something that we lack is a succession process that is openly discussed and treated with transparency as to the criteria for the choice of the successor. 1 2 3 4 5 6 7 NA VI.7 Our executives have the habit providing information, positive and negative, in addition to what is obligatory, to all interested parties, to facilitate the correct comprehension and evaluation of the organization. 1 2 3 4 5 6 7 NA 344 Section VII: Management Control Totally disagree Totally agree VII.1 There exists a clear and formal process for the evaluation of the performance of the CEO and of his/her executive group, in reaching the objectives for the shareholders. 1 2 3 4 5 6 7 NA VII.2 Financial controls and the indicators of return are regularly followed by the shareholders and/or their representatives. 1 2 3 4 5 6 7 NA VII.3 We adopt a rigorous internal control system to monitor the operations of our company. 1 2 3 4 5 6 7 NA VII.4 The levels of autonomy of the managers are negotiated guaranteeing agility of decision without releasing them from responsibility for the results arising therefrom. 1 2 3 4 5 6 7 NA VII.5 Within the present control mechanism situations have arisen where the managers have taken decisions harmful to the company, which were not detected in time by the shareholders due to their having adopted a more laissez-fair oversight. 1 2 3 4 5 6 7 NA Section VIII: Impartiality Totally disagree Totally agree VIII.1 In our situation, an error of a family manager in the performance of his/her functions deserves the same sort of treatment given to a nonfamliy manager. 1 2 3 4 5 6 7 NA VIII.2 The criteria for the evaluation of the performance of the family manager do not differ in any detail from those used for the nonfamliy manager. 1 2 3 4 5 6 7 NA VIII.3 In our company the criteria used in the selection to key positions prioritize the level of competence of the candidate, and are completely independent of kinship. 345 1 2 3 4 5 6 7 NA VIII.4 The economic objectives of the company are always prioritized, even if they are in detriment to the well-being of the family members. 1 2 3 4 5 6 7 NA VIII.5 Historically, the company has been a natural path for giving sanctuary to family members interested in joining it. 1 2 3 4 5 6 7 NA VIII.6 The owning family perceives the company as a source of opportunity for career development for future generations. 1 2 3 4 5 6 7 NA VIII.7 Family affection leads to unconditional acceptance of behaviour, even if inappropriate. 1 2 3 4 5 6 7 NA VIII.8 Even if a family manager does not discharge his/her executive functions with excellence, the tendency is that we choose to let him/her stay on. 1 2 3 4 5 6 7 NA Section IX: Board of Directors IX.1 Is there a Board of Directors formally established? ❑ Yes How many members does it have? ____________ ❑ No IX.2 Are there any independent members on the Board (members who do not participate in the management nor belong to the owner-families)? ❑ Yes How many? __________ ❑ No ❑ NA Reply to all the questions below, independently of the existence or not of a formal Board of Directors in your organization: 346 Totally disagree Totally agree IX.3 In the constitution of a Board it is more important that all family groups feel that they are adequately represented than to guarantee the presence of Board Members of recognized competence and experience. 1 2 3 4 5 6 7 NA IX.4 A Board of Directors that tends merely to validate the opinion of the CEO regarding the strategic direction of the organization without any questioning would never be accepted in our organization. IX.5 The members of the controlling family of our company definitely recognize (or would recognize in the event of future installation) the Board of Directors as the maximum authority and highest court for strategic decisions. 1 2 3 4 5 6 7 NA 1 2 3 4 5 6 7 NA IX.6 It is understandable that family board members should have greater authority in strategic decisions, than external members, because of their connections and interest in the business. 1 2 3 4 5 6 7 NA IX.7 The presence of board members from outside the family is cause for discomfort when they demand the sharing of strategic information and invade the family space in the company 1 2 3 4 5 6 7 NA Section X: Understanding of social rules Totally disagree Totally agree X.1 The family active in management can clearly distinguish family from professional issues, insuring that each is dealt within its proper place. 1 2 3 4 5 6 7 NA X.2 In a succession process the family CEO, retiring or taking up another position has demonstrated (or would demonstrate in the event of a future succession) total acceptance of his/her new role without imposing him/herself on the successor. 1 2 3 4 5 6 7 NA X.3 It is acceptable that a CEO, member of the controlling family and consequently an interested part in the company results, rejects having his/her performance formally evaluated. 1 2 3 4 5 6 7 NA 347 X.4 The shareholders, family or otherwise, that do not participate in the management, understand their role and ensure their rights through the proper channels, without directly interfering in the day-to-day running of the company. 1 2 3 4 5 6 7 NA Section XI: Existing or latent tensions Please, evaluate the existing or latent tensions between family members, staff and shareholders, using the scale from 1 to 7 - from the existence of no tension of any kind to a high tension situation. Please, do not leave any pair in blank. When one of the actors does not exist, obviously no analysis is applicable. In this case, just circle the option NA. No tension of any kind XI. Family owner/managers versus XI. Nonfamliy managers High tension 1. Nonfamliy managers 1 2 3 4 5 6 7 NA 2. Family managers 1 2 3 4 5 6 7 NA 3. Non-management family members 1 2 3 4 5 6 7 NA 4. Nonfamliy shareholders 1 2 3 4 5 6 7 NA 5. Board members 1 2 3 4 5 6 7 NA 6. Non-management family members 1 2 3 4 5 6 7 NA 7. Board members 1 2 3 4 5 6 7 NA 8. Minority shareholders 1 2 3 4 5 6 7 NA versus XI. Controlling family (majority shareholders) versus 348 XI. Controlling family (majority shareholders) versus 9. External board members 1 2 3 4 5 6 7 NA XI. versus 10. CEO 1 2 3 4 5 6 7 NA Chairman B. Information of the company (group) surveyed B.1 Year of foundation or acquisition by the controlling family: ____________________ B.2 Who at present is at the head of management in the company (CEO)? ❑ A family manager From which generation? ___________ ❑ Nonfamliy manager Since when __________ B.3 Is the present Chairman a member of the owner family? ❑ Yes ❑ No ❑ NA B.4 Is the company: a Publicly Traded Corporation, a Privately Held Corporation or a Limited Liability Company? _____________________________________________ B.5 What is the number of permanent staff maintained by the company? ❑ up to 19. ❑ between 20 e 99. ❑ between 99 e 499. ❑ 500 or more. B.6 Into which annual invoicing band does your company fall? ❑ lower than R$250 thousand. ❑ between R$250 thousand and R$2.5 million. ❑ between R$2.5 million and R$60 million. ❑ between R$60 million and R$170 million. ❑ over R$170 million. 349 B.7 Regarding economic and financial performance, would you say that the results of your company are: ❑ lower than average for the sector to which it belongs. ❑ average for the sector to which it belongs. ❑ above average for the sector to which it belongs. C. Your data (optional) Name: Contact address: Post code / City / State: Phone / e-mail: Company where you work / position: Reply to the question below that identify the type of bond that you have with the company: C.1. Are you a member of the owner family? If yes, which generation do you belong to? ______ ❑ Yes ❑ No C.2. Are you a shareholder in the company? ❑ Yes ❑ No C.3 Are you a member of your companies’ Board of Directors? ❑ Yes ❑ No C.4 Do you work in the company? ❑ Yes ❑ No Thank you very much for completing this Survey! 350