SME Performance Separating Myth From Reality (PDFDrive)

SME Performance
SME Performance
Separating Myth from Reality
John Watson
Professor of Accounting and Finance, The University of
Western Australia
Edward Elgar
Cheltenham, UK • Northampton, MA, USA
© John Watson 2010
All rights reserved. No part of this publication may be reproduced, stored in a

retrieval system or transmitted in any form or by any means, electronic,
mechanical or photocopying, recording, or otherwise without the prior
permission of the publisher.
Published by
Edward Elgar Publishing Limited
The Lypiatts
15 Lansdown Road
Cheltenham
Glos GL50 2JA
UK
Edward Elgar Publishing, Inc.

William Pratt House
9 Dewey Court
Northampton
Massachusetts 01060
USA
A catalogue record for this book is available from the British Library
Library of Congress Control Number: 2009940729
ISBN 978 1 84542 977 5
Printed and bound by MPG Books Group, UK

02
Contents
PART I BACKGROUND
1 Introduction 3
PART II SME PERFORMANCE
2 Defining and measuring SME failure/success 11

3 The effects of age, size, industry and the economy on SME
failure rates 31
PART III COMPARING THE PERFORMANCE OF

FEMALE- AND MALE-CONTROLLED SMES
4 Failure rates 47
5 Relating outputs to inputs 53
6 Adjusting for risk 59
PART IV GROWTH FINANCING FOR SMES
7 A qualitative analysis 69
8 A quantitative analysis 87
PART V NETWORKING AND SME PERFORMANCE
9 The association between networking and performance 101

10 Networking: comparing female- and male-controlled SMEs 116
PART VI CONCLUSIONS
11 Conclusions, implications and areas for future research 133
References 140
Index 153
v
PART I
Background
1. Introduction
1.0 BACKGROUND AND MOTIVATION
I first became interested in the small and medium enterprise (SME)

sector in 1989 as the result of a visit to my local branch of the Institute of
Chartered Accountants in Australia. While waiting in the reception area,
a brochure encouraging SME owners to seek the advice of a chartered
accountant caught my eye. The brochure argued that SME owners could
maximize their chances of success (reduce their chances of failure)1 by
seeking the advice of a properly qualified professional. The brochure also
pointed to the extremely high mortality rate for SMEs, as noted in the fol-
lowing quote by the then National President of the Institute of Chartered
Accountants in Australia:
The statistics on the longevity and mortality of small business in Australia show
a very disturbing picture. Nearly half go into receivership within three years
of commencement, and about 80 percent are out of business within ten years.
(Cohen 1987, p.6)
I had two problems with this quote and its use to promote the services of
chartered accountants. First, the extremely high failure rate referred to
did not seem to reflect the experiences of SME owners within my local
community. For example, I was aware of a number of SME owners who
had run very successful businesses over many years and who had eventu-
ally closed their businesses only when they felt it was time to retire. Were
these owners failures because they eventually retired and closed their busi-
nesses? I think not! Second, I was particularly concerned that such a highly
regarded professional body, of which I was a member, might have either
knowingly or unknowingly been using misleading statements to promote
the services of its members. While I strongly believe that SME owners can
benefit from obtaining professional advice, I do not believe the services of
chartered accountants (or any other service provider) should be promoted
on the basis of misleading statements.
Coincidentally, I also happened to be searching for a PhD topic at
that time, and therefore decided that an examination of SME failure
rates could potentially make a significant contribution to knowledge. As
3
4 SME performance
the first step down the PhD trail I undertook a literature review and it
soon became clear that the high mortality rate for SMEs referred to in
the Institute of Chartered Accountant’s brochure was simply reflecting
the consensus opinion at that time. The following quotes illustrate those
views: ‘The odds are well stacked against success, since in the United States
about 50 per cent of new ventures fail in the first two years and only a tiny
minority last ten years’ (Bannock 1981, p.34); ‘There is a lot of statistical
information in the literature on small business failures. The consensus of
opinion seems to be that between 50 and 60 per cent of such ventures fail
within three years of starting’ (Leslie, Magdulski and Champion 1985,
p.27); ‘Four out of five new firms fail within the first five years’ (Phillips
and Kirchoff 1989, p.65); ‘the literature . . . suggests that between one-
third and a half of new firms cease trading in their early years’ (Cromie
1991, p.44); and ‘Every year more than 100,000 new businesses open their
doors. The hazards, however, are so great that 95 percent eventually fail’
(Thankappan and Hammer 1980, p.1).2
However, I found much of the available literature confusing because
of the variety of definitions (or proxies) used to describe SME failure or
success. As a result, and in the absence of any contrary evidence, I felt that
dubious statistics suggesting very high failure rates for SMEs might have
become part of the folklore and received wisdom on this subject. It was
also common at that time (and, unfortunately, still occurs occasionally
today) for SME conference speakers to begin their presentations with a
justification of the importance of their topic based on the very high failure
rates prevalent within the sector.
The following quote further helped to convince me that an examination
of SME failure rates could make a useful contribution to our knowledge
concerning SME performance and the risks involved for the would-be
entrepreneur:
Like the weather, small business failure is the subject of much discussion . . .
But unlike the weather . . . there is . . . a dearth of timely, reliable, and relevant
information on small business failure rates. (Cochran 1981, p.50)
Therefore, while much had been written about SMEs, and in particular
about SME failure rates, reliable statistics on small business failure were
scarce and, as will be seen later, had typically been produced or inferred
from databases designed for other purposes.3 As stated by Scott and Lewis
(1984, p.49) ‘the absence of good statistical evidence leads to the growth
of myths and half truths’ and, as noted by Stanworth (1995, p.59), these
myths get ‘reported by the media, perpetuated by spokespeople for the
industry and subsequently accepted by the wider public’. Without reliable
Introduction 5
information on the subject, these perceptions are permitted to continue

