Small business

Small business start-ups: start-ups

success factors and support
implications
217
Kathryn Watson, Sandra Hogarth-Scott
University of Bradford Management Centre, Bradford, UK and
Nicholas Wilson
Leeds University Business School, Leeds, UK

Introduction
The research reported in this paper relates to an empirical study of business
start-ups and their survival and growth or, conversely, their failure (exit) in the
early years following start-up. A cohort of new start small or micro businesses
which were set up with the support of a Training & Enterprise Council (TEC) in
the West Yorkshire area are investigated. TECs are a government funded
support agency for small and medium-sized businesses which were set up by a
British Government initiative in 1988. By way of background information a
section outlining the support services provided by the TEC is included in this
paper.

Research focus
The focus of the research reported in this paper is to provide information
concerning the start up, development and growth of new small businesses and
the factors associated with various outcomes: failure (ceased trading) and
survival (continued trading). Furthermore, among the survival sub-set the
characteristics associated with growth potential are identified. Initial results
suggested a higher failure rate among female business founders and this issue is
investigated in detail.
New small businesses are considered to be especially vulnerable in the
infancy period following start up. Many fail to develop into thriving, prosperous
businesses. An understanding of the determinants of success and failure in new
small businesses and the motivations of the founders in establishing a new
business should provide valuable insights into the support needs of new
businesses and their founders in the early years.
The primary focus of the study is on the new businesses and their founders
rather than on support agencies. However, the study has implications for the
International Journal of
Entrepreneurial Behaviour &
Research, Vol. 4 No. 3, 1998,
An earlier version of this paper was presented at the 14th Annual Entrepreneurship Research pp. 217-238. © MCB University
Conference, sponsored by Babson College-Kauffman Foundation, in June 1994. Press, 1355-2554
IJEBR type of business support services offered by agencies such as the TECs at a local
4,3 level and these are discussed in the conclusions.
Specific research objectives are to:
• profile the characteristics of those businesses which continued to trade
(survivors) in comparison with those that had ceased trading (failures)
218 within the first three years;
• identify common characteristics of the failure sub-sample;
• investigate small business founders’ motivations for business start-up;
• determine the characteristics of small businesses and their founders
associated with business success (survival and growth);
• examine the problems encountered in operating a small business in the
first three years following start-up.
The investigation of these broad objectives is carried out with a view to making
recommendations for more appropriate business support services at a local level.
The businesses investigated in this study were set up towards the end of a
decade during which there had been a dramatic increase in the number of small
businesses in the UK. This unprecedented expansion was halted by the recession
which began in 1990. Furthermore, government policy in the 1980s had strongly
promoted an enterprise culture. Thus, the sample of business founders under
analysis embarked on new business formation at a time when the social and
political environment encouraged entrepreneurship but during their early years
of development in the early 1990s they experienced severe economic recession.
While recognizing that the economic, political and social environment will affect
the performance of new businesses, the focus of this study is on the businesses
and their founders and internal determinants of business performance and not
on the broader external environment.

Conceptual framework
The approach taken in this paper is primarily empirical. That is, a large number
of personal, business and environmental characteristics of businesses are
measured at start-up and related to eventual outcomes (failure, survival,
growth).
While there is a mass of literature concerning small businesses and
entrepreneurship there is no generally agreed theoretical framework for
carrying out research in this field. In order to facilitate our own research and
understanding of a vast literature on business formation we developed a
research framework based on the extant literature. The framework has been
designed specifically to aid research relating to very small or micro businesses
and may not be considered useful where larger businesses are being
investigated. The research framework is shown in diagrammatic form in
Figure 1.
 Small business
start-ups
ENTREPRENEURIAL BEHAVIOUR

• Experience • Industry sector/business format

• Socioeconomic background
INTERNAL
• Labour and technology 219
• Skills and knowledge • Financial base
• Personality attributes, traits ENVIRONMENT • Strategies and plans
• Values, expectations • Management and resources

Characteristics of the Characteristics of the

Founder Business

THE THE
FOUNDER BUSINESS

The Enterprise

The Business The Business

Infrastructure Customers

• Competitors Market Segments:

• Suppliers EXTERNAL • Business or consumer
ENVIRONMENT • Geographic, demographic
• Banks
• Lifestyle or consumption patterns
• Government • Organisational characteristics
• Support agencies • Purchase behaviour

Figure 1.
MACRO ECONOMIC ENVIRONMENT Analytical framework

Traditional analyses of small firm success/failure have used statistical models

of financial characteristics taken from models of established businesses (see
Keasey and Watson, 1991). While this approach provides useful performance
models for established businesses, it is inappropriate for assessing the viability
of new ventures where the skills and aspirations of the entrepreneur are likely
to be of paramount importance.
Figure 1 sets out some of the factors that are purported to be important in
affecting business outcomes. These factors were identified by a review of the
literature on small businesses and entrepreneurs. Clearly there are many
determinants of business success and growth even in very small businesses.
Successful entrepreneurship is undoubtedly a complex phenomenon and both
internal and external factors impact on business performance. As in large
IJEBR businesses, the characteristics of the business, the business infrastructure and
4,3 the particular customer markets served are important variables affecting
business performance.
However, with small businesses and particularly new ventures, the influence
of the founder in defining the business concept and mode of operation is of
220 paramount importance. In our conceptual framework the business enterprise
consists of both the founder and the business entity. The Committee of Inquiry
on Small Firms known as the Bolton Report (1971) emphasized the special role of
the founder in small businesses. Thus it is essential to investigate the
characteristics, background experiences and motivations of the founder as well
as other influencing factors within the internal and external environment.

