Auditing standards and the expectations gap: evidence from Spain

Int. J. Critical Accounting, Vol. 4, Nos. 5/6, 2012 Auditing standards and the expectations gap: evidence from Spain Emiliano Ruiz-Barbadillo*, Nieves Gómez-Aguilar and Estibaliz Biedma-López Facultad de Ciencias Económicas y Empresariales, University of Cadiz, Duque de Nájera, 8, 110 Cádiz, Spain Fax: 00-34-956-015386 E-mail: [email protected] E-mail: [email protected] E-mail: [email protected] *Corresponding author Nieves Carrera Instituto de Empresas Business School, Pinar 15, 1st floor, 28006, Madrid, Spain E-mail: [email protected] Abstract: This article examines the impact of auditing standards intending to reduce the audit expectations gap on auditors’ behaviour. Specifically, we examine whether the adoption of an auditing standard ongoing concern evaluation in Spain is associated with changes in auditors’ behaviour. Our empirical results suggest that the implementation of the new standard did not impact on auditors’ propensity to issue going concern opinions; on auditors’ interest on maintaining their reputation; and on the influence of auditors’ economic incentives. Our results indicate that the implementation of auditing standards seems not to have an impact on day-to-day auditing practices, and therefore, auditing standards related to going concern evaluation fail to reduce the expectations gap. Our results also suggest that, in spite of the differences between the regulatory framework of auditing in Spain and in Anglo-Saxon countries (e.g., degree of state intervention), the impact of such standards on auditors’ behaviour are comparable. Keywords: audit report; going concern; auditing standards; expectations gap; auditing profession; auditor independence; auditor reputation; Spain. Reference to this paper should be made as follows: Ruiz-Barbadillo, E., Gómez-Aguilar, N., Biedma-López, E. and Carrera, N. (2012) ‘Auditing standards and the expectations gap: evidence from Spain’, Int. J. Critical Accounting, Vol. 4, Nos. 5/6, pp.483–510. Biographical notes: Emiliano Ruiz-Barbadillo is a Professor of Accounting at University of Cádiz, Spain. His research interest includes auditor independence and auditing and corporate governance. His publications have appeared in The Accounting History, The European Accounting Review, International Journal of Auditing, Accounting Forum, Journal of Management and Governance, Accounting, Auditing and Accountability Journal, Auditing: A Journal of Practice and Theory, and Journal of Business Ethics. Copyright © 2012 Inderscience Enterprises Ltd. 483 484 E. Ruiz-Barbadillo et al. Nieves Gómez-Aguilar is a Lecturer at University of Cádiz, Spain. She has authored numerous research articles on auditor independence and auditor change. Her publications have appeared in The European Accounting Review, International Journal of Auditing, Accounting Forum, Journal of Management and Governance, Accounting, Auditing and Accountability Journal, and Auditing: A Journal of Practice and Theory. Estibaliz Biedma-López is a Lecturer at University of Cádiz, Spain. She has authored numerous research articles on auditor independence and auditing and corporate governance. Her publications have appeared in Accounting Forum and Journal of Management and Governance. Nieves Carrera received her degree in Economics and her PhD in Business Administration and Quantitative Methods from University Carlos III. She was a Lecturer at Manchester University before she joined IE Business School in 2005. She has authored numerous research articles on the role of women in the auditing profession, on the rotation policy of auditing firms and on the structure of the audit market. Her publications have appeared in The European Accounting Review, Accounting, Auditing and Accountability Journal, and Auditing: A Journal of Practice and Theory. 1 Introduction The going concern concept is one of the key assumptions underlying financial reporting. An auditor’s responsibility to assess whether substantial doubts exist about a client’s ability to remain a going concern has been the subject of intense debate (e.g., Jones, 1996; Geiger and Raghunandan, 2002). It has been argued that this decision is one of the most complex and difficult situations faced by auditors (Louwers, 1998; Louwers et al., 1999). Although the auditing profession has gradually assumed this responsibility, there has been some debate about whether auditors are professionally qualified to make evaluations when there is reasonable doubt as to the continued existence of the client (Humphrey et al., 1992; Jones, 1996; Arnold et al., 2001). Moreover, it is argued that the auditors are no better placed than the users of financial statements to predict going concern uncertainties and, therefore, their evaluation of uncertainties is no better than the evaluation that financial statement users might make themselves [Jones, (1996), p.2]. Under these circumstances, an expectations gap emerges in which financial statement users expect more from an audit than the auditor expects to accomplish when taking on responsibilities to assess company bankruptcy risk. The expectations gap is one of the main issues confronting the auditing profession (Sikka et al., 1998). Different auditing approaches give different insights about the role of auditing regulation and how it may contribute to reduce the expectations gap. On the one hand, the postulate approach to auditing argues that the implementation of auditing standards and regulations should seek to benefit society (e.g., Mautz and Sharaf, 1961). On the other, perspectives of a more critical nature argue that auditing and accounting regulations are of benefit to auditors rather than society. From this perspective regulation merely becomes a symbolic gesture, in which benefits are more apparent than real. It takes the stance that auditing standards are not affecting auditing behaviour, instead merely acting upon external perceptions of auditors, while giving the impression that standards act in favour of the users of audit reports (e.g., Puxty et al., 1997). Auditing standards and the expectations gap 485 The effect auditing standards may have on reducing the audit expectations gap is an empirical question. Our study aims to explore the impact of new auditing standards on auditors’ behaviour. Specifically, we analyse the impact of a new auditing standard for going concern evaluation on auditors’ behaviour in Spain for the period 1991–1996. In Spain, the audit profession is regulated by the state and has some characteristics that differ from the widely studied Anglo-Saxon settings (Bougen, 1997; Bougen and Vázquez, 1997; García-Benau et al., 1999a). In May 1993, a new auditing standard called ‘Norma Técnica de Auditoría sobre el Principio de Empresa en Funcionamiento’ (NTA-May/1993; auditing standard on the going concern assumption; BOICAC, 1993) was published by the Instituto de Contabilidad y Auditoría de Cuentas (ICAC, Institute of Accounting and Auditing), the Spanish body responsible for the regulation of accounting and auditing. The new standard was designed to highlight going concern uncertainties and provide guidance for going concern evaluation. Before its adoption there were no detailed auditing standards on the evaluation of going concerns. Therefore, the issuance of the NTA-May/1993 created a new scenario, which enabled us to explore the question of whether the adoption of new standards impacts on auditors’ behaviour. Our analysis, based on a sample of companies quoted on the Spanish stock market, indicate that the new auditing standard did not modify the probability of auditors to issue a qualified going concern audit report for those companies with reasonable doubts about their continuity. Our results also suggest that the likelihood of auditors qualifying audit reports and disclosing going concern uncertainties is related to auditor reputation and company financial health. This investigation may be of interest for several reasons. Our study aims at enhance knowledge about the impact of new auditing standards on auditors’ behaviour. By doing this, we attempt to contribute to existing auditing literature about the effectiveness of auditing standards in reducing the audit expectations gap as well as investigating a relevant issue in the realm of practice. Regulators and professional bodies as well as users of financial information may benefit from an investigation on the impact that auditing standards on auditors’ behaviour specially when dealing with complex tasks such as the assessment of going-concern uncertainties. We expect to complement previous studies that have examined similar issues in other contexts (e.g., Carcello et al., 1995). Auditing research is dominated by studies focusing on Anglo-Saxon settings. By analysing the Spanish context, where the audit profession is regulated by the state (e.g., Bougen, 1997; Bougen and Vázquez, 1997; García-Benau et al., 1999a), we aim to provide further insights about the significance of auditing standards for auditing practice and the role of professional bodies in such an environment. Although the analysis will be context-specific, it will be of interest to a broader audience, given the recent pressures to reduce the degree of self-regulation in those countries where traditionally the profession has enjoyed a significant regulatory power (for example, in the UK and the USA). Our results will allow us to reflect upon in which extent the effectiveness of auditing standards in reducing the expectations gap in settings characterised by a higher involvement of the state (e.g., Spain) differs from the situation in those countries with a high degree of self-regulation. The remainder of the paper is organised as follows: The next section provides a literature review. In the third section we describe the Spanish context and analyse the expectations gap in Spain. The fourth section develops our research questions. Sections 5 and 6 describe the methodology and results respectively. The last section provides a discussion of our findings and the limitations of the study. 486 2 E. Ruiz-Barbadillo et al. Prior literature Albeit with a certain degree of reluctance, the auditing profession has gradually assumed the responsibility of evaluating a company’s ability to remain a going concern. This responsibility has been imposed through the issuance of several auditing standards [for example, in the USA, the Statement on Auditing Standards (SAS) No. 59, ‘The auditor’s consideration of an entity’s ability to continue as a going concern’, AICPA, 1988]. Primarily, these auditing standards are designed to describe the procedures to be followed when doubts arise on a company’s ability to remain a going concern. These standards require auditors to consider both the factors against the going concern assumption and the so-called mitigating factors. When evaluating the relative importance of both types of factors, the auditor must consider the underlying conditions from which all of the informative evidence is produced, and then analyse carefully the nature of the company’s business, the management’s plans and the conditions of the industry. The assessment process must enable the auditor to form a judgement concerning the company’s ability to continue its activities in the future. Auditing literature has paid special attention to the examination of auditors’ behaviour when they issue audit reports questioning their clients’ ability to continue to exist as a going concern. Since the seminal article of Altman and McGough (1974), many studies have provided valuable evidence of the auditing profession’s behaviour when facing going concern uncertainties (Hopwood et al., 1989; McKeown et al., 1991; Koh, 1991; Citron and Taffler, 1992; Hopwood et al., 1994; Mutchler et al., 1997). These studies suggest that auditors issue relatively few modified going concern audit reports for companies presenting observable risks of failing to remain a going concern. For example, only 15% of non-financial companies that went bankrupt between 1987 and 1994 in the UK were qualified on a going concern basis in their audit report prior to failure (Citron and Taffler, 2001). Conclusions drawn from these studies question the effective value of auditors’ work in relation