unchallenged, and the ‘danger is that believers, acting in the faith, may
take actions which have unintended consequences in the real world’ (Scott
1982, p.239). Further, policy decisions by governments and others with
an interest in the small business sector based on such perceptions are
likely to be suspect. For instance, the assumed high risk of small business
failure has been cited as justification for the high rates of return demanded
from this sector by bankers and venture capitalists (Phillips and Kirchoff
1989).
This is not to say that the mortality rate for small business could not be
lowered ‘if the proper help is available and accepted’ (Said and Hughey
1977, p.37). Also, the Wilson Committee’s interim report (1979, p.35)
stated that ‘the major source of financial advice to small businesses is their
accountant . . . But in practice this advice appears often to be confined to
questions of audit and taxation’. The Committee recommended that ‘the
accountancy bodies should take steps to ensure that their members are
both equipped and encouraged to take a more active role in providing
adequate advice to their smaller business clients’. Indeed Reynolds (1987)
found that a major factor related to small firm survival was the amount
of attention given to financial matters. Similarly, Potts (1977, p.93) found
that ‘successful companies rely more heavily on accountants’ information
and advice than do unsuccessful companies’.
My PhD, therefore, had three primary objectives (Watson 1995). First,
I wanted to get a better understanding of what might constitute failure
or success within the SME sector, that is, what definition(s) of failure
and success might be the most appropriate. Having decided on the most
appropriate definition(s), the next step would be to measure prevailing
failure rates within the Australian SME sector. Finally, I hoped to dem-
onstrate that failure rates could be reduced, and performance improved,
if SME owners sought (and acted on) appropriate advice from profes-
sional groups such as accountants. So started my journey down the path
of trying to better understand and measure SME performance and, in so
doing, to expose as a myth the belief that SMEs experience failure rates
considerably higher than that experienced by large organizations.
Having undertaken an extensive literature review, the next step in the
process was to gain access to appropriate data that would allow me to
answer the questions I had identified. The problem I confronted, however,
was the almost total absence of any data that could be used for this
purpose. The only body that regularly obtained information (particularly
financial information) from the SME sector in Australia was the Taxation
Department and, because of confidentiality concerns, its data could not be
accessed. As can sometimes happen, one night I awoke from a deep sleep
6 SME performance
with an idea! I would try to access information on SMEs from managed