Training and enterprise councils

The TECs were launched by the British government in 1988 with the far
reaching objective “to foster economic growth and contribute to the regeneration
of the community by strengthening the skill base and assisting local enterprise
to expand and compete effectively”. The focus of this study is not on whether the
TECs are a successful or unsuccessful policy initiative but on what services and
assistance they should provide at a local level.
In order to maintain confidentiality, both the study area and the support
agency (TEC) involved in the study are described in only general terms. The
strategic aims of the TEC involved in the study are to enhance the survival and
continuing success of new and small businesses. The small business sector plays
a major role in the local economy and generation of new jobs. Thus, the TEC’s
objective is to strengthen the support network to new and existing businesses
and encourage the spirit of enterprise in the district. In order to achieve this, the
TEC provides a support package to new and small businesses which comprises
the following:
• initial seminars offering advice and guidance for potential entrepreneurs;
• financial support to individuals who wish to start their own business;
• training to help in the preparation of a business plan;
• training to help in the preparation of a business plan specially tailored to
the special needs of women;
• assessment of business plans for TEC financial support linked to
counselling visits during the first year of trading;
• a comprehensive business information and counselling service to
enhance business start up, survival and growth;
• a series of training modules to help small businesses improve their
performance.
Financial support to individuals who wished to start their own business was Small business
provided through an allowance (called the Enterprise Allowance) of £50 UK per start-ups
week for a period of up to 18 months (subsequently reduced to a maximum
period of one year).
It is a requirement of TEC support that participants have a business plan and
a cash flow forecast when they start up in business. Founders only qualify for
financial support if they have been unemployed for a period of six weeks or 221
more. Ongoing help and advice to enhance the prospects of business survival is
available throughout their first year of trading through a business information
and advice service and through a series of business seminars.
In the case of administering the above support service the TEC clearly has as
an objective the effective distribution and provision of funding, training and
support among competing prospective business ventures. Implicitly the TEC
has the objective of ensuring a high rate of survival during the period of support
by the agency. Information concerning survival/failure rates for the TEC area
under investigation are not publicly available. However, the desire to improve
survival rates for TEC supported businesses during the early years following
start up was an important instigator for this empirical research study.

Literature review
In this section we provide a brief review of the extant literature relating to small
business start ups and success and failure in small businesses. Given the vast
amount literature in these areas this summary should not be regarded as a
comprehensive review – it merely serves to highlight some of the important
issues relating to our research topic.

Motivations for business start-up

Storey (1994) synthesizes some of the explanations which have been provided by
researchers for variations in the birth of small firms over time, space and sector.
The distinction is made between the work of industrial economists, focusing on
the structure-conduct-performance paradigm, (e.g. Acs and Audretsch, 1989);
and labour market economists who examine new firm formation as a decision
exercised by the individual in the context of the labour market. In making that
decision the individual is influenced by a variety of factors such as work
experience, motivation, personality, family environment and societal norms, etc.
These latter influences have been the prime focus of explanations of new
business formation provided by non-economists and it is this approach which
has been adopted in our investigation of the motivations for new business
formation.
Research using the labour market economists’ approach is derived from the
work of Knight (1921), who argued that an individual could exercise choice by
being in one of three states: unemployment, paid employment or self-
employment. Changes in the relative prices of these three states would induce
IJEBR some individuals to move from one state to another. Thus, an individual has to
4,3 consider income levels associated with seeking new paid employment, becoming
self-employed, or remaining unemployed. If unemployment is high, that person
will be more likely to consider either self-employment or unemployment.
Knight’s framework draws our attention to the influence of “push” factors in
the decision to start a new business. Similarly, Mayes and Moir (1990) argue that
222 the relative attractiveness of self employment and setting up small firms
increased as it became difficult for the greater number of unemployed to find
employment. Self-employment may be riskier and may result in a lower level of
earnings than full-time paid employment with an established firm. Nevertheless,
where the effective comparison is with being unemployed and receiving benefit,
self-employment and small-firm employment may look much more attractive.
In addition to “push” factors one needs to take into account the fact that some
individuals are more likely than others to become self-employed owing to a
variety of “pull” factors. For example, Gray (1990) notes that the lure of personal
independence is an important “pull” factor in the decision to seek a career as a
small business owner although “push” factors such as redundancy, recession,
blocked promotion play a stronger part for many self-employed.
Other researchers focus on human capital factors in the decision to become
self-employed. For example, US empirical work presents consistent evidence that
ceteris paribus, educational attainment levels are positively associated with a
move into self-employment/new business formation (Evans and Leighton, 1990).
Birley and Westhead (1993) found that the incubator organisation, i.e. the last
organisation worked for prior to start-up, can have a significant influence on the
decision to start-up and the location and characteristics of the business. The
incubator organisation provided personal contacts, market knowledge,
managerial skills and industry knowledge.
It is expected that “push” factors may be an important motivator for business
start up among our sample of small businesses. However, we are interested to
find out whether “pull” factors are also an important motivator among our
sample. The motivation for start up is likely to have implications for the sort of
support services and training needs required by business founders and may also
have a bearing on their ultimate success or failure in business ownership.