to the evaluation of the going concern assumption. Users of audit reports do not understand the apparent reluctance of auditors to raise alarms when there are substantial doubts about an entity’s ability to continue. This issue has given substance to the concept of the expectations gap (Asare, 1990; Mutchler and Williams, 1990; Humphrey, 1997). The audit expectations gap refers to “the differences between what the public expects from an audit and what the auditing profession prefers the audit objectives to be” [Sikka et al., (1998), p.299]. The debate regarding the expectations gap usually arises after cases of legal actions against members of the profession, investigations by public bodies and general criticism concerning the profession’s behaviour. As Willmott (1991, p.109) notes, when doubts arise concerning the profession’s non-fulfilment of its social responsibilities, a debate ensues concerning a possible change of control of the regulatory mechanisms. This criticism is very dangerous for the auditing profession because, as Sikka and Willmott (1995) reveal, misgivings about the independence of auditors are doubly damaging to the profession because they threaten not only to devalue the material and symbolic value of a core area of expertise but also jeopardise auditors’ capacity to preserve and expand other lucrative growth areas, such as non-audit services. One of the measures adopted by regulators and professional bodies to face the auditing expectations gap is the codification of auditing knowledge through the issuance of auditing standards (Hooks, 1991; Martens and McEnroe, 1991; Fogarty et al., 1991; Humphrey, 1997). Auditing standards are designed to help auditors to evaluate material Auditing standards and the expectations gap 487 evidence rationally so they can reach an effective professional judgement. In other words, auditing standards provide operational guidance concerning the skills expected from auditors, the reasonable care with which they are expected to perform their duties and the professional tasks that must be undertaken (Willekens et al., 1996). From this perspective, new auditing standards play two different roles. On the one hand, standards specify what an auditor’s responsibility is with regard to each particular professional aspect and define issues upon which he/she must or must not take action1. On the other, standards become the basis for professional opinion-forming. By issuing auditing standards the profession attempts to show that they are ‘doing something’ to meet society’s expectations (Fogarty et al., 1991). Particularly in times of harsh criticism, there has been a definite pattern of increased standards issuance, in an attempt to signal change (Byington and Sutton, 1991). The persistence over time of the expectations gap raises questions about whether the issuance of auditing standards has the genuine objective of reducing this gap. These questions have been a subject of controversial debate, mainly because the empirical results are mixed. While some studies found that the relative frequency of modified audit reports increases after the adoption of new standards (de Fond et al., 2000), others researchers reveal that standards seem to have no effect upon auditors’ behaviour when an auditor evaluates a company’s risk of failing to remain a going concern (e.g., Carcello et al., 1995). In the USA, several studies have analysed the effect of SAS No. 59, ‘The auditor’s consideration of an entity’s ability to continue as a going concern’ (AICPA, 1988). This standard was designed by the Auditing Standard Board (ASB) to help close the gap between users’ expectations and auditors’ performance. While Raghunandan and Rama (1995) found that the standard contributed to reduce the gap, Carcello et al. (1995) found that “it does not appear that the gap between users’ expectations and audit firm reporting has been reduced” [Carcello et al., (1995), p.133]. Carcello et al. (1997) argue that a potential explanation for these contradictory findings is the different treatment of the data for the transition period (Carcello et al., 1997). Expectations gap concerns in the UK led to the introduction of SAS No. 600, ‘Auditors’ reports on financial statements’ and SAS No. 130, ‘The going-concern basis in financial statements’. Citron and Taffler (2004) found that SAS No. 600 caused an increase in reporting of going-concern uncertainties because of its less contentious nature compared with the previous auditing standard in the UK (SAS No. 58, ‘Reports on audited financial statements’). However, these authors did not find any impact of SAS No. 130 on auditors’ behaviour. These inconclusive results about the contribution of auditing standards to reduce the expectations gap have been used to make critical interpretations concerning the professions’ modus operandi and the attempts to reduce the expectations gap by means of strengthening self-regulation (e.g., Sikka et al., 1998). As noted by Abbott (1988), the ability of a profession to sustain its jurisdiction lies partly in the power and prestige of its academic knowledge because the public perceives a relationship between the abstract professional knowledge and effective professional. Therefore, the abstract knowledge serves to legitimate professional work [Abbott, (1988), pp.53–55]. In this sense, Citron (2003) argues that the development and eventual promulgation of an auditing standard can be viewed as an exercise of power by the profession in order to promote its own view regarding the nature and scope of ethical auditor behaviour. Lee (1995, p.59) states that the histories of the UK’s and the US’ 488 E. Ruiz-Barbadillo et al. standard-setting processes suggest a delicate procedure, managed by the professional bodies, of balancing economic self-interest against public interest. The issuance of auditing standards could be seen as a response to the profession’s desire to create an appearance of standardisation and the existence of a rational basis of experience (Humphrey et al., 1992). When an expectations gap arises, visible changes in the extent of standards publication may “provide the perception of significant change to external parties” [Byington and Sutton, (1991), p.318]. If the objective of standards is to influence users’ perceptions, its effect, according to Byington and Sutton (1991), will be less than expected because the only purpose of these standards will be to formalise traditional practices that the profession had already been using before the adoption of the new standards2. Such an effect is possible because auditing standards are designed, formulated and interpreted from the viewpoint of the profession itself (Zaid, 1997). The issuing of ‘ineffective’ standards has other objectives that serve as a signal, however. It has been stressed that the auditing profession issues auditing standards to minimise professional effort and maximise legal protection for its members (Gaa, 1991). For example, auditing standards may pursue the profession’s interest of minimising the risks of civil liability, without actual alteration of auditor behaviour. Auditors prefer to have a regulatory framework however inefficient, since the standards act as a benchmark for what is regarded as diligent behaviour and thereby limit the risk of adverse legal rulings (Sikka, 1992). Thus, the auditing profession channels the tensions they face into a rational strategy of ‘doing nothing’. In particular, they change the nature of auditing standards with regard to going concern assessment by implying the undertaking of more responsibilities without this entailing any alteration of day-to-day practices (Fogarty et al., 1991; Humphrey et al., 1992; Lee, 1995)3. The success of such a strategy rests on the institutional environment in which the auditing profession operates. Audit practice and perceived practices may be decoupled. That is, the practice of auditors as a profession may not be the same as that inferred by others’ readings and their reliance upon auditing standards [Chandler, (1997), p.414]. The abstract knowledge – or cognitive basis – of the auditing profession, however, does not provide individuals with clear, tangible symbols, meaning that the aims of auditing are intangible objectives whose achievement will always be open to interpretation (Fogarty et al., 1991). This environment provides favourable conditions for enabling the profession to separate its external image from its operating processes – and, therefore, “to the public’s mistaken belief that abstract professional knowledge is continuous with practical professional knowledge” [Abbott, (1988), p.54]. This in turn has enabled the auditing profession to issue reports as a ritual whose aim is to ‘paper over’ the existence of an expectations gap in order to legitimise the profession’s actions (Mills and Bettner, 1992). In summary, from a critical perspective auditing standards might lack the objective of solving the expectations gap problem (Williams, 1984; Neu, 1991; Hook, 1991; Carpenter and Dirsmith, 1993). In order to obtain empirical evidence on the effect of an auditing standard upon auditors’ behaviour we will analyse the auditors’ reporting decisions pre- and post-implementation of a particular standard in Spain. Next section validates the existence of the expectations gap in the context of our study, that is, the Spanish audit market. Auditing standards and the expectations gap 3 489 The audit expectations gap in Spain The 1988 Spanish Audit Law established for the first time statutory audit obligations for medium and large limited companies in Spain. The previous accounting and auditing regulatory framework had very negative effects upon the transparency and quality of the accounting information issued by companies (García-Benau and Humphrey, 1992) so the new regulation received a positive response by users of accounting information. While the Spanish auditing profession also welcomed this law because it institutionalised the auditing function and created a market in which they could perform their activity, the professional bodies fiercely criticised its apparently interventionist tone (Bougen and Vázquez, 1997; Bougen, 1997). The Spanish Audit Law established the ICAC, a body responsible to the Ministry of Economy, and which has considerable powers regarding the monitoring of auditors’ work. This agency is responsible for the approval and publication of auditing standards prepared by professional auditing bodies (García-Benau et al., 1993). According to Bougen (1997), the absence of an auditing tradition together with conflicts between professional associations of auditors4 that broke out during the debate of the auditing bill in the parliament, gave the government the necessary justification for a regulatory model where the ICAC and professional bodies shared different aspects of the regulation and monitoring of auditors. Some authors have argued that the interventionist tone of the law, which was widely contested by the auditing profession, was an indicator of the concerns of the government about the fulfilment of expectations by the auditing profession due to otherwise such extensive government control mechanisms would have been unnecessary (Bougen and Vázquez, 1997). The empirical evidence, however, suggests that at that time, the expectations placed on the profession were highly favourable (García-Benau et al., 1993; García-Benau and Humphrey, 1992). Only a few years after the regulation of auditing came into force, the atmosphere in the Spanish auditing environment was far from the optimism generated at the time of the launch of the Spanish Audit Law. Several financial scandals that led to wide social repercussions in the media (for example, the Banesto case) meant that in the early 1990s there was a gap between what the public expected from the auditing profession and what auditors were doing (García-Benau, 1994; Gonzalo-Angulo, 1995; Ruiz-Barbadillo, 1996; García-Benau et al., 1999a). The question ‘where were the auditors?’ and issues such as auditors’ responsibility to issue a going concern opinion when a business was about to fail became relevant on public debates. The controls of the ICAC on the profession and the subsequent sanctions imposed on some auditors5 (Amesti, 1995) reveal that only a few years after the legal recognition of the auditing function in Spain, the profession began to suffer a large loss of credibility (e.g., Lizcano-Álvarez, 1999). The Spanish auditing profession did not ignore these criticisms. At first, the auditing bodies argued that the expectations gap arose because the users of financial statements did not understand the scope, nature and objectives of the auditing function (see García-Benau et al., 1999a). As in many other countries, the Spanish professional associations attempted to protect themselves, arguing that the expectations gap was a natural and non-controversial issue and that the profession was just a victim of the financial scandals (Expansion 20 December 1994, Picazo, 1995). The profession could not, however, maintain its policy of shifting the blame for the expectations gap for long, mainly because the ICAC was threatening to introduce changes of a more interventionist 490 E. Ruiz-Barbadillo et al. nature in audit regulation (Ruiz-Barbadillo, 1996). Therefore the profession itself underwent several structural changes and a protracted debate ensued concerning the role that auditors should play with regard to companies’ going concern evaluations (El País 27 November 1994). There was at first a certain amount of resistance on the part of the profession’s members to accept responsibility for going concern evaluations (Blasco, 1994; Picazo, 1995; Marín, 1995). However, empirical research carried out by Prado (1993) indicated that in 1992, 57% of auditors opined that they should take on this responsibility. This study also reveals that auditors themselves were aware that social pressure would oblige them to take on greater degrees of responsibility in going concern evaluations [Prado, (1993), p.43]. At that time in Spain, there were no specific auditing standards ongoing concern evaluation. The Spanish Audit Law did not explicitly refer to the obligation of auditors to evaluate their clients’ risk of bankruptcy. Its Article 5, however, indicated that auditors must give their opinion if the non-application of one or more accounting principles is relevant to the faithful representation of the entity’s state of affairs that the financial statements must provide. Due to the going concern assumption is one of the accounting principles specified in the Spanish accounting legislation, Article 5 gave an implicit reference to auditors’ responsibility in relation to going concern assessment. In addition, Ley de Sociedades Anónimas of 1989 (Public Companies Act, 1989) in its article 209 stated that an auditor must mention in her/his report observations upon any fact that might represent a risk to the company’s financial situation. The Technical Auditing Standards issued by the ICAC in 1991 (ICAC, 1991) state in Section 1.5.3 that auditors must pay attention to those situations and circumstances that might bring into question the ability of the company to continue its activities. Although these rules state that auditors have the duty of alerting users of accounting information by reporting their clients’ risks of bankruptcy, none of them furnish detailed procedures for performing the task of going concern evaluation. Regulators and professional associations responded to the expectations gap debate by issuing a new auditing standard. Specifically, on 31 May 1993 the ICAC published a new standard called ‘Norma Técnica de Auditoría sobre el Principio de Empresa en Funcionamiento’ (thereafter ‘NTA-May/1993, auditing standard on the going concern assumption’, BOICAC, 1993). This standard aimed to provide guidance on the application of the going concern assumption. In particular, it specified the scope of an auditor’s responsibility when evaluating a company’s risk of failing to remain a going concern, simultaneously giving operational guidance, enabling the auditor to form her/his opinion upon this question. The change in the Spanish scenario caused by the adoption of the new standard NTA-May/1993 constitutes an appropriate research setting in which to investigate the impact of new auditing standards on auditors’ behaviour. Some comments should be done regarding the specificities of the Spanish audit market and Spanish audit regulation. In 1988, the Spanish Audit Law established that auditors had to be contracted for a period that must not be less than three years or greater than nine (the law was applied for the first time in 1990). The aim of such model was to avoid an excessively long contractual period between auditor and auditee that might, over time, reduce the auditor’s professional scepticism and impartiality (García-Benau et al., 1993; Prado et al., 1995; López-Combarros, 1996; Arruñada and Paz-Ares, 1997). In 1995, the profession’s lobbying succeeded in getting the rotation rule repealed (Gonzalo-Angulo, 1995; Amesti, 1995; Petit, 1995). After the removal of mandatory rotation, auditors were contracted on an annual basis, thus increasing Auditing standards and the expectations gap 491 companies’ ability to influence auditors’ behaviour. Given the fact that contracts are now on an annual basis, companies can change auditor simply by not renewing a contract. It is also necessary to analyse carefully the characteristics of the Spanish audit market. Although after the approval of the Spanish Audit Law the audit market was characterised by rapid growth, after 1993 the market was noticeably saturated. This stagnation led to significant competitive pressures, creating an environment where auditors feel under pressure to satisfy the interests of the auditee, who is able to hire and fire them at will (Paz-Ares, 1996). Unlike other European countries, Spanish legislation does give any safeguards to prevent auditees’ opportunistic behaviour when changing their auditors. In the context of a highly competitive auditing market and the ease with which audit firms can be replaced, auditors are at a negotiating disadvantage compared with company managers. Managers can easily solve any conflict arising in their relationship with auditors by switching to another audit firm. Once a change of auditor has occurred, the dismissed auditor has no right to be heard in the shareholders’ general meeting and explain her/his position. Moreover, there is no public document in which third parties are able to find the specific reasons given by the auditee for changing its auditor. In short, as empirical research has shown (Gómez-Aguilar and Ruiz-Barbadillo, 2003) the threat of auditor contract termination has had significant ramifications in the conduct of auditing in the Spanish context. Finally, with respect to the loss of reputation and litigation risk, during the first years of mandatory audits in Spain, the ICAC published the sanctions imposed on auditors, as well as the reasons why they had been sanctioned (García-Benau et al., 1999b). This, however, provoked concerns within the Spanish auditing profession, given the fact that the sanction files could be used to bring legal proceedings against auditors. The auditing profession undertook a vigorous and very successful lobbying campaign aimed at the regulatory bodies to prevent the publication of the files used to sanction auditors for negligent behaviour (Ruiz-Barbadillo et al., 2000; Humphrey et al., 2003). The result of this campaign was that the profession’s behaviour became opaque, making civil liability proceedings extremely complex. Moreover, the Spanish system establishes that in civil liability cases, the affected party must demonstrate negligence on the part of the auditor. This is practically impossible in cases against auditors, given the opacity of auditing processes and practices. Thus, even though there have been numerous financial scandals in which auditors have been involved to date, there has been no condemnatory sentences against auditors (Ruiz-Barbadillo et al., 2000). However, this low litigation risk context is very interesting for the purpose of our study because it enabled us to measure directly auditors’ independence without any interference or pressure influencing their decisions. 4 Research questions From a normative point of view, the goal of auditing standards is to ensure a uniform level of quality when providing auditing services and a benchmark of the minimum level of acceptable professional competence. In this situation, a change in auditing standards or the implementation of new standards would entail a change in the objectives, procedures and techniques of auditing and, therefore, a change in auditors’ behaviour should be expected. The aim of NTA-May/1993, the new auditing standard dealing with going concern evaluations, was to enable the auditor to identify those companies suffering going concern problems. We assume that the degree to which auditing standards related 492 E. Ruiz-Barbadillo et al. to going concern evaluations are fulfilled must be related to the detection of companies in financial distress. The new audit standard implemented by the ICAC not only defines auditor responsibility but also describes the process of evaluation of an entity’s ability to continue its activities. Ceteris paribus, the implementation of NTA-May/1993 was expected to increase the probability of auditors detecting and reporting going concern uncertainties. We hypothesised that the relative frequency of modified audit reports would show an increase after the adoption of the new standard. We analysed the following research question: RQ1 Does the implementation of NTA-May/1993 increase the likelihood of auditors’ issuing a qualified going concern audit report to distressed companies? Authors such as Wilkerson (1987), Raghunandan and Rama (1995) and Reynolds and Francis (2001) have shown that professional audit behaviour when issuing qualified audit opinions cannot be seen as a simple process of evaluating evidence. There are economic incentives associated with the type of opinion given by the auditor that are inextricably bound to the auditor’s reporting decision. These are related to the costs that must be assumed by the auditor in the event of losing the client (if the client disagrees with the type of opinion received) and litigation costs and loss of reputation (in those cases where the auditor is found to be issuing a more favourable opinion than the client deserves). It is relevant to analyse whether the implementation of the new standard NTA-May/1993 had an impact upon auditor’s economic incentives. Abbott (1988) states that it is essential for an occupational group to control its abstract system of knowledge in order to maintain its professional stature. In the case of auditing, its power rests in its vagueness: the disconnection between what is tested and what effects are achieved from this testing (Power, 1997). The issuing of an auditing standard has a significant impact upon the abstract system of auditing knowledge. As an operational guide, the standard defines the audit process and the techniques that must be applied by the auditor. This standardisation demystifies the process, for example, when considering the evaluation of whether a company is a going concern, thereby eliminating any abstract or vague features. Covaleski et al. (2003) found that once knowledge becomes standardised, a profession can no longer earn high profits through its application. If higher professional standards result in a higher reputation either for individual firms or for the profession as a whole [Gaa, (1991), p.86] differentiation through reputation becomes more difficult. Therefore, standardisation of auditing procedures makes more difficult to earn returns through strategies of differentiation and, as a result, auditing firms would expect to obtain lower returns from investments in reputation. RQ2 Does the implementation of NTA-May/1993 diminish auditors’ interest in their reputation? As mentioned above we expect that the standardisation of the audit process will make easier to detect when the auditor has not fulfilled her/his obligations. Importantly, they may be used in courts to determine whether auditors have used reasonable skill and care in carrying out their duties [Chandler, (1997), p.414]. The gradual exposure of the accountancy and auditing body of knowledge through conceptual statements, standards and guidance recommendations