shopping centres. My logic was that the owners of these centres were big
businesses and, as such, were almost certain to keep records concerning
their tenants. The managers in charge of a managed shopping centre were
also likely to know the reasons surrounding the demise of any of their
clients. Indeed, I later discovered that managers were routinely required
to write reports, normally monthly, on each of their tenants. Further,
given that the success of a managed shopping centre depends largely on
the success of the tenants, I felt that shopping centre managers were likely
to expend considerable effort screening new tenants and providing them
with ongoing support and advice. Subsequent discussions with a number
of shopping centre managers confirmed this belief. Therefore, I would
suggest that the role played by shopping centre managers is similar, in
many respects, to the role that an external accountant (business adviser)
might play. For this reason I expected the failure rates for businesses
located within managed shopping centres to be lower than those applying
to the wider population of SMEs. If this expectation could be confirmed,
it would provide support for the notion that accessing (and acting on)
appropriate advice increases the probability of SME survival.
I should also note that during the course of completing my PhD
I became aware of a growing body of literature expressing the view
that female-controlled SMEs underperformed male-controlled SMEs.
Unfortunately, my PhD was not designed to examine this issue. However,
shortly after completing my PhD the Australian Federal Government
commissioned a substantial longitudinal (four-year) study into the per-
formance of Australian businesses. The SME data collected in that study
were subsequently made available to researchers (in a confidentialized
form), thereby permitting many interesting questions concerning SME
performance (such as gender effects and the role of networking) to be
explored in a way that had not previously been possible. Results from
analysing that data have allowed me, at least for Australian SMEs, to
expose as a myth the belief that female-controlled SMEs underperform
male-controlled SMEs and also to clarify a number of other issues, such
as the benefits of networking and the relationship between growth and the
availability of external funding.
In summary, I hope the material contained in this book will help to
dispel a number of myths related to SME performance. In particular, I
hope to convince the reader that: SMEs do not suffer from excessively
high failure rates; female-owned SMEs do not underperform male-
owned SMEs (when appropriate adjustments and controls are incorpo-
rated into the analysis); SME growth is not limited by a lack of external
funding; female SME owners do not find it more difficult than male SME
Introduction 7
owners to access external funding; and female SME owners are not dis-
advantaged, relative to male SME owners, in terms of their networking
activities.
It should be noted, at the outset, that the focus of this book is on the
individual SME owner and, therefore, implicit in this book is the notion
that ‘failure’ is ‘bad’ and that reducing ‘failure’ rates is ‘good’. However,
from a societal perspective it can be argued that some level of ‘failure’
is ‘good’ because it allows inefficient operators to be replaced by more
efficient operators (Schumpeter 1942). Indeed Knott and Posen (2005,
p.638) found that, within the banking sector, ‘Excess entry and subse-
quent failure increase aggregate industry efficiency.’ Similarly, using real
options reasoning, McGrath (1999, p.16) noted that ‘A high failure rate
can even be positive, provided that the cost of failing is bounded.’
1.1 OUTLINE OF THE REMAINING CHAPTERS
Having provided the background and motivation for this book, Part II
outlines the various issues that need to be considered if we want to get
a better understanding of: what constitutes failure; the rate of failure
within the SME sector (Chapter 2); and how economic and other factors
(such as age of business) are likely to impact reported SME failure rates
under alternative definitions of failure (Chapter 3). Part III compares
male- and female-controlled SMEs on a number of dimensions, such as:
business closure rates (Chapter 4); return on assets (Chapter 5); and risk-
adjusted returns (Chapter 6). Part IV examines the relationship between
external funding and firm growth using both a qualitative (Chapter 7)
and a quantitative approach (Chapter 8). This is followed in Part V by
an examination of the association between networking and firm perform-
ance (Chapter 9) and the differences in networking activities for male and
female SME owners (Chapter 10). Finally, Part VI (Chapter 11) concludes
the book with a summary of the key findings from the earlier chapters and
with some suggestions for future research.
I trust that the material provided in this book will help clarify a
number of important misconceptions relating to SME performance to
ensure that policy decisions by governments, bankers, service provid-
ers and any other groups with an interest in the SME sector are based
on reliable statistical analysis and not on unsubstantiated myths that
have been permitted to flourish in the absence of such evidence. Indeed,
anyone with an interest in SMEs should find the material presented in
the remainder of this book essential to a proper understanding of SME
performance.
8 SME performance
NOTES
1. It should be noted that in the literature (particularly the early literature) it is generally
assumed that firms that have not failed are successful. More recently some researchers
have moved away from a dichotomous definition of success/failure (such as bankrupt/
not bankrupt) to more continuous measures (for example, percentage growth in sales or
return on assets).
2. Further examples of similar comments can be found in Massel (1978) and Scott (1982).
3. Excellent literature reviews are provided by Berryman (1983) and Cochran (1981).
PART II
SME performance
In this section I intend to dispel the myth that SMEs suffer from exces-
sively high failure rates. Chapter 2 looks at how we might consider defin-
ing and measuring SME failure and success and then Chapter 3 examines
the likely impact of age, size, industry and the state of the economy on
reported failure rates. It is important that researchers, policy makers and
others with an interest in the SME sector adopt appropriate definitions of
failure and success or, at the very least, make clear any potential limita-
tions with the definition being used. Otherwise, as noted in Chapter 1,
inappropriate policies or actions might be adopted to the potential detri-
ment of SME owners and the future health of the economy.
For example, if we define failure as bankruptcy proceedings being initi-
ated against a firm, then it is likely that failure rates will increase during
periods of recession (as interest rates increase) and will decrease during
good economic times. However, if the sale of a business is used as the
definition of failure, then good economic times are likely to be associated
with an increase in the level of SME failures (as owners take the opportu-
nity to sell their businesses and retire, or to move into paid employment,
or to look for other opportunities) and periods of recession are likely to
be associated with lower failure rates (because owners will have limited
opportunities to sell their businesses). Such starkly contrasting outcomes
(depending on which failure definition is adopted) could, potentially, be a
source of some confusion for policy makers interested in the health of the
SME sector. Similarly, if as the result of an inappropriate definition being
used to measure SME failure, groups such as bankers are led to believe
that SMEs have very high failure rates, they might be unwilling to lend
to this sector or, if they do so, they might charge an interest rate premium
(Phillips and Kirchoff 1989) or insist on other conditions which are likely
to inhibit new firm start-ups and growth.
10 SME performance
It should be noted that the focus in this section is on dichotomous