Factors influencing success and failure

As highlighted in our conceptual framework, a multitude of factors can impact
on business outcomes. In this section we discuss the literature which is pertinent
to certain business performance outcomes, namely failure, survival and growth.
Since the Bolton Report (1971), a mass of research has sought to discover if
there are any clear characteristics shared by the owners of small businesses that
distinguish them from other members of the economically active population and
what characteristics are conducive to small business success. The general
conclusion appears to be that there is no simple pattern. Rather, the evidence
points towards a complex set of interrelated factors that increase or decrease the Small business
probability that an individual will become the owner of a successful small start-ups
business (Stanworth and Gray, 1991, p. 151). Ray (1993) suggests that there is no
ideal-type personality or marginal set of attributes that guarantee success for a
new venture. In order to understand why some individuals become
entrepreneurs and some are more successful than others, according to Ray, three
key elements must be addressed: personality or attributes; background and 223
experience; and skills. Ray (1993) posits that the probability of launching a
successful business is not based on a fixed set of attributes but on an infinite
variety of combinations in which an individual’s positive attributes might
outweigh their negative attributes.
While the idea that small businesses and their founders are heterogeneous
was made explicit by the Bolton Inquiry, certain commonalties were emphasized.
Founders were likely to be both owners and managers and supported by the
family. Their closeness to the firm explains their involvement, flexibility, special
role in risk-taking and innovation, and their fervently guarded sense of
independence. Money, it was suggested, was not their prime source of
motivation, there is a quality of life issue – personal involvement in owning and
managing one’s own firm led to greater satisfaction on a number of fronts all
associated with the notion of independence. Gray (1990) states that the strong
desire of many small business owners to retain personal control and business
independence has been well recognized as a key factor limiting the growth of
many potentially successful small enterprises.
Some authors view business success from an entirely different perspective,
for example, Osborne (1993), in a study of entrepreneurial success, rejects the
idea that success is equated with entrepreneurial competence. Osborne
recommends a shift from a focus on the personality or characteristics of the
business founder to the firm’s underlying business concept and capacity to
accumulate capital. In starting up a business, Osborne suggests that
entrepreneurs should start or buy a business which has a hospitable
environment; understand how customers assess product/service benefits; avoid
markets dominated by one or several companies with product or price
leadership; understand the underlying economics in which the firm operates and
be cautious about starting a business where capital requirements suppress or
eliminate discretionary cashflow.

Problems experienced by small firms

Studies which focus on the problems experienced by small firms and how the
successful owner-managers overcome such problems lend some support for
Osborne’s argument (Osborne, 1993), especially concerning the managerial
competences required to understand the nature of specific markets. However,
these studies also draw attention to the importance of personal and family
situations impacting on the performance of the business.
IJEBR For example, Cromie (1991), in a study of male and female owners of young
4,3 firms in the demarrage phase of development (period between launch and take
off into sustained growth), found that young organisations experienced
problems primarily in the areas of accounting and finance, marketing and the
management of people. In addition, personal problems attributed to the
workload, the loneliness of running a new venture and ambiguity of owner
224 managing causes some concern. The age and family circumstances of the
founders under investigation will have a bearing on success/failure. Those
founders with a stable and supportive family structure and/or family experience
of self-employment may be better able to cope with the demands of a new
venture. There were few significant differences in the problems experienced by
the genders. However, women were more likely to cite problems concerning the
financing of their ventures.
Similarly, Smallbone (1991) investigated the problems faced by new and small
firms which were supported by a local enterprise agency situated in an outer
London borough. He found that the most frequently cited problem was
marketing or selling the product or service, followed by financial control and
unsuitable premises. Also, problems varied accordingly to the stage of
development: 30 per cent said raising finance was a problem in the pre-start
phase. Establishing a market and a customer base was expected to be a
significant hurdle for those new ventures that were attempting to establish
products or services outside of the founder’s immediate area of expertise or
skills.

Research propositions
Based on our conceptual framework and the literature review the following
broad propositions were developed:
P1. There are likely to be distinctive differences between successful business
founders and unsuccessful business founders concerning their
background, experience, skills and knowledge which they bring to the
new business venture.
P2. There are likely to be distinctive differences between successful business
founders and unsuccessful business founders concerning their
motivations for starting a business and objectives in business ownership.
P3. The owners of growth businesses are more likely to possess
entrepreneurial characteristics than the owners of static or non-growth
businesses.

Methodology
This empirical study investigates the characteristics of a cohort of new start
small businesses set up during a period of economic recession (1989-1993) by
founders who had experienced various spells of unemployment prior to start-up.
In the initial phase, at least, these founders can be thought of as making the Small business
transition from unemployment to self-employment. In the light of the literature start-ups
review, the study specifically addresses the founders’ motivations for business
start-up, their personal background, education, employment history and
experience, problems encountered in operating the business, training and advice
received from support agencies, and their aims, objectives and expectations for
the future of their businesses. Data relating to the performance of the businesses 225
were obtained from a detailed postal questionnaire together with basic business
data such as industry sector, ownership structure, employment, period of
trading and capital employed.
A random sample of 504 new businesses was taken from the TEC records of
supported businesses. All 504 business owner managers were surveyed by a
postal questionnaire which was mailed first class to their private homes. The
questions used included dichotomous, multi-choice and attitudinal scaled
questions plus a small number of open-ended questions and covered the various
characteristics deemed important from the literature, i.e. background details of
the business; details about the founders’ background and employment history;
the start-up period of the business; details about the business now and future
prospects (completed by those still trading); and problems experienced and
reasons for discontinuance (completed by those who had since ceased trading).
A covering letter was sent with the questionnaire, which aimed to encourage
participation in the survey and assure participants of the confidentiality of
individual survey responses. To ensure a good response rate, the business
founders who did not return the questionnaire were contacted again, as follows:
a reminder letter was sent to all founders who had not returned the questionnaire
after one week; and a telephone call was made to all those who had not returned
the questionnaire after two weeks. Thus, 437 reminder letters were mailed and
an attempt was made to contact 310 founders by telephone.
A total of 166 (33 per cent) usable survey questionnaires were returned
completed by the cut-off date in 1994. If the number of founders who had
changed addresses and were not contactable (gone aways) are taken into account
the response rate achieved was 35.5 per cent. Telephone contact with founders
revealed that some of those who had ceased trading or only traded for a short
time had assumed we would not be interested in their experiences or views. We
were able to assure them that this was not the case and encourage their
participation in the survey together with those who were still trading. As with all
surveys non-response bias is a problem, which limits generalisability. In order to
minimise this problem every effort was made to obtain a good response rate. The
distribution of the respondent businesses was found to be similar to the
population of TEC-supported new start small businesses on the criteria of
business sector, age and start-up capital.
Statistical analysis of the generated data was performed using SPSS/PC+. An
analysis of the frequency of responses to the questionnaire and the mean (sd) of
IJEBR individual variables was undertaken in order to describe and understand the
4,3 sample.
In one section of the questionnaire respondents were asked to indicate on a
five-point Likert scale the importance they attached to various motivations for
business start-up from a list of 11 items. Factor analysis was performed in order
to distinguish different patterns of responses in this data. The “still trading”
226 versus “ceased trading” categorisation was used to construct a dichotomous
variable which we refer to as the survival/failure sub-sample. Within the
survival category a further breakdown was performed in order to differentiate
between “static” versus “growing” businesses. This categorisation was based on
whether respondents intended to increase the number of people employed in
their businesses in the next one or three years.
The initial focus of the research was on the characteristics of the
survival/failure sub-samples and to a lesser extent the static/growth sub-
samples. T-tests and chi-square tests were conducted in order to identify any
significant differences in sub-sample characteristics and responses. The
multivariate techniques of discriminant and logit analysis were then employed to
isolate the combinations of variables associated with survival/failure. T-tests
and factor analysis were used to examine gender specific problems and
differences in the background/experiences between men and women.