have removed some of the mystique of accountancy, and made it easier for non-accountants to criticise the professional practice of accountants Auditing standards and the expectations gap 493 (Lee, 1995). Regulation risks cause ‘defensive auditing’ with auditors using only hard evidence (information which is observable and verifiable for third parties) to support their opinions. As a result, audits can become trivial (Arruñada, 2000). When the risk of litigation becomes high, auditors have more incentives to behave and to be seen behaving independently. From this perspective we would expect that if the economic dependence on a particular client influences on auditors’ decision to issue a qualified audit report, such an influence should be lower after the issuance of a new standard ongoing concern assessment. On the other hand, higher professional standards may result in higher cost to auditing firms of performing audits. These costs may be partially reduced by passing them on to auditees, for example, through subsidisation by fees from non-audit services [Gaa, (1991), p.86]. As a result, we could expect more competition in the audit market. From this point of view, we would expect that auditors will become more dependent on the revenues generated by existing clients after the implementation of a new auditing standard. Thus, under the assumption that the economic dependence affects auditors’ likelihood to issue a qualified going concern audit report, we would expect an increase of the economic dependence on existing clients after the issuance of the new standard. The following research question attempts to test such a relationship. RQ3 5 Does the implementation of NTA-May/1993 influence on the economic dependence of auditors on their existing clients? Data and methods 5.1 Sample We examined a sample of non-financial companies quoted on the Spanish stock market during the years 1991–1996, inclusive. The initial sample consisted of 2,070 observations of companies. Empirical evidence suggests that auditors do not usually issue qualified reports for non-stressed companies that suddenly fail (McKeown et al., 1991; Geiger and Raghunandan, 2001). Therefore, it is important to perform the analyses using a sample of companies with similar underlying uncertainties that might be the reason for a modified going concern opinion (Wilkerson, 1987). Consistent with prior research (Hopwood et al., 1994; Geiger and Raghunandan, 2001), we selected our sample using the following indicators: negative working capital; a loss from operations in two consecutive financial years; and stockholders’ equity being lower than the capital figure 6. After the application of these criteria, our final sample consisted of 412 observations, of which 123 (30%) refer to the period before the issuance of the audit standard (1991–1992), while 289 (70%) refer to the subsequent period (1993–1996). We identified 65 qualified going concern reports (16%), of which 23 were issued in the pre-standard period and 42 in the post-standard period. 5.2 Dependent variable and experimental variables To study the impact of NTA-May/1993 on the behaviour of the auditing profession, we examined audit reports of companies both before and after the new standard. The 494 E. Ruiz-Barbadillo et al. dependent variable in our model is the type of audit opinion. The dichotomous variable GC was coded ‘1’ if the audit opinion was modified for going concern and ‘0’ otherwise. The variable of interest to examine the first research question is STANDARD. This is a dichotomous variable that differentiates the period before and after the adoption of the new auditing standard. The NTA-May/1993 standard was issued on 31 May 1993 and became effective in the fiscal year 1993. Therefore, the variable STANDARD was coded ‘1’ for the pre-standard period (1991–1992) and ‘0’ for the post-standard period (1993–1996). This variable enabled us to analyse whether the likelihood of receiving a modified going concern report on a distressed company changed subsequent to the adoption of the auditing standard. de Angelo (1981) argues that large audit firms have more incentives to avoid reputation-damaging criticism compared to small audit firms. This is because large firms have more audit clients and, therefore, have more to lose from loss of reputation. Since the failure to issue a going concern opinion when this is justified can harm the auditor’s reputation, we expect a positive relationship between cost of reputation loss and the auditors’s probability of issuing a going concern opinion. The variable AUDITORSIZE is used as a proxy to capture the reputation effect. As in previous studies, AUDITORSIZE is calculated by total sales of the clients of an auditor divided by total sales of clients of the whole audit market (Francis and Wilson, 1988). The interaction of the variable AUDITORSIZE with STANDARD will allow us to examine the second research question of this study. We expect that auditors’ concern for the upholding of their reputation decreases after a standard is issued because of the standardisation of the audit knowledge. After that occurs, we expect that the reputation interaction term (STANDARD – AUDITORSIZE) will show a negative relation with the dependent variable, diminishing the value of the coefficient of the variable that represents auditor’s reputation. A variable related to auditor economic dependence is included in the model (DEPENDENCE). The auditor may lose the contract if a qualified audit report is issued (Craswell, 1988). The greater the size of the client, the greater the quasi-rents that the auditor obtains and therefore the more significant is the loss of the contract with the client. We expect a negative relationship between the relative size of the client and the reception of a going concern evaluation. The variable DEPENDENCE was calculated as follows: client sales divided by sales revenues of all audit clients. The interaction of DEPENDENCE with the variable STANDARD will allow us to examine the third research question. We expect that the interaction term related to auditors’ economic dependence of a client (STANDARD – DEPENDENCE) will show an association with the dependent variable. A priori we are not able to determine the sign of the relationship. On the one hand, after the issuing of a going concern standard an auditor is more likely to be exposed to litigation, making independence more important in reporting decisions and, as a result, the auditor will be less likely to behave in a non-independent manner. On the other, the issuance of an auditing standard may increase the competition in the audit market, making auditors more dependent on the revenues generated by their existing clients. Auditing standards and the expectations gap 495 5.3 Control variables The probability of receiving a qualified going concern audit report is also affected by the client company’s financial situation. Thus, the going concern evaluation of companies is clearly associated with the economic cycle (Rosman et al., 1999). The inclusion of control factors enabled us to manage the companies’ economic and financial situation when the sample is divided into pre-and post-standard periods. The control variables are as follows. The first variable is a proxy of the company’s losses. A company that reports accounting losses has a higher probability of receiving a qualified going concern opinion (Reynolds and Francis, 2001; de Fond et al., 2002). We expect a positive association between the variable LOSS and the likelihood of issuing a qualified going concern opinion. The variable DISTRESS tries, through a bankruptcy prediction model7, to calculate the company’s probability of bankruptcy. It could be expected that the higher the likelihood of failure, the more straightforward the judgement would be as to whether a client is suffering going concern problems. The variable CLIENTSIZE was included as a proxy of a company’s size. Craswell et al. (2002, pp.260–261) note that litigation costs that may be incurred after a large client fails have the potential to condition auditors to be conservative in their opinion. Default status has been found to be a powerful predictor of qualified going concern opinion (Raghunandan and Rama, 1995). Different studies have found that covenant violations are positively associated with receiving a qualified going concern opinion (e.g., de Fond et al., 2002). Thus, we introduced a fourth control variable, LEVERAGE, to capture proximity to covenant violations because firms close to violation are likely to have high leverage [de Fond et al., (2002), p.1257]. LEVERAGE is measured as the ratio of total debt to total assets. Finally, McKeown et al. (1991) show that the greater the lag between financial year end and the audit report date, the greater the likelihood of auditors issuing qualified going concern opinions. Firstly, audit report lag may indicate auditor–client opinion negotiations; managers and auditors may engage in prolonged negotiations when there are going concern uncertainties. Secondly, audit report lag may be indicative of the extent of the auditor’s test work, with auditors being more likely to discover financial distress when they carry out extensive testing (McKeown et al., 1991; Raghunandan and Rama, 1995; Carcello et al., 1995; Louwers, 1998). AUDITLAG is calculated as the number of days between fiscal year end and the audit report issue date. To sum up, we expect to find a positive relationship between the probability of the firm receiving a qualified going concern opinion and LOSS, DISTRESS, CLIENTSIZE, LEVERAGE and AUDITLAG. 5.4 Models To test the impact of the auditing standard on the likelihood of receiving a modified going concern audit opinion we used the following logistic regression model: ⎛ STANDARD, LOSS , DISTRESS , CLIENTSIZE , LEVERAGE , ⎞ P (GC = 1) = f ⎜ ⎟ ⎝ AUDITLAG , AUDITORSIZE , DEPENDENCE ⎠ 496 E. Ruiz-Barbadillo et al. where GC = 1 if audit opinion was modified for going concern; 0 otherwise STANDARD = 1 if the financial year is 1993, 1994, 1995 and 1996, 0 otherwise LOSS = dichotomousvariable coded 1 if the client reported loss in current or prior years; 0 otherwise DISTRESS = probability of failure calculated using Zmijewski (1984 ) coefficients LEVERAGE = ratio of total debt to total assets CLIENTSIZE = natural logarithm of total assets AUDITLAG = number of days between fiscal year end and the audit report date AUDITORSIZE = total sales of clients of an auditor divided by total sales of clients of the audit market DEPENDENCE = client sales divided by sales revenues of all audit clients. In this model, we introduced two interaction terms to test the second and third research questions. To test the impact of the auditing standard on auditors’ economic incentives we use the following logistic regression model: ⎛ STANDARD, LOSS , DISTRESS , CLIENTSIZE , ⎞ ⎜ ⎟ P (GC = 1) = f ⎜ LEVERAGE , AUDITLAG, AUDITORSIZE , DEPENDENCE , ⎟ ⎜ STANDARD − SIZE , STANDARD − DEPENDENCE ⎟ ⎝ ⎠ where, besides the previous defined variables: STANDARD – AUDITORSIZE = interaction term STANDARD × AUDITORSIZE STANDARD – DEPENDENCE = interaction term STANDARD × DEPENDENCE 6 Results 6.1 Univariate analysis In this section, we present univariate results concerning the variable STANDARD and the variables of interest for our research question. Table 1 describes the contingency table for STANDARD and the dependent variable (GC), that is, the type of opinion. Table 1 Contingency table* Report period Clean report GC qualified report Total Pre-standard (1991–1992) 100 (81.3%) 23 (18.7%) 123 (100%) Post-standard (1993–1996) 247 (85.5%) 42 (14.5%) 289 (100%) 347 (84%) 65 (16%) 412 (100%) Total Note: *percentages in rows Twenty three of the 123 (18.7%) companies in our sample for the period before the adoption of NTA-May/1993 received a qualified going concern audit report. For the period subsequent to the implementation of the new standard (1993–1996), 42 out of 289 Auditing standards and the expectations gap 497 companies (14.5%) received such a report. In order to analyse the existence of association or dependency between the temporal variable and the type of audit report, we apply the chi-square test of independence. The null hypothesis that there is no relationship between the issuing