measures of failure and success because the early research in this area
was generally constrained to such measures by the available data. For
instance, studies relying on bankruptcy statistics had little choice other
than to define as failed those firms that were placed into bankruptcy.
Implicit in such studies is the notion that firms that have not been placed
into bankruptcy are successful, which clearly might not always be the case
as a business might cease with significant losses to the owner(s) but with no
losses to creditors. Similarly, studies relying on business closure statistics
have little choice other than to define as failed those firms that discontinue
operations. However, in relatively recent studies, Headd (2003) and Bates
(2005) both report that about a third of SME owners considered that their
businesses were successful at the time of closure. In many of these cases the
owners were simply retiring or had found ‘a superior alternative’ (Bates
2005, p.344).
Chapter 2 will now examine a number of alternative definitions of
failure that have been suggested (used) in the literature and discuss some
of the potential problems with each definition. It will also attempt to
provide an indication of the likely rate of SME failure that might be
reported under each definition. Chapter 3 will then examine the likely
impact firm age, firm size, the industry in which the firm operates and the
state of the economy might have on reported failure rates, again depend-
ing on the definition of failure being used.
By the end of this section I trust the reader will be convinced that
SMEs do not have excessively high failure rates and will have a greater
appreciation of the factors that have given rise to this myth.
2. Defining and measuring SME
failure/success
2.0 INTRODUCTION
How to adequately assess SME failure and success has long been a con-
troversial issue because the type of data routinely available to assess the
performance of large businesses has simply not been available for the SME
sector. Cochran (1981) suggested that the lack of a reliable measure of
failure was a major obstacle to understanding and alleviating the causes
of small business failure and Scott and Lewis (1984, p.49) noted that
‘[o]ne practical implication of this is that ill-founded policy must necessar-
ily follow’. Prior to looking more closely at some of the commonly used
indicators of SME failure and success, this chapter will consider various
attributes that might be considered when selecting a measure of perform-
ance for research or other purposes. Later in the chapter, I will discuss the
likely failure rates that might be expected using the various performance
measures suggested in the literature. This will enable readers, based on
the performance indicator(s) they believe to be the most appropriate, to
draw their own conclusions concerning the potential risks SME owners
confront.
2.1 CRITERIA FOR SELECTING A MEASURE OF

PERFORMANCE
Prior to reviewing a number of alternative definitions of failure that have

been used (or suggested) in the literature, it might be useful to consider
some attributes that a definition should possess if it is to be useful in meas-
uring and analysing business failure and success. In particular, the fol-
lowing attributes could be considered: objectivity/verifiability; relevance/
representational faithfulness; reliability/freedom from bias; and simplicity/
parsimony (Watson and Everett 1993).
11
12 SME performance
Objectivity/Verifiability
The use of an objective measure makes replication of results by researchers

working independently easier, and any conclusions are therefore likely to
be more generalizable. By way of contrast, results obtained using a sub-
jective measure are likely to be more difficult to replicate and might not,
therefore, gain the same level of acceptance. For this reason, it is advisable
when choosing a definition of failure to select a measure that is as objective
as possible, that is, a measure that can easily be confirmed by independent
researchers examining the same (or a similar) sample of SMEs.
Relevance/Representational Faithfulness
The selected measure should faithfully represent that which it purports to

describe; otherwise it could be considered irrelevant. There is little point in
having an objective measure that is not relevant. If the measure is irrelevant
then conclusions drawn from the results might be, at worst, misleading
or, at best, disregarded. A further consideration is that different measures
might be relevant to different users. For example, while bankers might be
interested in bankruptcy rates, SME owners might be more concerned with
the return they can expect to achieve on their investment (both financial
and time) or other non-financial rewards, such as ‘proving you can do it’.
Reliability/Freedom from Bias
It is important that the measure selected should, within reason, be free

from bias. As will be seen later, some failure definitions suggested in the
literature are biased against certain types of businesses. For example, larger
businesses are more likely to be placed into bankruptcy while smaller busi-
nesses are more likely to discontinue. Therefore, it is important (if poss-
ible) that the performance measure selected should yield reliable results
across a range of business types and situations.
Simplicity/Parsimony
As a rule, simple measures are less prone to error and should, there-
fore, be preferred to more complex measures. Studies adopting simple/
parsimonious measures are also more easily replicated and, therefore,
potentially more generalizable.
Ultimately, the choice of a failure/success measure is likely to involve a

compromise between the various criteria discussed above. There might not
Defining and measuring SME failure/success 13
be a single measure that has all the desirable attributes and which meets the
needs of all users (such as: credit providers; owners and potential owners;
advisers to small business; and policy makers). For example, it is possible
that some objectivity might have to be sacrificed to obtain a measure that
is more relevant in a particular situation or for a particular user group.
We can now turn to an examination of a number of SME failure
measures commonly referred to in the literature.
2.2 ALTERNATIVE DEFINITIONS OF SME