Survey results
Sample characteristics
Frequency schedules were prepared to describe the total sample of 166
respondents. The key characteristics are presented briefly below:
• at the time of the survey 123 respondents (74 per cent) were still trading
and 43 (26 per cent) had ceased trading;
• 70 per cent of the businesses were sole proprietorships;
• 67 per cent of the businesses operated from the home of the respondent;
• 69 per cent of the respondents were male and 31 per cent are female;
• 71 per cent of the respondents were married;
• 63 per cent of the respondents had CSEs, GCSEs or O Levels;
• 22 per cent of the respondents had technical qualifications;
• 19 per cent of the respondents had A or A/S Levels;
• 20 per cent of the respondents had a degree;
• 95 per cent of the respondents had previously been in paid employment;
• 28 per cent of the respondents had previously been self-employed;
• 51 per cent of the respondents had been unemployed for six months or
less prior to start-up;
• 41 per cent of the respondents had been unemployed only once.
Motivations for start-up Small business
One section of the questionnaire asked respondents to indicate on a five-point start-ups
Likert scale the importance of various motivations for starting up in business
from a list of 11 such motivations. The responses were subjected to the statistical
technique of factor analysis in order to distinguish different patterns of
responses among respondents. The factor analysis reduced the data to four
patterns of responses which we have labelled as follows: 227
(1) Entrepreneurial factor (26.8 per cent):
• be independent/own boss;
• use own creative skills;
• do enjoyable work;
• frustrated with previous job.
(2) Personal-opportunistic factor (14.6 per cent):
• able to work from home;
• make a lot of money;
• further career objectives.
(3) Market-opportunistic factor (11.1 per cent):
• meet a service or need;
• exploit a market opportunity;
(4) Financial needs factor (10.3 per cent):
• earn a reasonable living;
• get off the dole.
The total variance accounted by the four factors was 62.8 per cent and was
deemed statistically acceptable by the appropriate diagnostic tests. Factor scores
were calculated for each respondent and these four variables were than used in
the further statistical analysis.

Comparison of sub-samples: survivors v. failures

T-tests of means were used to identify the significant differences between two
sub-samples, defined as survivors (still trading) and failures (ceased trading). A
summary of these results is as follows:
The survivors sub-sample were more likely to be male and to have two or
more financial dependents. The failures sub-sample were more likely to have
secondary education qualifications (CSE, GCSE, O Level) and to have children
under the age of six years. Analysis of the differences in background/
experiences between the genders is presented in a separate section below.
Concerning employment history, the survivors were more likely to have been
in full time employment as opposed to part-time or temporary employment and
to have been in their last job for more than one year. Also, they were more likely
IJEBR to have been in senior/middle management or in skilled manual employment than
4,3 those who had ceased trading; the latter are more likely to have worked for a
family business. There is a very significant difference between the two sub-
samples concerning apprenticeships: 36 per cent of respondents had served an
apprenticeship and 86 per cent of these were still trading. This suggests that those
who had served an apprenticeship were more likely to be successful, particularly
228 if their business is related to their trade. The survivors were more likely to have
been unemployed only once and their period of unemployment prior to starting
up a business was more likely to have been less than six months..
In line with the factor analysis, referred to above, the t-tests showed that the
two sub-samples differed concerning their motivations for start-up. The
survivors attached greater importance to being their own boss, being
independent, using creative skills and frustration in last job was a more
important push factor. The failures trading attached greater importance to
being able to work from home. Earning a reasonable living and doing enjoyable
work was important for both sub-samples.
The two sub-samples differed in relation to advice/training received prior to
start-up. The failures were more likely to have obtained advice/training in
writing/using a business plan, market research, communications/promotions.
Whereas, the survivors were more likely to have obtained advice/training in
computer skills.
With regard to problems experienced in running the business, obtaining
finance was significantly more important for the failures. For the survivors,
finding/keeping appropriate skills; management skills; and keeping up with
changes in the marketplace were more important. Important problems where
there was no difference between the two sub-samples were finding/keeping sales
outlets, marketing or selling the product or service and dealing with the
competition.