of the audit standard and receiving a qualified audit report cannot be rejected (χ2 = 1.127, p = 0.288). Statistically, both variables are independent; therefore, it seems that the issuing of the standard did not lead to a greater percentage of distressed companies receiving a going concern qualification. At a univariate level the results suggest that the ICAC resolution NTA-May/1993 concerning the going concern principle has not impacted upon the auditors’ decision on whether to issue a modified going concern audit report. Table 2 presents the percentage of qualified going concern audit reports issued by different audit firms for the two sub-periods (pre- and post-standard). Auditing rules standardise issues such as the definition of criteria for identifying companies that have problems with remaining a going concern, and the specific evaluation procedure for evaluating the underlying risks to a company’s activities. Auditors are obliged to comply with these rules, so it could be argued that their behaviour will tend to become uniform. However, previous research has found that auditors do not alter their behaviour subsequent to the adoption of new auditing standards (Carcello et al., 1995, 1997). In order to analyse whether there is an association between an individual auditor’s propensity to issue a qualified report and the issue of NTA-May/1993, the chi-square test of independence was performed. The objective was to test whether the auditor’s propensity to issue a qualified audit report is not associated with the date of the adoption of the new auditing standard (that is, whether it fell before or after the adoption of the new auditing standard). Table 2 Auditors’ propensity to issue qualified going concern reports in the pre- and post-standard periods* Pre-standard (1991–1992) Post-standard (1993–1996) Number of qualified reports (%) Number of qualified reports (%) AA 10 (18) 9 (8) 4.189 (0.041) CL 3 (60) 2 (13) 4.356 (0.037) DT 2 (33) 5 (42) 0.117 (0.732) EY 3 (27) 9 (23) 0.085 (0.774) KPMG 3 (13) 6 (14) 0.011 (0.918) Auditor χ2 (p-value) PW 0 (0) 4 (19) 1.768 (0.184) OTH 2 (15) 7 (17) 0.092 (0.761) TOTAL 23 (19) 42 (14) 1.127 (0.288) Notes: Arthur Andersen (AA); Coopers & Lybrand (CL); Deloitte & Touche (DT); Ernst & Young (EY); KPMG Peat Marwick (KPMG); Price Waterhouse (PW); other auditors (OTH) As mentioned above, NTA-May/1993 did not alter auditors’ propensity to qualify audit reports (χ2 = 1.127; p = 0.288). However, results differ when the propensity of individual audit firms to issue a qualified going concern audit report is analysed. AA issued qualified reports for 18% of its financially distressed clients in the period 1991–1992 (before the auditing standard NTA-May/1993). This percentage dropped to 8% after the standard (1993–1996). Moreover, the chi-square results indicate that there is a 498 E. Ruiz-Barbadillo et al. relationship between the period under investigation and the propensity of AA to issue qualified going concern audit reports. Similar results are found for CL, with the percentage of clients with qualified audit reports decreasing significantly from the period 1991–1992 to the period 1993–1996. The chi-square test for the remaining auditors was not significant; therefore, the null hypothesis of independence between the issuance of the new standard and the issuance of modified going concern audit reports cannot be rejected. It is worth noting that DT, KPMG and OTH increased the percentage of qualified audit reports for distressed clients, while the opposite happened in the case of EY. Several factors justify these mixed findings. First, auditing standards suggest only a single, partial methodology to follow, and one that on occasions can be both ambiguous and too general [e.g., Gonzalo-Angulo, (1995), p.620]. Second, an auditor is solely responsible when judging her/his client’s ability to remain a going concern. Thus the standards will always need an interpretation process, which justifies the existence of different patterns of behaviour among different professionals (Gibbins and Mason, 1988). Thirdly, the use of standards-based auditing procedures increases auditing costs (Byington and Sutton, 1991), including staff training and the development of specific procedures within the audit work programme. The cost of applying the standards will affect audit firms in different ways, depending on their size and their client portfolio. Indeed, assuming that the increase in costs can be transferred entirely to clients, the smaller the auditor’s client portfolio, the greater will be the transferred costs borne by clients. In short, the standards could entail a loss of competitive advantage among auditors with a small client portfolio. Such auditors will therefore be less inclined to fulfil the requirements established in the standards. We can therefore speculate that their judgement will not be as accurate. Finally, another factor that may explain differences between auditors is the technology employed by the audit company (Kinney, 1986; Cushing and Loebbecke, 1986). Empirical evidence reveals that an auditor’s judgement is affected by the level of technology employed in their operating approach: the less structured a firm’s auditing process is, the more important the individual auditor’s judgement becomes (Newton and Ashton, 1989; Mutchler and Williams, 1990). This therefore leads to differences in the propensity of different auditors to qualify clients in similar circumstances. 6.2 Multivariate analysis Table 3 reports the results of the first logistic regression model. It describes the estimated coefficients for individual variables and several indicators of the goodness of fit of the model. The chi-square test was used to test the statistical significance of the model. The test indicated that the model is highly significant (χ2 = 100.67; p = 0.00). The model demonstrated relatively good explanatory power, with a pseudo R2 = 0.27 and the level of correct classifications at 70.60%. Our first research question attempts to analyse whether the implementation of a new auditing standard ongoing concern evaluation had an impact on the likelihood of auditors to issue a qualified audit report. Therefore, we are primarily concerned with the statistical significance of the coefficient on STANDARD. The results show that this variable is not significant (α = 0.05). Ceteris paribus, this finding involves that the implementation of the auditing standard NTA-May/1993 did not have any impact on the likelihood of auditors to issue a qualified audit report. Auditing standards and the expectations gap Table 3 499 Logistic regression results Variable Expected sign Coefficient (Wald’s test) p-value STANDARD + 0.260 (0.801) 0.371 LOSS + 2.863 (13.887) 0.000 DISTRESS + 1.173 (4.126) 0.042 LEVERAGE + 1.315 (3.746) 0.053 CLIENTSIZE + 4.723 (1.855) 0.173 AUDITLAG + 0.005 (2.617) 0.106 AUDITORSIZE + 1.261 (4.242) 0.039 DEPENDENCE – –0.585 (1.086) 0.297 Intercept ? –12.588 (8.338) 0.004 Notes: χ2 = 100.667; p = 0.000; pseudo R2 = 0.265; % correct classification: 70.6%; Variable definitions: STANDARD = 1 if the financial year is 1993, 1994,1995 and 1996, 0 otherwise LOSS = dichotomous variable coded 1 if the client reported loss in current or prior years; 0 otherwise DISTRESS = probability of failure calculate dusing the coefficients of Zmijewski (1984) LEVERAGE = ratio of total debt to total assets CLIENTSIZE = natural logarithm of total assets AUDITLAG = number of days between fiscal year end and the audit report date AUDITORSIZE = total sales of clients of an auditor to total sales of clients of the audit market DEPENDENCE = client sales divided by sales revenues of all audit clients The proxy of auditors’ reputation (AUDITORSIZE) has a positive (1.26) and significant coefficient (α = 0.05). This result suggests that ceteris paribus audit firms with considerable investment in reputation show a higher propensity for issuing qualified going concern audit reports. In doing so, they try to protect themselves from the consequences of being found to report more favourably on their client than is justified. The variable DEPENDENCE, proxy for auditors’ economic dependence upon a particular client, is not statistically significant. Our findings suggest that in the case of companies with distress symptoms the loss of a client is not an important consideration for the auditor. A potential explanation for this lack of significance for the economic dependence of auditors could be that, given the characteristics of these companies, which have a high probability of failure, their auditors know that, eventually, they will lose their quasi-rents. From an economic point of view auditors may consider that the risk of issuing a clean audit report for a distress company is higher than the risk of losing such a client after considering all the present and future returns, from audit and non-audit services, expected to gain from such a client. With the exception of CLIENTSIZE and AUDITLAG, the remaining control variables are all significant (see Table 3). The coefficient for the variable LOSS is positive (2.86) and significant (p-value = 0.00). The positive sign of its coefficient was expected, as companies reporting losses have more probability of receiving a modified going concern report. The variable DISTRESS shows similar results being its coefficient positive (1.26) 500 E. Ruiz-Barbadillo et al. and significant at α = 0.05. As expected, our results suggest that the greater a company’s probability of bankruptcy, the greater the probability of receiving a qualified going concern report. The coefficient for the variable LEVERAGE is also significant and positive as expected suggesting that the greater a company level of indebtedness, the greater its probability of receiving a qualified going concern report. The remaining control variables (CLIENTSIZE and AUDITLAG) are not significant; that is, our results indicate that client complexity and audit report lag do not impact on the likelihood of receiving a qualified audit report. Logistic regression with interaction terms results Table 4 Expected sign Coefficient (Wald’s test) p-value STANDARD Variable + 0.609 (1.641) 0.200 LOSS + 2.853 (13.839) 0.000 DISTRESS + 1.173 (4.126) 0.042 LEVERAGE + 1.263 (3.304) 0.069 CLIENTSIZE + 4.914 (1.992) 0.158 AUDITLAG + 0.005 (2.637) 0.104 AUDITORSIZE + 2.267 (5.426) 0.020 DEPENDENCE – –0.636 (0.424) 0.515 STANDARD – AUDITORSIZE – –1.578 (1.743) 0.187 STANDARD – DEPENDENCE + 0.104 (0.008) 0.930 Intercept ? –13.019 (8.802) 0.003 Notes: χ = 102.851; p = 0.000; pseudo R = 0.271; % correct classification: 73.2%; Variable definitions: STANDARD = 1 if the financial year is 1993, 1994,1995 and 1996, 0 otherwise LOSS = dichotomous variable coded 1 if the client reported loss in current or prior years; 0 otherwise DISTRESS = probability of failure calculate dusing the coefficients of Zmijewski (1984) LEVERAGE = ratio of total debt to total assets CLIENTSIZE = natural logarithm of total assets AUDITLAG = number of days between fiscal year end and the audit report date AUDITORSIZE = total sales of clients of an auditor to total sales of clients of the audit market DEPENDENCE = client sales divided by sales revenues of all audit clients STANDARD – AUDITORSIZE = interaction term STANDARD × AUDITORSIZE STANDARD – DEPENDENCE = interaction term STANDARD × DEPENDENCE 2 2 To sum up, our results from the logistic regression model without interactions indicate that implementation of the new auditing standard on the going concern assumption (NTA-May/1993) did not have any impact on the likelihood of distressed firms receiving a qualified going concern audit report. Company financial distress indicators such as the losses in the previous years, the probability of failure and the level of leverage of the Auditing standards and the expectations gap 501 company as well as the proxy for auditor reputation seem to affect the likelihood of auditors issuing a qualified audit report for a sample of distressed companies. Table 4 shows the results of testing the logistic regression model with interactions. This model is used to test research questions second and third of this paper. Statistically, this model shows a better adjustment than the previous model (χ2 = 102.85; p = 0.00), with a pseudo R2 