FAILURE
As noted by Bruno and Leidecker (1988, p.51):
No two experts agree on a definition of business failure. Some conclude that

failure only occurs when a firm files for some form of bankruptcy. Others
contend that there are numerous forms of organizational death, including
bankruptcy, merger, or acquisition. Still others argue that failure occurs if the
firm fails to meet its responsibilities to the stakeholders of the organization,
including employees, suppliers, the community as a whole, and customers, as
well as the owners.
Because there are no formal reporting requirements for the majority of

SMEs, it is difficult (if not impossible) to obtain sufficient reliable informa-
tion to measure their performance in an economic sense, that is, the rate
of return on capital. Instead, most studies have relied on some recorded
event as a surrogate measure of failure (Watson and Everett 1996a). The
two events for which data have been most readily available are the discon-
tinuance (sale or closure) of a business and the initiation of formal bank-
ruptcy proceedings. Between these two extremes, two further definitions
(namely termination of the business to prevent further losses; and failure
to ‘make a go of it’) have been proposed by Ulmer and Nielsen (1947) and
Cochran (1981), respectively. These four potential definitions will now be
examined to see how they perform against the various criteria outlined in
the previous section.
Discontinuance (Sale or Closure)
This definition of failure is the least homogeneous. Fredland and Morris

(1976, p.7) argued that discontinuance is a proxy for SME failure because
it suggests that resources have been shifted to ‘more profitable oppor-
tunities’. However, there are a number of problems with this definition
of failure. As noted by Garrod and Miklius (1990, p.143), ‘In empirical
14 SME performance
studies, it is sometimes not possible to distinguish between change of own-

ership and exit.’ This can result in an extremely broad definition of failure,
which might include as failed, businesses that are sold to make a profit
or because the owner wishes to retire for age or health reasons (Churchill
1952). Examples of studies that have defined failure to include all discon-
tinuances (both discontinuance of ownership and closure of the business)
include: Hutchinson, Hutchinson and Newcomer (1938); Churchill (1952);
Star and Massel (1981); Ganguly (1985); Stewart and Gallagher (1986);
Phillips and Kirchoff (1989); Baldwin and Gorecki (1991); and Williams
(1993). These studies reported average failure rates in the first five years of
life ranging from a low of 31% to a high of 80%. On a per annum basis, the
average reported failure rates varied from 6.5% to 11%.
Examples of studies that have limited their definition of failure to busi-
ness closure (that is, businesses that were sold but continued to operate
were not treated as failures) include: Tauzell (1982); Hamilton (1984);
Price (1984); Birley (1986); Reynolds (1987); Cooper, Dunkelberg and
Woo (1988); Bates and Nucci (1989); Dunne, Roberts and Samuelson
(1989); Dekimpe and Morrison (1991); Bates (1995); Stanworth (1995);
Headd (2003); Forsyth (2005); Box (2008); and Esteve-Pérez and Mañez-
Castillejo (2008). These studies reported average annual failure rates
varying from 3% to 17%. Interestingly, even though this is a much nar-
rower definition of failure, the dispersion in reported failure rates reported
by these studies is far greater than that reported in the previous paragraph
for studies that defined as failed any business that was closed or sold.
It should be noted that discontinuance of ownership as a definition
of failure can be biased against unincorporated businesses (sole traders
and partnerships) because whenever a business that is operating as a sole
trader or partnership is sold it is typically treated as a discontinuance of
one business and the start-up of another (particuarly where databases such
as the UK VAT register are being used). However, a transfer of some or
all of the shares in a company is typically not treated as a discontinuance.
This inconsistency of treatment can lead to a serious bias in which sole
traders and partnerships appear to discontinue (and by implication fail)
more often than incorporated entities.
It should also be noted that where only business closure is used as the
definition of failure, reported failure rates exclude businesses sold to new
owners irrespective of the reason for the sale (that is, even if the business
was running at a loss). To the extent that large businesses are more likely
to be taken over or sold (rather than liquidated) when they are perform-
ing poorly, failure rates reported using this definition will again be biased
against smaller concerns. Also, in many service industries a business
might have to close when the key operator retires. To label this situation
as a failure would clearly be inappropriate. For example, as noted earlier,