Long term business aims and prospects

For the survivors the most important long term aim was to “earn income for me
and my family” (86 per cent), followed by “be better than the competition” (71 per
cent), “grow the business” (51 per cent) and “become a major player in the area”
(35 per cent). To “create a profitable business which I can sell at a later date” was
important for only 28 per cent of the survivors.
The founders’ perceptions of the prospects for their businesses for the next
year was that 72 per cent expected their turnover to grow, 63 per cent expected
their profit to grow and 53 per cent expected the size of their overall market to
grow. The growth prospects for the next three years were perceived rather more
optimistically; turnover (81 per cent), profit (80 per cent) and market size (67 per
cent).

Reason for discontinuance of the business

Of the 43 respondents who had ceased trading, 19 (45 per cent) had gone into full
time employment, 12 (29 per cent) were unemployed and two (5 per cent) were
either a housewife or househusband. The remaining ten (26 per cent) described Small business
their current status as “other”, which included retirement and sickness start-ups
claimants. The most important reason for giving up was because “business not
earning enough money” (65 per cent); followed by “poor trading conditions”;
“cash flow problems” and “personal reasons” (42 per cent); then “poor long term
prospects” (40 per cent) and “obtaining full time employment” (30 per cent).
229
Comparison of sub-samples: growth and static businesses
A comparison was made between two sub-samples based on their growth in
employment intentions. The “growth” group intended to increase the number of
employees in the business in the next one or three years whereas as the “static”
group did not. T-tests of means identified 38 significant differences between
these two sub-samples as shown in Table I.
Notably, the growth sub-sample were more likely to want to introduce new
products, increase customers and be better than the competition. Also, they
wished to increase turnover and profits and reduce costs. They were more likely
to have obtained additional finance since start-up, were less likely to operate the
business from home and were more likely to have both large competitors and
large buyers. Also, the growth sub-sample were more likely to have received
advice in leadership/motivation, staff training and recruitment. The static group
were more likely to be over 50 years old, to have started up to enable work from
home and to actually work from home and it is more likely that their longest
spell of unemployment was over one year.

Multivariate analysis: characteristics of survivors and failures

A considerable volume of literature has grown up around the problem of
predicting individual corporate failure, using primarily company financial
information and variants of the Z-score technique. Multivariate models
(discriminant, logit, multi-logit and hazard functions) are constructed by
analysing a large number of financial ratios for a sample of “healthy” and
“failed” companies. Statistical analysis is then used to identify a small number of
key ratios which can be weighted and combined in order to classify firms into
“healthy” or “failing” categories. The essence of this technique has been to
identify those characteristics from a given stock of companies which increase the
risk of failure. Such models were built using data from a sample of failed and
non-failed small firms in the UK (Keasey and Watson, 1986). Increasingly, these
models have incorporated non-financial characteristics as both a means of
improving predictive accuracy and understanding the reasons underlying
business failures as opposed to simply identifying symptoms.
Both discriminant and logit techniques were used in an attempt to isolate the
start-up characteristics of the new businesses in our sample which might be
associated with survival/failure. Initially, discriminant analysis was applied in
order to identify clusters of key discriminating variables. A within-sample
classification accuracy of 87-90 per cent was obtained using this technique. A
model determining the probability of survival-failure was then estimated using
IJEBR Significance Significance
4,3 (per cent) (per cent)

The growth group (approx. 45 respondents)

Start-up reason Aims for the business
To further career <1 To grow the business <1
230 To make a lot of money <1 To be a major player <1
To exploit a market opportunity <1 To be better than competition <1
To be own boss <5 To create a profitable firm <1
Choice product/service Objectives
Choice due to knowledge/ To introduce new products <1
experience gained in previous To increase employment <1
employment <1 To increase customers <5
Employment To increase profits <5
More S/M management <1 To reduce costs <5
More made redundant <5 To increase sales revenue < 10
More full-time in last job < 10 To build good customer relations < 10
Advice received Problems
Leadership/motivation < 10 Finding/keeping people <1
Staff training/recruitment < 10 Obtaining premises <1
Business circumstances Finding/keeping skills <5
Do not operate from home <1 Management skills < 10
Have large competitors <5 Prospects
Have large buyers < 10 Growth in turnover <1
Obtained additional finance since Growth in profit and market size
start-up <5 in next year and next three years <1

The static group (approx. 72 respondents)

Background Employment
More over 50 years old <1 More manual workers < 10
More secondary/comprehensive More worked in government
Table I. education < 10 organisations < 10
Comparison of growth Business operation More part-time or temporary < 10
and static sub-samples Start-up to enable work from home < 1 More longest spell of unemployment
(student’s t-test) More work from home <1 over one year < 10

a logit estimator. The dependent variable was constructed as (1=FAILURE;

0=SURVIVOR). The advantage of using the logit estimator is that a wide variety
(non-normally distributed) of qualitative characteristics of the businesses can be
built into the multivariate model and the significance of individual variables in
the model are more easily determined. Logit models provide a probability of
failure as the output i.e. a score between 0 and 1; cut-off points of 0.5 are usually
used to identify failing and surviving businesses.
Table II presents the results of this analysis. The estimated model achieved a
within-sample classification accuracy of approximately 90 per cent and appears
to be well specified. A number of interesting results emerge from this analysis
which support previous results and have policy implications. Concentrating on
the variables in the equation that are significant at 5 per cent level or better,
Variable Coefficient Significance
Small business
start-ups
FACTOR1 –1.924 0.002
FACTOR3 1.936 0.001
CASH1 1.705 0.200
CHANGE –1.355 0.300
SEX –2.625 0.014 231
DEPEND –0.455 0.221
TUNEM 0.509 0.021
NCO 0.179 0.015
NAGE2 –1.317 0.160
NLONG –3.867 0.002
NSTATUS 3.543 0.019
SELFE –3.080 0.016
REL –1.328 0.085
NSIZE01 –1.183 0.168
NTYPE2 1.103 0.243
NTYPE3 –9.184 0.780
Constant –3.687 0.102
Classification accuracy 89.52 per cent
Chi-square 141.38
Note:
FACTOR1: Factor score (entrepreneurial)
FACTOR3: Factor score (market opportunistic)
CASH1: Start-up capital below £5,000 = 1; above = 0
CHANGE: Change business plans early on = 1; did not = 0
SEX: Male = 1; Female = 0
DEPEND: Number of dependants
TUNEM: Number of times unemployed
NCO: Number of companies worked for
NAGE2: Middle age range = 1; other = 0
NLONG: Worked for one company for six years or more = 1; otherwise = 0
NSTATUS: Never had a skilled job = 1; skilled = 0
SELFE: Previously self-employed = 1; otherwise = 0 Table II.
REL: Have relatives who are self-employed = 1; otherwise = 0 Logit estimates:
NSIZE01: Previously worked for a small company = 1; otherwise = 0 probability of failure