of 27% and the level of correct classifications at 73.2%. Results are similar to those reported in the previous model. To test whether the relevance of reputation for auditors change before and after the new standard was implemented and whether the economic dependence of auditors varies once the new auditing standard was implemented we behaviour of auditors in relation to the research questions we analyse the coefficients of the interaction terms STANDARD – AUDITORSIZE and STANDARD – DEPENDENCE. The coefficients of both interaction terms are not statistically significant. Specifically, the lack of significance of the coefficient for STANDARD – AUDITORSIZE suggests that the behaviour of auditors in relation to the relevance of reputation for auditors did not vary from the period previous to the implementation of NTA-May/1993 and after its implementation. Therefore, in relation to our second research question the results suggest that the implementation of NTA-May/1993 did not have an impact on auditors’ incentives related to their reputation. As reported in the model without interactions, this model also indicates that the economic dependence of auditors does not impact of auditors’ likelihood to issue a qualified audit report. Consistently with this result, we find that the coefficient for the interaction term STANDARD – DEPENDENCE is not significant. Therefore, in relation to our third research question there is no impact of the implementation of the new standard on surprising. While the model suggest that the reputation of auditors impacts on the likelihood of auditors to issuing a going concern audit opinion, the relevance of reputation does not vary when we differentiate the period before and after the implementation of NTA-May/1993. Regarding the economic dependence of auditors, results for the interaction terms are consistent with the lack of significance of the economic dependence of auditors as a predictor of the likelihood of issuing a qualified audit report. Further discussion of these results is provided in the next section. 7 Discussion and conclusions This paper analyses the effects of a new auditing standard related to going concern evaluation on Spanish auditors’ behaviour. Even though the debate on the expectations gap has not been as intense in Spain as in other countries (for example in the UK), comparable reactions by the Spanish profession and regulators can be found. The audit expectations gap refers to “the differences between what the public expects from an audit and what the auditing profession prefers the audit objectives to be” [Sikka et al., (1998), p.299]. The standard NTA-May/1993 implemented by the ICAC can be regarded as the response of regulators and the auditing profession to the emergence of the audit expectations gap in Spain. The new standard attempted to minimise the gap between what the public expect auditors to achieve and what auditors can reasonable be expected to accomplish – in other words, to reduce the reasonableness gap (Porter, 1993) – by increasing the profession’s responsibility on reporting going concern uncertainties and outlined a specific process to follow when evaluating a company’s ability to continue its 502 E. Ruiz-Barbadillo et al. activities. With the implementation of a standard ongoing concern evaluation the profession showed society what they can reasonable expect from auditors’ work without this necessary meaning that auditors modify auditors’ day-to-day practices. Our results suggest that the implementation of the standard NTA-May/1993 did not affect whatsoever upon the likelihood of auditors to issue qualified going concern audit reports. Therefore, in relation to the effectiveness of the new auditing standard our findings suggest that it does not appear that the gap between users’ expectation and audit firm reporting has been reduced as a result of the implementation of the new standard ongoing concern uncertainties. Our results are consistent with those reported by Carcello et al. (1995) when analysing the impact of SAS No. 59 in the USA. Our empirical results also suggest that the implementation of NTA-May/1993 did not affect auditors’ interest in their reputation. In particular, our model indicates that while reputation is a significant variable to explain auditors’ likelihood to issue a going concern audit report, no change was observed after the implementation of NTA-May/1993. Therefore, although it could be expected that auditors were less concern about their reputation due to the standardisation of audit knowledge (Gaa, 1991) no such an effect was found in our model. This result also suggests that the audit market exerts a more effective control over the auditing profession than legal mechanisms do. The auditing profession claims that external regulation is unnecessary because the audit market will pressure auditors to be seeing as behaving independently – that is maintaining their reputation. Given that the auditing standard analysed in this paper has not had any effect on the Spanish auditing profession’s behaviour in relation to their interest in reputation, our results seem to support this argument. In relation to the relevance of the economic dependence of auditors when considering their reporting decisions our results indicate that the economic dependence of auditors, defined as the importance of an audit client in its auditors’ portfolio, did not affect the likelihood of auditors to issue qualified audit reports. Moreover, the implementation of the new auditing standard did not impact on the economic dependence of auditors. While previous studies have found that the economic dependence of auditors influences on auditors’ reporting decision (e.g., Krishnan and Krishnan, 1996; Vanstraelen, 2002), our results do not confirm such a result. Some characteristics of the Spanish audit market such as the existence of the mandatory audit rotation policy from 1990 to 1995 may explain the lack of significance of such variable. Economic dependence may be a less important factor in an environment where the auditor makes decisions knowing that after a number of years its clients are required to change their auditors by law. From a normative point of view, the lack of significance of such a variable could be interpreted as an indicator of auditors’ independent behaviour: Their reporting decisions seem not to be affected by the importance of a particular client in her/his portfolio of clients. Finally, our model indicates that variables related to the company’s financial situation are predictors of the probability of issuing a qualified going-concern audit report. Taken together our results revel that the implementation of NTA-May/1993 in Spain did not have an impact on auditors’ behaviour. These findings lead us to make some observations regarding the role of auditing standards, the profession’s behaviour and the expectations gap. Auditing standards related to the going concern assumption such as NTA-May/1993 in Spain, SAS No. 59 in the USA or ISA 570 Going Concern issued by the International Auditing Practices Committee (IAPC) merely establish a process, sometimes in very ambiguous terms, to evaluate the circumstances of risk of bankruptcy (see for instance Auditing standards and the expectations gap 503 Carcello et al., 2003). These standards usually refer to the large volume of information that an auditor must gather when facing uncertainties about the client’s ability to continue as a going concern. Even though the determination of a wide range of indicators of going concern uncertainties can be relatively easy, determining an evaluation process that takes into account all of the circumstances can be an extremely complex task. These standards do not usually provide a relationship between causes (factors) and effects (discontinuity), thus leaving the auditor to make these links (Koh, 1991). Moreover, crises in companies are a complex phenomenon and often unpredictable because not all companies present observable symptoms of financial distress (Hopwood et al., 1994). In this regard, an auditor may end up taking on excessive responsibilities in a situation where auditing standards are unhelpful in the identification of companies that may fail to remain a going concern. Therefore, as Willekens et al. (1996) indicate, the issuance of an auditing standard does not necessarily lead to an improvement in the quality of auditors’ work. The standard’s effectiveness and capacity will depend on the detail of specification within the standard itself and the level of firmness established a posteriori with regard to their fulfilment. In Spain, the ICAC resolution NTA-May/1993 is deliberately ambiguous and does not provide a precise relationship between the causes of a company’s crisis and the decision that the auditor must make (Gonzalo-Angulo, 1995; Guiral, 2003). As our results suggests, no changes were observed before and after the implementation of the auditing standard. Prior studies have found comparable characteristics and results when analysing auditing standards adopted by the auditing profession and regulators in other countries (e.g., Carcello et al., 2003). The lack of definition and ambiguity characterise the process of elaboration, adoption and implementation of auditing standards. When facing an expectations gap, the auditing profession has always had a considerable interest in monitoring the debate concerning others’ expectations of its work output [Humphrey et al., (1992), p.139]. From a critical perspective, auditing standards might be seen as designed to achieve objectives other than those of solving the expectations gap problem (Williams, 1984; Neu, 1991; Carpenter and Dirsmith, 1993). In particular, the adoption of auditing standards could be seen as part of the strategy adopted by the accounting profession to monopolise expertise (Hooks, 1991; Fogarty et al., 1991; Humphrey, 1997). Successful claims to esoteric knowledge have important outcomes, namely professional authority and autonomy. In other words, professionals involved have the right to use their specialised knowledge in a manner that they see as most appropriate. Moreover, the profession modifies its knowledge system in response to social demands. As in other continental European contexts (for example, for Belgium see de Beelde, 2002), in Spain the state plays a significant role in the definition of auditing. In the Spanish Audit Law (1989), the government considered auditing not as a profession along the lines set for the established professions of lawyers or doctors8 but as an activity, monitored and regulated exclusively by the state. The law created a governmental body, the ICAC, responsible for the regulation of accounting and auditing. This model differs significantly from the situation in Anglo-Saxon countries where the accounting profession has traditionally enjoyed a high degree of self-regulation, autonomy and self-control and where the laissez-faire approach has left the field open for action by the accountancy bodies [Puxty et al., (1997), p.327]9. Auditors need to be able to define the boundaries of ethical behaviour effectively, in order to keep their powers of self-regulation. The accounting profession tries to avoid having regulation imposed by government due to government legislation changes 504 E. Ruiz-Barbadillo et al. unethical behaviour into illegal behaviour [Brien, (1998), p.392]. It has been argued that the ability of the accounting profession to continue to self-regulate is therefore largely dependent on upholding the belief that, as regulator, its professional bodies will keep a certain distance or independence from the accounting firms they regulate, and thereby continue to serve the public interest (Puxty et al., 1997). In Spain, auditing standards are elaborated by professional bodies and approved by the ICAC. An immediate consequence of the interventionist tone of the Spanish Audit Law is that, in the event of external pressures questioning the social value of auditing, the ICAC unilaterally may elaborate and implement auditing standards if professional bodies do not provide a response to such pressures (Lizcano-Álvarez, 1999). The increasing pressures on the auditing profession after several financial scandals occurred in early 1990s and the emergence of the