both Headd (2003) and Bates (2005) found that a significant number of
businesses closed while successful, calling
into question the use of ‘business closure’ as a meaningful measure of busi-

ness outcome. It appears that many owners may have executed a planned exit
strategy, closed a business without excess debt, sold a viable business, or retired
from the work force. (Headd 2003, p.51)
Similarly, Bates (2005, p.344) notes that business closure ‘is not necessar-
ily rooted in failure or even performance that lags behind expectations;
departure requires only that a superior alternative has become available’.
In summary, using discontinuance as a measure of failure has the
advantage that it can be a relatively objective (verifiable) and simple
measure. However, it might be a biased measure if sole traders and part-
nerships are treated differently from companies. Further, it is difficult to
see how recording as failures all businesses that are closed or sold, irre-
spective of the reason for the sale or closure, is likely to provide useful or
relevant information for a number of key interest groups such as: credit
providers; owners and potential owners; advisers to small business; and
policy makers.
Bankruptcy/Losses to Creditors
Dun and Bradstreet (1979) classify all businesses that are placed into
bankruptcy, or cease operations with resulting losses to creditors, as
failed. The implication is that continuing businesses and businesses that
cease without any losses to creditors (although there might have been
losses to the owners) are regarded as successful (non-failed). This appears
to be a very narrow definition of failure and excludes many businesses that
might commonly be regarded as having failed, for example, businesses
that are barely breaking even and, therefore, not providing a reasonable
income or return for their owners (Land 1975). Examples of studies using
bankruptcy (losses to creditors) to define failure include: Massel (1978);
Cahill (1980); Hall and Young (1991); Lowe, McKenna and Tibbits
(1991); Harada (2007); and Hudson (1997). While reasonably homoge-
neous in terms of the way failure is defined, these studies generally only
examine a cohort of failed firms and typically provide no information on
the overall population of small businesses (that is, failed and non-failed
firms). These studies, therefore, do not usually provide annual failure rate
estimates. For the few studies where annual failure rates are estimated they
range from 0.43% to 1.3%.
As with discontinuance, bankruptcy has the advantage of being an
16 SME performance
objective (verifiable) and simple measure of business failure and is cer-

tainly a relevant measure for credit providers. For other users, however, it
can lack relevance given there might be a large number of businesses that
have ceased trading with substantial losses to the owners but without any
losses to creditors. Few would argue that these businesses were successful
and yet they would not be reported as failures under this definition. In
so far as larger businesses (because of larger borrowings) are more likely
than smaller businesses to be placed into formal bankruptcy, there is also
a potential for this definition to be biased in favour of smaller businesses
and against larger businesses because, in the absence of formal bankruptcy
proceedings being initiated, researchers are unlikely to be able to deter-
mine whether a business closed with losses to creditors.
Disposed of to Prevent Further Losses
Ulmer and Nielsen (1947, p.11) defined as failed ‘firms that were disposed
of (sold or liquidated) with losses to prevent losses’. Losses in this context
include the owner’s capital, and a business could therefore be regarded as
having failed even though there might not have been any losses to creditors.
Defining failure to include businesses that were sold or ceased to prevent
further losses appears more relevant for owners and potential owners,
advisers to small business and policy makers than using a measure based
on either discontinuance or bankruptcy. However, this measure is neither
as simple nor as objective as either bankruptcy or discontinuance as it
requires information from someone associated with the business. Because
such information would not generally be available from external (third-
party) sources, failure statistics reported using such a measure could be
difficult to verify and this might explain its limited use.
Failing to ‘Make a Go of It’
Cochran (1981, p.52) suggested that ‘failure should mean inability to

“make a go of it”, whether losses entail one’s own capital or someone
else’s, or indeed, any capital’. This definition is wider than that suggested
by Ulmer and Nielsen (1947) as it would presumably include as failed any
business that was not earning an adequate return or meeting other owner
objectives. The difficulty with this definition is that most studies have
relied on business closure or sale to trigger the classification of the business
as either failed or non-failed. However, many businesses may continue
operating even though they would be classified as having failed under this
definition. In addition, an adequate return is hard to define, as many small
business proprietors might be willing to accept low financial returns as the
cost of independence, making it difficult (and possibly even inappropriate)