associated with the probability of survival are: sex, age, work experience, skills,
previous experience of self-employment or having a close relative in self-
employment. The model suggests that the surviving category is typically
middle-aged males who have acquired a skill with one employer and are
applying this experience and skill in their new business, and have experienced
only short spells of unemployment. The entrepreneurial factor (i.e. those setting
up for positive reasons: to be independent, to be creative, to do enjoyable work,
and frustration with previous employer) is associated with survival. Factors
which appear to reduce the probability of survival are: setting up to exploit a
market opportunity or meet a perceived service need (Factor 3), gender, long
and/or frequent spells of unemployment and little long-term employment
experience with one employer.
IJEBR Comparison of sub-samples: men and women
4,3 Fifty-two (31 per cent) of the respondents were women and 114 (69 per cent) were
men. At the time of the survey 30 (58 per cent) of the women and 93 (82 per cent)
of the men were still trading (22 women and 21 men had ceased trading). Further
analysis (t-tests, factor analysis) was carried out to see if there were any gender
specific problems or differences in background, experience which resulted in a
232 higher proportion of women terminating their business operation.
There were significant differences between the genders in relation to previous
employment history; motivations for business start up; training/advice received
prior to start up and reasons for discontinuance. There was an age difference
between the two sub-samples; the women sub-sample were more likely to be
under 30 years and men sub-sample were more likely to be between 30 and 50
years. The men were more likely to have no children and the women more likely
to have one or two children.
With regard to previous employment history, men were more likely to have
been in full time employment; in senior/intermediate management or manual
work; to have served an apprenticeship; and to have worked for a private
company. Women were more likely to have been in part-time or temporary
employment; junior management, clerical, other professional or small business
services. Men were more likely to have been made redundant. Of the female
respondents, 16 were not unemployed prior to start-up but entered small
business from maternity leave, career break, retirement or following husband’s
redundancy. Excluding these respondents, there were no differences between
men and women concerning their unemployment history (length of
unemployment, number of times unemployed).
Men were more likely to state that their reasons for starting up are: to earn a
reasonable living; to get off the dole; to make a lot of money; to earn income for
me and my family; and because of frustration with previous job. There was no
significant difference between the genders concerning the following start-up
reasons: to further career; to do enjoyable work; to be independent/own boss; to
use creative skills; to meet a service need in an area; to exploit a market
opportunity and to enable work from home.
T-tests showed significant differences indicating that women received more
training/advice on obtaining finance; writing/using a business plan; selling
skills; market research and time management. Overall the amount of
training/advice received prior to start-up was greater for women than for men.
The means for men and women for each of 14 variables (measured 1-0) are given
in Table III. An estimate of the overall amount of training/advice received by
men and women is indicated by the sum of means. A t-test, significant at less
than 5 per cent, showed that overall the women sub-sample had received more
training than the men sub-sample.
There were virtually no differences between men and women concerning
problems experienced in running the business and their aims and objectives for
their businesses. The only difference being that men were more likely to state
that finding a product/service to provide is a problem and women were more
likely to state that increasing sales revenue is an objective. Other variables where
Training/advice received in the Men Women Total sample Small business
following areas, prior to start-up mean mean mean start-ups
How to obtain finance 0.3333 0.4808 0.3795
How to write/use a business plan 0.6491 0.8462 0.7108
Bookkeeping 0.3772 0.5000 0.4157
Financial planning/budgets 0.3860 0.4423 0.4036 233
Creditor/debtor control 0.1667 0.1538 0.1627
Taxation 0.2544 0.2692 0.2590
Personnel law 0.0877 0.0962 0.0904
Leadership and motivation 0.0965 0.1346 0.1084
Staff training and recruitment 0.0614 0.0577 0.0602
Selling skills 0.2193 0.3846 0.2711
Market research 0.3421 0.5577 0.4096
Time management 0.1053 0.2308 0.1446 Table III.
Communications and promotions 0.2018 0.3077 0.2349 Level of training/
Computer skills 0.0965 0.1538 0.1145 advice received prior
Total advice 3.4685 4.6154 to start-up