expectations gap led to Spanish professional associations, although reluctantly, to design a new auditing standard ongoing-concern assessment (Blasco, 1994; Picazo, 1995; Marín, 1995). The threat of changes of a more interventionist nature (Ruiz-Barbadillo, 1996) worked as a powerful incentive for the profession to ‘do something’. As a result, the standard NTA-May/1993 was elaborated by professional bodies and finally approved by the ICAC in 1993. Similar evidence obtained in other settings characterised by a higher degree of self-regulation by the accountancy profession has warned that the future self-regulation of the accounting profession is not guaranteed because of the growing threats to auditor independence in the modern business environment. Moreover, some commentators talk about ‘the death of self-regulation’ (Thomas, 2003). For example, following recent financial scandals such as Enron and WorldCom in the USA, the SEC and Congress decided that the self-regulatory model had failed, thus clearing the way for a new era of regulation of the accounting profession [Thomas, 2003; see US Congress, (2002), Title I]. Similarly, the Department of Trade and Industry (DTI, 2003) in the UK has stated that the responsibility for setting independence standards for auditors and for monitoring the auditing of listed companies and other significant entities should be transferred from professional accountancy bodies to independent regulators. Several academics have also shown concern about the effectiveness of the profession’s self-regulatory processes (see, among others, Sutton, 1997; Thomadakis, 2004). Our results cast some doubts about the effectiveness of those changes leading to an increase of governments’ involvement in auditing regulation. In Spain, where the state controls auditing regulation through a state agency, the profession attempts, by issuing auditing standards, to show that they are ‘doing something’ to meet society’s expectations and to signal change. That is, the profession shows a similar behaviour as that reported in studies carried out in settings characterised by a minimal or significantly lower involvement of the state in the system of professional organisations (for example, in the UK and in the USA, see Byington and Sutton, 1991; Fogarty et al., 1991). Proposals for a higher degree of government intervention, therefore, may have a limited impact on the day-to-day practice of auditing, in particular, when attempting to reduce the expectations gap. As our evidence suggests the existence of regulatory bodies controlled by the government does not guarantee the effectiveness of auditing standards. Regulators and other commentators might wish to reflect on the extent to which a higher involvement of government agencies in auditing regulation (as a result of reducing the self-regulatory power of the profession) may actually help to enhance the social value of auditing. Auditing standards and the expectations gap 505 The empirical results of this study should be viewed in light of three limitations. First, there may be other important client or audit firm variables that are uncontrolled in our model. As noted by Vanstraelen (2002), a potential omitted variable problem in any model of going-concern opinion can never entirely be overcome. Second, the number of companies with going-concern reports in our sample (n = 65) is relatively small, possibly raising questions about the robustness of the regression results. Finally, we have used two proxy variables to measure auditors’ incentives. Although such proxy variables have been used consistently in previous research, we believe that other proxy variables, such as the audit and non-audit fees received by the audit firm from a particular client as a proxy for economic dependence, may provide additional insights into the relationship between auditors’ economic dependence and their decision to issue a going concern audit report. Due to data unavailability it was not possible to use such variable in our study. Further research on the implementation of ISA 570 Going Concern in different countries may provide insights about the role of auditing standards in shaping auditing practices and auditors’ behaviour when facing going concern decisions. Such research would contribute to understand the extent to which the impact of such auditing standards on auditors’ behaviour and auditing practice is influenced by institutional factors and whether ISA 570 contributes to reduce the expectations gap in comparison with previous national standards ongoing concern assessment. Acknowledgements The authors would like to thank the financial support provided by the Spanish Ministry of Education and Science ECO2010-21627. References Abbott, A. (1988) The System of Professions: An Essay on the Division on Expert Labor, University of Chicago Press, Chicago. Altman, E. and McGough, T. (1974) ‘Evaluation of a company as a going concern’, Journal of Accountancy, December. American Institute of Certified Public Accountants (AICPA) (1988) ‘The auditor’s consideration of an entity’s ability to continue as a going concern’, Statement on Auditing Standards No. 59, AICPA, New York. Amesti, C. (1995) ‘La Actuación de los Auditores en la Sociedad Anónima: la Responsabilidad de los Auditores de la Sociedad Anónima’, Cuadernos de la Revista de Derecho Bancario y Bursátil, No. 3. Arnold, V., Collier, P., Leech, S.A. and Sutton, S.G. (2001) ‘The impact of political pressure on novice decision makers: are auditors qualified to make going concern judgements?’, Critical Perspectives on Accounting, Vol. 12, No. 3. Arruñada, B. (2000) ‘Audit quality: attributes, private safeguards and the role of regulation’, The European Accounting Review, Vol. 9, No. 2. Arruñada, B. and Paz-Ares, C. (1997) ‘Mandatory rotation of company auditors: a critical examination’, International Review of Law and Economics, Vol. 17. Asare, S. (1990) ‘The auditor’s going-concern decision: a review and implications for future research’, Journal of Accounting Literature, Vol. 9. Blasco, J.J. (1994) ‘El Informe de Auditoría: Expectativas’, Partida Doble, Noviembre. 506 E. Ruiz-Barbadillo et al. BOICAC (1993) ‘Resolución de 31 de mayo de 1993, del Instituto de Contabilidad y Auditoría de Cuentas, por la que se publica la Norma Técnica de Auditoría sobre la aplicación del principio de empresa en funcionamiento’, Boletín Oficial del Instituto de Contabilidad y Auditoría de Cuentas, No. 13, ICAC, Madrid. Bougen, P. (1997) ‘Spain, July 1988: some observations on becoming professional’, Accounting, Organizations and Society, Vol. 22, No. 8. Bougen, P. and Vázquez, D. (1997) ‘Debating Spanish audit legislation: the audit law of 1988’, The European Accounting Review, Vol. 6, No. 1. Brien, A. (1998) ‘Professional ethics and the culture of trust’, Journal of Business Ethics, Vol. 17, No. 4. Byington, J. and Sutton, S. (1991) ‘The self-regulation profession: an analysis of the political monopoly of the audit profession’, Critical Perspectives on Accounting, Vol. 2. Carcello, J.V., Hermanson, D.R. and Huss, H.F. (1995) ‘Temporal changes in bankruptcy-related reporting’, Auditing: A Journal of Practice & Theory, Vol. 14, No. 2. Carcello, J.V., Hermanson, D.R. and Huss, H.F. (1997) ‘The effect of SAS no. 59: how treatment of the transition period influences results’, Auditing: A Journal of Practice & Theory, Vol. 16, No. 1. Carcello, J.V., Hermanson, D.R. and Neal, T.L. (2003) ‘Auditor reporting behaviour when GAAS lack specificity: the case of SAS no. 59’, Journal of Accounting and Public Policy, Vol. 22. Carpenter, B. and Dirsmith, M. (1993) ‘Sampling and the abstraction of knowledge in the audit profession: an extended institutional theory perspective’, Accounting, Organizations and Society, Vol. 18, No. 1. Chandler, R.A. (1997) ‘Conflict, compromise and conquest in setting auditing standards: the case of the small company qualification’, Critical Perspectives on Accounting, Vol. 8. Citron, D. and Taffler, R. (1992) ‘The audit report under going concern uncertainties: an empirical analysis’, Accounting and Business Research, Vol. 22, No. 88. Citron, D. and Taffler, R. (2001) ‘Ethical behaviour in the U.K. audit profession: the case of the self-fulfilling prophecy under going concern uncertainties’, Journal of Business Ethics, Vol. 29, No. 4. Citron, D. and Taffler, R. (2004) ‘The comparative impact of an audit report standard and a going-concern standard on going-concern disclosure rates’, Auditing: A Journal of Practice & Theory, Vol. 23, No. 2. Citron, D.B. (2003) ‘The UK’s framework approach to auditor independence and the commercialization of the accounting profession’, Accounting, Auditing and Accountability Journal, Vol. 16, No. 2. Covaleski, M.A., Dirsmith, M.W. and Rittenberg, L. (2003) ‘Jurisdictional disputes over professional work: the institutionalization of the global knowledge expert’, Accounting, Organizations and Society, Vol. 28. Craswell, A.T. (1988) ‘The association between qualified opinions and auditor switches’, Accounting and Business Research, Winter. Craswell, D., Stokes, J. and Laughton, J. (2002) ‘Auditor independence and fee dependence’, Journal of Accounting and Economics, Vol. 33. Cullinan, C. and Sutton, S. (2002) ‘Defrauding the public interest: a critical examination of reengineered audit processes and the likelihood of fraud detection’, Critical Perspectives on Accounting, Vol. 13, pp.297–310. Cushing, B. and Loebbecke, J. (1986) Comparison of Audit Methodologies of Large Accounting Firms, Accounting Research Study No. 26, American Accounting Association, New York. de Angelo, L. (1981) ‘Auditor size and audit quality’, Journal of Accounting and Economics, Vol. 3. de Beelde, I. (2002) ‘Creating a profession ‘out of nothing’? The case of the Belgian auditing profession’, Accounting, Organizations and Society, Vol. 27. Auditing standards and the expectations gap 507 de Fond, M.L., Raghunandan, K.K. and Subramanyam, K.R. (2002) ‘Do non-audit service fees impair auditor independence? Evidence from going concern audit opinions’, Journal of Accounting Research, Vol. 40, No. 4. de Fond, M.L., Wong, T.J. and Li, S. (2000) ‘The impact of improved auditor independence on audit market concentration in China’, Journal of Accounting and Economics, Vol. 20. Department of Trade and Industry (DTI) (2003) Review of the Regulatory Regime of the Accountancy Profession: Legislative Proposals, London, available at http://www.cabinetoffice.gov.uk/servicefirst/index/consultation.htm. Fogarty, T., Heiau, J. and Knutson, D. (1991) ‘The rationality of ‘doing nothing’: auditor responses to legal liability in an institutionalized environment’, Critical Perspectives on Accounting, Vol. 2. Francis, J.R. and Wilson, E.R. (1988) ‘Auditor changes: a joint test of theories relating to agency costs and auditor differentiation’, The Accounting Review, Vol. 58, No. 4. Gaa, J. (1991) ‘The expectations game: regulation of auditors by government and the profession’, Critical Perspectives on Accounting, Vol. 2. García-Benau, M.A. (1994) ‘La Pregunta ¿Dónde Estaban los Auditores? Ha Llegado a España’, Boletín AECA, Vol. 36. García-Benau, M.A. and Humphrey, C. (1992) ‘Beyond the audit expectations gap: learning from the experiences of Britain and Spain’, The European Accounting Review, Vol. 1, No. 2. García-Benau, M.A., Ruiz-Barbadillo, E., Humphrey, C. and Husaini, W. (1999a) ‘Success in failure? Reflections on the changing Spanish audit environment’, The European Accounting Review, Vol. 8, No. 4. García-Benau, M.A., Garrido, P., Vico, A., Moizer, P. and Humphrey, C. (1999b) ‘La Calidad del Servicio de Auditoría: Los Auditores Vistos por sus Clientes’, Revista Española de Financiación y Contabilidad, Vol. 28, No. 102. García-Benau, M.A., Humphrey, C., Moizer, P. and Turley, S. (1993) ‘Auditing expectations and performance in Spain and Britain: a comparative analysis’, International Journal of Accounting, Vol. 28. Geiger, M. and Raghunandan, K. (2001) ‘Bankruptcies, audit reports, and the reform act’, Auditing: A Journal of Practice & Theory, Vol. 20, No. 1. Geiger, M. and Raghunandan, K. (2002) ‘Going-concern opinions in the ‘new’ legal environment’, Accounting Horizons, Vol. 16, No. 1. Gibbins, M. and Mason, A. (1988) Professional Judgment in Financial Reporting, Canadian Institute of Chartered Accountants, Toronto. Gómez-Aguilar, N. and Ruiz-Barbadillo, E. (2003) ‘Do Spanish firms change auditor to avoid a qualified audit report?’, International Journal of Auditing, Vol. 7, No. 1. Gonzalo-Angulo, J.A. (1995) ‘Presentación: La Auditoría una Profesión en la Encrucijada de los Noventa’, Revista Española de Financiación y Contabilidad, Vol. 84. Guiral, A. (2003) ‘El Modelo de Revisión de Creencias como Aproximación Psicológica a la Toma de Decisiones de Auditores y Analistas Bancarios’, Doctoral thesis, Universidad de Alcalá, Madrid. Hooks, K. (1991) ‘Professionalism and self-interest: a critical view of the expectations gap’, Critical Perspectives on Accounting, Vol. 2, pp.109–136. Hopwood, K., McKeown, J. and Mutchler, J.F. (1989) ‘A test on the incremental explanatory power of opinions qualified for consistency and uncertainty’, The Accounting Review, Vol. 1. Hopwood, K., McKeown, J. and Mutchler, J.F. (1994) ‘A reexamination of auditor versus model accuracy within the context of the going-concern opinion decision’, Contemporary Accounting Research, Vol. 10, No. 2. Humphrey, C. (1997) ‘Audit expectations’, in Turley, S. and Sherer, M. (Eds.): Current Issues in Auditing, Paul Chapman Publishing, London. 