for anyone other than the SME owner to assess a firm’s performance using
this definition.
While this definition of failure appears to be the most relevant (par-
ticularly for owners and potential owners, advisers to small business and
policy makers) it is clearly the least objective and, therefore, results from
studies using this definition are likely to be difficult to verify; this might
explain why it has also been virtually ignored as an SME performance
measure.
In summary, from a review of the literature, there are at least four defi-
nitions (or proxies) that have been used (or suggested) to describe SME
failure and success. At one extreme, all businesses that are sold or cease to
operate are classified as having failed (referred to as discontinuance). At
the other extreme, only businesses that are either placed into bankruptcy
or cease with losses to creditors are considered to have failed. Between
these two extremes, Ulmer and Nielsen (1947) defined failure as termina-
tion to prevent further losses and Cochran (1981) suggested that failure
ought to be recorded only where the owner failed to ‘make a go of it’.
Each definition has appealing attributes. Unfortunately, no one defini-
tion is clearly superior on all the criteria identified as being important in
choosing a measure of failure. Also, different users might be interested in
different measures. As a result, it is difficult (if not impossible) to form a
consensus view regarding which definition is the most appropriate. For
this reason, and because reported failure rates might vary substantially
depending on the definition of failure used, researchers should clearly state
the measure of failure they have adopted and acknowledge any resulting
bias that might result. Similarly, policy makers need to be careful in inter-
preting the results of such studies, particularly if they intend to formulate
policy on the basis of reported findings.
2.3 AN EXAMINATION OF THREE CASE STUDIES
The following examples, based on real businesses from my local com-

munity, will help demonstrate some of the difficulties inherent in the way
failure and success have been conceived of in the past (and possibly still
are today).
Example 1: Watch Repair and Jewellery Shop
A Swiss couple founded this business soon after they immigrated to

Australia. Both husband and wife worked in the business, with the wife
18 SME performance
attending to customers while her husband spent most of his time in the
workshop. This business operated for some 30 years and provided the
couple with a good standard of living and a comfortable retirement. When
the couple were ready to retire they closed the business and the lease for
the premises was taken over by new owners, who established a coffee and
cake shop. Was the watch repair and jewellery shop a failure?
Example 2: Bookshop
A gentleman in his early forties started this business. It was very success-
ful and on the strength of its success the owner was able to borrow a large
sum of money from a bank to invest in a property deal. Unfortunately,
the property deal was not successful and the bank placed the owner (and
the bookshop) into bankruptcy and then sold the bookshop to recover
the monies owing. The proceeds from selling the bookshop were suffi-
cient to cover all the owner’s debts and the business continued under new
management/ownership. Should this business be classified as a failure,
either at the time it was placed into bankruptcy or when it was sold?
Example 3: Fruit and Vegetable Shop
The manager of the shopping centre in which this business was located
felt that the business could be significantly improved. However, the owner
of the business rejected all of the manager’s suggestions. Given that the
rent paid for the premises occupied by this business included a component
based on gross sales, the manager ultimately decided not to renew the lease
and, as a result, the business closed. However, shortly afterwards, the busi-
ness reopened in another location. Should this business be classified as a
failure when it ceased to operate in the managed shopping centre?
These three examples illustrate some of the difficulties inherent in attempt-

ing to determine SME failure and success rates. Table 2.1 illustrates how
each of these businesses is likely to be viewed under the various definitions
of failure discussed above.
The first point to note from Table 2.1 is that all the businesses would be
classified as having failed under at least one of the definitions. The watch
repair and jewellery shop would be recorded as having failed if discontinu-
ance or business closure is the definition of failure being used. However,
the owners of the business (and anyone who knew the business well)
would undoubtedly consider it to have been very successful. Similarly,
if discontinuance of ownership is the performance measure being used,
the bookshop would also be regarded as having failed, although again
Table 2.1 Classification of business as failed/non-failed under the various

definitions of failure
Definition of Failure Watch repair Bookshop Fruit and Overall

and jewellery vegetable failure rate
shop shop (%)
Discontinuance (of Failed Failed Not Failedb 66
ownership and/or
business)
Business closurea Failed Not Failed Not Failedb 33
Bankruptcy Not Failed Failedc Not Failed 33
Disposed of to Not Failed Not Failed Not Failed 0
prevent further
losses
Failing to ‘make of Not Failed Not Failed Failedd 33
go of it’
Notes:
a. Note that business closure is a subset of discontinuance.
b. Assuming the researcher was aware that this business had relocated.
c. This business would most likely be classified as having failed unless the researcher was
able to access detailed information about its performance and the reason why it was the
subject of a forced (bankruptcy) sale.
d. Based on the opinion of the shopping centre manager, which was probably not shared
by the business owner.
there is little doubt that the business itself was very successful. A similar
conclusion regarding the bookshop would almost certainly be reached if
bankruptcy is the performance measure being used, unless the researcher
spent considerable time looking into all the circumstances surrounding the
bankruptcy proceedings and the reason for the forced sale of this business.
Interestingly, the fruit and vegetable shop would not have been considered
a failure under any of the definitions except failing to ‘make a go of it’, and
then only if the opinion of the shopping centre manager (rather than that
of the owner) was sought. In summary, it would seem that if we were to get
an expert opinion on the success or failure of each of these three businesses,
it is most likely that the fruit and vegetable shop would be considered the
least successful. However, this business is the least likely to be recorded
as a failure by researchers using secondary data where they are unable to
access any first-hand knowledge concerning individual businesses and the
reasons for their discontinuance.
The second point to note from Table 2.1 is the wide variation in
reported failure rates (from 0% to 66%) depending on the definition
selected and how that definition is implemented in specific cases. Of
20 SME performance
particular concern is the fact that the first and last definitions in Table
2.1 give exactly opposite results for each of the three examples provided.
I believe these examples are not uncommon, and we should therefore be
very careful in interpreting failure rate statistics derived from secondary
data sources where it has not been possible to obtain any input from a
person (or persons) knowledgeable about the businesses being examined.
This is an issue that should be of concern to policy makers and others
with an interest in the SME sector and suggests that researchers need
to be very careful in selecting and implementing performance measures
within this sector. Further, given that there is unlikely to be one single
performance measure that is the most appropriate in all cases, research-
ers need to carefully articulate any shortcomings in the measure(s) they
adopt.
I will now explore the likely SME failure rates that could be expected
based on the various definitions discussed above.
2.4 ESTIMATED FAILURE RATES UNDER