there was no difference between men and women were: how long in last job, size
of organisation last worked for; qualifications; type of businesses (e.g.
partnership etc); changing plans since start-up; operating from home or business
premises; previous experience of self-employment; close relatives who have run
a business; and family help in running the business.
For those who had ceased trading there were differences between men and
women concerning their reasons for discontinuance. From a list of ten reasons
for discontinuance, t-tests showed that men were more likely to have ceased
trading because they obtained full time paid employment and women were more
likely to have ceased trading due to personal reasons.
Thirteen (59 per cent) of the 22 women and five (24 per cent) of the men who
had ceased trading cited personal reasons in relation to their discontinuance of
the business. These respondents were asked to specify what specific personal
reasons were applicable. Examples of the personal reasons stated by women
were: health; needed at home; difficult balance between work and home; got
married; had another baby; now divorced (ex-husband business partner);
separated; chose wrong business partner; heavy work; thought I was getting
nowhere; bought wrong stock which did not sell; easier to work for a company;
and ceased trading in order to travel round the world. For the men who cited
personal reasons in relation to their discontinuance; examples were: injury and
advised to stay out of manual work; had another child; pressure/problems in
running the business; and changing to providing a different service.
Eight of the variables cited as reasons for discontinuance were reduced to
three patterns of responses using factor analysis. Only two respondents cited
“someone else took over the business” as a reason for discontinuance and only
one respondent cited “went into full time education” as a reason; consequently,
these variables were not used in the factor analysis.
IJEBR The three factors produced by the analysis are:
4,3 (1) Business marginality (44.7 per cent)
• cash flow problems;
• lack of finance;
• poor long term business prospects;
234
• poor trading conditions;
• business not earning enough money;
• increased competition.
(2) Obtained full-time paid employment (16.5 per cent).
(3) Personal reasons (13.8 per cent).
The total variance accounted for by the three factors was 75 per cent. Factor
scores were calculated for each respondent who had ceased trading and t-tests
on the factor scores revealed significant differences between men and women on
Factor 2 and Factor 3. There was no significant difference between the genders
concerning Factor 1 – business marginality. These results are consistent with the
findings above and confirm that personal reasons, mainly related to family or
problems in coping with running both a home and a business, are important
factors causing a higher rate of discontinuance by women. For men an important
reason for discontinuance is finding full time paid employment.

Concluding remarks
Summary and discussion
As expected, the results of this empirical study confirm many of the findings of
other researchers in the same field, especially concerning the heterogeneity of
small businesses and their founders and the influence of “push” and “pull”
factors in the decision to start a new business venture. However, this study has
taken our understanding of the issues concerning small business start-up and
early development a stage further, as outlined in this section. Furthermore, one
notes that the study has produced consistent results using a broad range of
statistical analysis techniques.
The study confirms our propositions that even among the very small
businesses there are considerable differences in:
• the owners’ personal backgrounds and experiences and problems
encountered in running their businesses (P1);
• motivations for start-up and objectives in running a business (P2);
• growth orientations and objectives for business development (P3).
Also, it is apparent that there are significant differences concerning:
• the level of training and advice received from support agencies;
• where applicable, the reasons for discontinuance of the business.
P1 Small business
Although most of the respondents had been unemployed prior to starting their start-ups
business (some for only a short time), there were, of course, differences in the
education levels and employment histories of the sample respondents. The
results concerning employment history confirm the findings of Birley and
Westhead (1993) that the incubator organisation can have a significant influence
on the business venture. Our results are inconclusive concerning education 235
levels, however, an interesting result was that those who had served an
apprenticeship were more likely to survive.
Consistent with the research findings of Cromie (1991) and Smallbone (1991),
our results show that marketing and finance are particular problems for small
business owner-managers. Finding/keeping sales outlets and marketing/selling
the product or service and dealing with the competition were important
problems for all respondents. Obtaining finance was more of a problem for those
who had ceased trading, and for those still trading keeping up with changes in
the marketplace and retaining necessary skills were more problematic. In
agreement with Osborne (1993) our results support the view that the owner-
manager’s underlying business concept is of considerable importance.
P2
Consistent with the labour market economists’ work concerning small business
formation we found differences in the importance of various “push” and “pull”
factors as motivators for business start-up. “Push” criteria such as redundancy,
unemployment, frustration with previous employment and the need to earn a
reasonable living are important motivators for start-up, more so for men than
women. However, “pull” criteria such as independence, being ones own boss,
using creative skills, doing enjoyable work and making a lot of money are more
important motivators and these are more closely associated with survival.
Factor analysis of the motivations for start-up shows that the pattern of
responses which we have labelled the entrepreneurial factor (be
independent/own boss, use creative skills, do enjoyable work, and frustrated in
previous employment) accounted for 26.8 per cent of the variance and z-scores on
this factor were higher for the survivors (still trading) than the failures (ceased
trading). The entrepreneurial factor was also significant in the multivariate
logistic regression analysis.
From a review of the literature on entrepreneurship it is apparent that the
variables included in our entrepreneurship factor are associated with
entrepreneurial behaviour and suggest a positive or “pull” motivation for
business start-up. However, the literature suggests that there are many other
characteristics associated with entrepreneurship, for example, risk-taking,
achievement, locus of control, innovation, etc. and consequently we recognize
that this issue needs further investigation, especially concerning very small
businesses.
Perhaps a rather unusual result from the multivariate analysis is that the
factor labelled market opportunistic factor (where start-up is to meet a service
need or exploit a market opportunity) is associated with business failure.
IJEBR P3
4,3 Comparing the growth and static sub-samples, it is clear that the growth group
had far more ambitious aims and objectives in the development of their
businesses. Also, managerial skills, particularly concerning employing people in
the business and giving leadership/motivation, are more important for these
business owners. The growth sub-sample had received more training/advice on
236 leadership/motivation, staff training and recruitment.
We recognize that our dependent variable for growth, based on intentions to
increase employment in the business, is a fairly crude measure of growth
intentions and that multivariate analysis is likely to provide greater insight into
this issue.

Training and advice

There were notable differences in training/advice received prior to start up.
Overall women received more training/advice than men. Our findings suggest
tentatively that potential new business founders seek and obtain training/advice
which they perceive as relevant to their needs. However, the area where most
training was obtained was in writing/using a business plan, which is a pre-
requisite of being accepted for sponsorship by the TEC. In many other subject
areas the amount of training received appears to be low, with mean levels for the
total sample below 0.5 on 13 variables measured on a 1-0 scale (see Table III).
This raises two questions: first, do applicants correctly perceive their training
needs prior to business start-up, and second, is high quality training in
appropriate areas readily available.