508 E. Ruiz-Barbadillo et al. Humphrey, C., García-Benau, M.A. and Ruiz-Barbadillo, E. (2003) ‘El Debate de la Responsabilidad Civil de la Auditoría en España: La Construcción del Discurso sobre la Limitación de Responsabilidades por la Corporaciones Locales’, Revista Española de Financiación y Contabilidad, Vol. 119. Humphrey, C., Moizer, P. and Turley, S. (1992) ‘The audit expectations: gap-plus ca change, plus c’est la même chose’, Critical Perspectives on Accounting, Vol. 3. Instituto de Contabilidad y Auditoría de Cuentas (ICAC) (1991) Normas técnicas de Auditoría, Resolution 19 January, Madrid, ICAC. Jones, F.L. (1987) ‘Current techniques in bankruptcy prediction’, Journal of Accounting Literature, Vol. 6. Jones, F.L. (1996) ‘The information content of the auditor’s going concern evaluation’, Journal of Accounting and Public Policy, Vol. 15. Kinney, W. (1986) ‘Audit technology and preferences for auditing standards’, Journal of Accounting and Economics, Vol. 8. Koh, H. (1991) ‘Model predictions and auditor assessment of going concern status’, Accounting and Business Research, Vol. 84. Krishnan, J. and Krishnan, J. (1996) ‘The role of economic trade-offs in the audit opinion decision: an empirical analysis’, Journal of Accounting, Auditing and Finance, Vol. 11, No. 4. Lee, T. (1995) ‘The professionalization of accountancy: a history of protecting the public interest in a self-interested way’, Accounting, Auditing and Accountability Journal, Vol. 8, No. 4. Lizcano-Álvarez, J. (1999) ‘Diecinueve Aspectos Clave y Asignaturas Pendientes de la Auditoría en España’, Actualidad Financiera, Vol. 4, No. 3. López-Combarros, J.L. (1996) ‘Propuestas para una Modificación de la Ley de Auditoría de Cuentas’, Partida Doble, October, Vol. 71. Louwers, T.J. (1998) ‘The relation between going-concern opinions and the auditor’s loss function’, Journal of Accounting Research, Vol. 36, No. 1. Louwers, T.J., Messina, F.M. and Richard, M.D. (1999) ‘The auditor’s going-concern disclosure as a self-fulfilling prophecy: a discrete-time survival analysis’, Decision Sciences, Vol. 30, No. 3. Marín, J.M. (1995) ‘De Qué y ante Quién son Responsables los Auditores: Equívocos que es Conveniente Aclarar’, Revista Española de Financiación y Contabilidad, Vol. 84. Martens, S. and McEnroe, J. (1991) ‘Interprofessional struggles over definition: lawyers, accountants and illegal acts’, Critical Perspectives on Accounting, Vol. 2. Mautz, R.K. and Sharaf, H.A. (1961) The Philosophy of Auditing, American Accounting Association, Sarasota, Florida. McKeown, J., Mutchler, J. and Hopwood, W. (1991) ‘Towards an explanation of auditor failure to modify the audit opinions of bankrupt companies’, Auditing: A Journal of Practice & Theory, supplement. Mills, S. and Bettner, M. (1992) ‘Ritual and conflict in the audit profession’, Critical Perspectives on Accounting, Vol. 3. Mutchler, J.F. and Williams, D. (1990) ‘The relationship between audit technology, client risk profiles and the going-concern opinion decision’, Auditing: A Journal of Practice & Theory, Fall, Vol. 9, No. 3. Mutchler, J.F., Hopwood, W. and McKeown, J. (1997) ‘The influence of contrary information and mitigating factors on audit opinions decisions on bankrupt companies’, Journal of Accounting Research, Vol. 35, No. 2. Neu, D. (1991) ‘Trust, impression management and the public accounting profession’, Critical Perspectives on Accounting, Vol. 3. Newton, J. and Ashton, R. (1989) ‘The association between audit technology and audit delay’, Auditing: A Journal of Practice & Theory, Vol. 8, No. 2, supplement. Paz-Ares, C. (1996) La Ley, el Mercado y la Independencia del Auditor, Editorial Civitas, Madrid. Auditing standards and the expectations gap 509 Petit, M.V. (1995) ‘La Supresión de la Regla de Rotación Obligatoria en el Nombramiento de Auditores de Cuentas por la Ley 2/1995, de 23 de Marzo, de Sociedades de Responsabilidad Limitada’, Revista General de Derecho, Vol. 609. Picazo, P. (1995) ‘La Responsabilidad del Auditor en la Detección de Fraudes y los Problemas de Gestión Continuada de la Empresa’, Partida Doble, Vol. 57. Porter, B. (1993) ‘An empirical study of the audit expectation – performance gap’, Accounting and Business Research, Vol. 93. Power, M. (1997) The Audit Society: Rituals of Verification, Oxford University Press, Oxford. Prado, J.M. (1993) ‘La Norma de Auditoría sobre la Aplicación del Principio de Empresa en Funcionamiento: Consideraciones de los Auditores’, Revista Técnica, Vol. 3. Prado, J.M., Gonzalo, I. and Martín, D. (1995) ‘La Situación de la Auditoría en España desde la Perspectiva de los Auditores’, Revista Española de Financiación y Contabilidad, Vol. 24, No. 84. Public Companies Act (1989) ‘Real Decreto Legislativo 1564/1989’, de diciembre, por el que se aprueba el texto refundido de la Ley de Sociedades Anónimas, BOE 310, December 27, Madrid. Puxty, A., Sikka, P. and Willmott, H. (1997) ‘Mediate interest: the accounting bodies’ responses to the McFarlane report’, Accounting and Business Research, Vol. 27, No. 4. Raghunandan, K. and Rama, D. (1995) ‘Audit reports for companies in financial distress: before and after SAS No. 59’, Auditing: A Journal of Practice & Theory, Vol. 14, No. 1. Reynolds, J.K. and Francis, J.R. (2001) ‘Does size matter? The influence of large clients on office-level auditor reporting decisions’, Journal of Accounting and Economics, Vol. 30. Rosman, A.J., Seol, I. and Biggs, S.F. (1999) ‘The effect of stage of development and financial health on auditor decision behaviour in the going-concern task’, Auditing: A Journal of Practice & Theory, Vol. 18, No. 1. Ruiz-Barbadillo, E. (1996) ‘Un Análisis de las Fases del Conflicto en el Entorno de Auditoría en España’, Revista Española de Financiación y Contabilidad, Vol. 25. Ruiz-Barbadillo, E., Humphrey, C. and García-Benau, M.A. (2000) ‘Auditors versus third parties and others: the unusual case of the Spanish audit liability ‘crisis’’, Accounting History, Vol. 5, No. 2. Sikka, P. (1992) ‘Audit policy making in the UK: the case of the auditor’s considerations in respect of going concern’, The European Accounting Review, Vol. 1, No. 2. Sikka, P. (2002) ‘The politics of restructuring the standard setting bodies: the case of the U.K.’s auditing practices board’, Accounting Forum, Vol. 26, No. 2. Sikka, P. and Willmott, H. (1995) ‘Illuminating the state-profession relationship: accountants acting as department of trade and industry investigators’, Critical Perspectives on Accounting, Vol. 6. Sikka, P., Puxty, A., Willmott, H. and Cooper, C. (1998) ‘The impossibility of eliminating the expectations gap: some theory and evidence’, Critical Perspectives on Accounting, Vol. 9. Spanish Audit Law (1989) ‘Ley 19/1988, de 12 de julio, de Auditoría de Cuentas’, BOE núm. 169/1988, de 15 de julio de 1988, Madrid. Sutton, M.H. (1997) ‘Auditor independence: the challenge of fact and appearance’, Accounting Horizons, Vol. 11, No. 1. Thomadakis, S.B. (2004) ‘Strengthening the credibility and reliability of financial audits: public interest oversight’, International Convergence and Public Oversight of Accounting and Auditing Standards, 17–20 May, OICV-IOSCO, Amman, Jordan. Thomas, C.W. (2003) ‘Enron and beyond: what’s the ‘WorldCom’ing’ to?’, CPA Journal, January, available at http://www.nysscpa.org/cpajournal/2003/0103/nv/nv1.htm. US Congress (2002) Sarbanes-Oxley Act, House of Representatives 3763, Public Law 107–204 107th, available at http://news.findlaw.com/hdocs/docs/gwbush/sarbanesoxley072302.pdf. 510 E. Ruiz-Barbadillo et al. Vanstraelen, A. (2002) ‘Auditor economic incentives and going-concern opinions in a limited litigious continental European business environment: empirical evidence from Belgium’, Accounting and Business Research, Vol. 32, No. 2. Wilkerson, J. (1987) ‘Selecting experimental and comparison samples for use in studies of auditor reporting decisions’, Journal of Accounting Research, Spring. Willekens, M., Steele, A. and Miltz, D. (1996) ‘Audit standards and auditor liability: a theoretical model’, Accounting and Business Research, Vol. 26. Williams, H. (1984) ‘Practitioners’ perspectives on going-concern issues’, CPA Journal, December. Willmott, H. (1991) ‘The auditing game: a question of ownership and control’, Critical Perspectives on Accounting, Vol. 2. Zaid, O.A. (1997) ‘Could audit standards be based on society’s values?’, Journal of Business Ethics, Vol. 16, No. 11. Zmijewski, M. (1984) ‘Methodological issues related to the estimation of financial distress prediction models’, Journal of Accounting Research, Vol. 22, supplement. Notes 1 2 3 4 5 6 7 8 9 Some authors have argued that auditing standards appear to be more interested in specifying what auditors must not do rather than in providing active guidelines upon what they must do (Asare, 1990; Hooks, 1991; Sikka, 1992). Similar behaviour is described by Citron (2003, p.255) when analysing the introduction of a new statement on independence. An example is provided by the issuance in the USA of nine new statements on auditing (‘the expectations gap standards’) as a way of forestalling further activities by Congress or the Securities Exchange Commission (SEC) that would threaten self-regulation, though to all intents and purposes, audit practice actually changed very little (Byington and Sutton, 1991). The results a decade later appear to be a regression in assumed professional responsibility rather an attempt to bridge the gap and move towards meeting the expectations of financial statement users (Cullinam and Sutton, 2002). There are three professional bodies of auditors in Spain: the Instituto de Censores Jurados de España (Spanish Institute of Sworn Auditors, ICJCE), the Registro de Economistas Auditores (Register of Economist Auditors, REA) and the Registro General de Auditores (General Register of Auditors, REGA). For example, in 2002 the ICAC fined DT for the irregularities of the audit of Gescartera in the accounting year 2000 (€540,000). Other examples are the fines to KPMG for the audit work in the company Huarte (approx. €552,000) and Andersen for its work on Corporación Banesto (approx. €566,000) (see, Expansión, 11 February 2002). These criteria are, to a great extent, similar to those factors considered by the Spanish standards that an auditor takes into account when considering the existence of going concern uncertainties. In a review study, Jones (1987) notes that all of the prediction models for bankruptcy published in the literature had similar explanatory power, due to the fact that they tended to include the same general factors: profitability, liquidity and financial risk. The model adopted by Spanish regulators differs also from some Continental European countries such as Belgium, where auditing was modelled as a profession similar to other established professions (de Beelde, 2002). For example in the UK the state does not directly regulate auditing. Instead, the state has relied upon the accountancy professional associations to regulate the auditing industry [Sikka, (2002), p.101].