Business discontinuance
The most important reasons for business discontinuance were: business not
earning enough money; poor trading conditions; cash flow problems and
personal reasons. Factor analysis of the reasons for discontinuance showed that
variables associated with business marginality were the primary factor for
discontinuance but that women were more likely to cease trading due to personal
problems, mainly associated with home and family, and men were more likely to
cease trading because they had obtained full time paid employment. It is
important to bear in mind that the sample of businesses under analysis was set
up during a period of severe recession in the UK, which no doubt had some
bearing on their performance in the early years. These results have given us
some insight into the problems experienced by women business founders,
however, there is clearly more research required in this area.

Implications for support agencies

Our study has shown that there are considerable differences in the
characteristics of small business owner-managers and the nature of businesses
which they have founded. This diversity has implications for support agencies
concerning their appraisal techniques and ongoing training and advice. These
implications are discussed below.
P1 Small business
A major issue for support agencies such as the TECs is how to assess start-ups
applications for business support schemes by potential business founders. In
this respect we suggest that our conceptual framework (see Figure 1) is a useful
indication of the complexity of issues which need to be addressed in such
appraisals, as follows:
First, the fundamental issue which must be addressed is the viability of the 237
business concept itself and this requires focusing on the external business
environment and market specific factors. It may well be that the analytical skills
of the prospective business founders are inadequate for making such an
assessment and that assistance or training is required in this area. Clearly the
use of business plans in screening procedures seeks to address this issue,
however, the problem with this is that the business plan then becomes an
obstacle to overcome in the selection procedure rather than a working document
for running the business. Furthermore, if the business plan is based on
erroneous assumptions it loses its validity as an analytical tool.
Second, appraisal should be made of the internal business environment and
especially whether the prospective founder has the appropriate background,
skills and experience to meet the demands of the new business venture. As Ray
(1993) points out, there is no ideal set of attributes that guarantee success; the
requirements of each business venture must be considered and an assessment
made of each individual’s training and business advice needs.

P2
At the outset it would be useful for both the business founder and the business
support agency to have a clear idea of the motivations behind the new business
venture and the founder’s objectives in developing the business. However, one
needs to bear in mind that during the recent recession many people have become
unemployed and a period of self-employment may enable them to retain their
position within the labour market. Conversely, if they move to unemployment it
is likely to be considerably more difficult to return to employment and their skills
and experience will be lost.
Furthermore, support agencies need to be aware that quality of life issues,
personal problems and family matters can be very important for small business
owners and can impact on the success of the business venture. Some
consideration needs to be given to the compatibility of the business venture with
other demands on the owner’s time and the family network from which the
founder can gain support. Constructive advice should be given on how to
manage a whole variety of demands (possibly conflicting) made upon the
business founder.

P3
Our study suggests that those entrepreneurial business founders with relatively
ambitious growth intentions need support on issues concerning the
management of people, especially recruitment, training, motivation and
IJEBR leadership. Furthermore, the management of change, especially keeping up with
4,3 changes in the external market environment, appear to be critical.
References
Acs, Z. and Audretsch, D. (1989), “Births and firm size”, Southern Economic Journal, Vol. 55,
pp. 467-75.
238 Birley, S. and Westhead, P. (1993), “A comparison of new businesses established by ‘novice’ and
‘habitual’ founders in Great Britain”, International Small Business Journal, Vol. 12 No. 1,
pp. 38-60.
Bolton Report (1971), Report of the Committee of Inquiry on Small Firms, chaired by J.E. Bolton,
Cmnd. 4811, HMSO, London.
Cromie, S. (1991), “The problems experienced by young firms”, International Small Business
Journal, Vol. 9 No. 3, pp. 43-61.
Evans, D.S. and Leighton, L.S. (1990), “Small business formation by unemployed and employed
workers”, Small Business Economics, Vol. 2 No. 4, pp. 319-30.
Gray, C. (1990), “Business independence – impediment or enhancement to growth in the 1990s?”,
paper presented to the 13th National Small Firms Policy and Research Conference, Harrogate.
Keasey, K. and Watson, R. (1991), “The state of the art of small firm failure prediction:
achievements and prognosis”, International Small Business Journal, Vol. 9 No. 4, pp. 11-28.
Keasey, K. and Watson, R. (1986), “The prediction of small company failure: some behavioural
evidence for the UK, Accounting and Business Research, Winter, pp. 49-57.
Knight, F.H. (1921), Risk, Uncertainty and Profit, Houghton Mifflin, New York, NY.
Mayes, D.G. and Moir, C. (1990), “The growth of small firms in the UK”, in Moir, C. and Dawson, J.
(Eds), Competition and Markets, Macmillan, Basingstoke, pp. 41-62.
Osborne, R.L. (1993), “Why entrepreneurs fail: how to avoid the traps”, Management Decision,
Vol. 31 No. 1, pp 18-21.
Ray, D.M. (1993), “Understanding the entrepreneur: entrepreneurial attributes, experience and
skills”, Entrepreneurship & Regional Development, Vol. 5, pp 345-57.
Smallbone, D. (1991), “Success and failure in new business start-ups”, International Small Business
Journal, Vol. 8 No. 2, pp. 34-45.
Stanworth, J. and Gray, C. (1991), Bolton 20 Years On: the Small Firm in the 1990s, Paul Chapman
Publishing Ltd, London.
Storey, D.J. (1994), Understanding the Small Business Sector, Routledge, London.