ALTERNATIVE DEFINITIONS
As noted earlier, lack of data is a major problem in trying to conduct

research on SMEs and this was the first major challenge I faced at the
commencement of my PhD. I was fortunate in gaining the support of
the Building Owners and Managers Association in Australia (BOAMA),
which was a key factor in enabling me to obtain information on over 5000
SMEs operating within 51 Australian managed shopping centres over the
period 1961–90.1 Before looking at the results from analysing this data it
should be noted that this sample is not representative of all Australian
SMEs for two reasons. First, the data is restricted to retail and service
businesses because it is these businesses that are normally located within
managed shopping centres. Second, because managed shopping centres
typically screen potential new tenants and provide ongoing support and
advice to existing tenants, we might expect the failure rates for such busi-
nesses to be lower than that applying to the broader population of SMEs.
With these two caveats in mind, Table 2.2 lists the primary reasons given
(by the shopping centre managers) for the sale or closure of businesses
located within Australian managed shopping centres over the period
1961–90 (see Watson and Everett 1996a for further details).
The first point to note from Table 2.2 is that by far and away the most
common reason for the sale or closure of a business was to realise a profit.
Indeed, if we sum reasons 4, 5 and 6a we can see that over half (1319 out
of 2543) the businesses that were sold or closed could not reasonably be
Table 2.2 Reasons for sale or closure of businesses located within

Australian managed shopping centres 1961–90
Reason for Sale/Closure Number Percent

1. Bankruptcy 179 3.4
2. To avoid further losses 415 8.0
3. Did not ‘make a go of it’ 267 5.1
4. Retirement or ill health 126 2.4
5. To realize a profit 916 17.6
6a. Other – not faileda 277 5.3
6b. Other – failed 34 0.7
2214 42.6
7. Unknown 329 6.3
Total sale or closures 2543 48.9
Continuing businesses 2653 51.1
Total start-ups 5196 100
Note: a. Other included, for example, marriage breakdowns, and in such cases the
shopping centre manager was asked to give his/her opinion as to whether the business had
been successful prior to its sale/closure or whether the owner(s) had failed to ‘make a go
of it’. Based on the manager’s assessment, the business was then classified as failed/not
failed.
Source: Adapted from Watson and Everitt (1996, Table 2).
classified as ‘failures’; a result that was subsequently supported by Headd’s

(2003) analysis suggesting that about 30% of US SME owners felt their
businesses were successful at closure.
Table 2.3 shows that of the 2543 businesses that were closed or sold,
about 42% might be considered ‘failures’ in that they were either: placed
into bankruptcy, ceased to prevent further losses or failed to ‘make a go
of it’. Table 2.3 also shows the annual failure rate recorded under each
failure definition, that is: 0.7% for bankruptcy; 2.3% to prevent further
losses; 4.1% failed to ‘make a go of it’; 9.4% if we include all discontinu-
ances (sale and closures); and 3.9% if we include only closed (rather than
sold) businesses.
Two important points emerge from the results provided in Table 2.3.
First, the reported failure rate can vary substantially depending on the def-
inition adopted; for example, SMEs discontinue at over ten times the rate
they go bankrupt. Second, the annual rate of business closure (3.9%) is
remarkably similar to that for businesses that were sold or closed because
they failed to ‘make a go of it’ (4.1%). This suggests that, in the absence of
Table 2.3 Analysis of reasons for SME sale or closure grouped by failure definition, 1961–90, in Australian managed
shopping centres
Reason for Definition of Failurea

Sale/Closure
Bankruptcy To Prevent Failed to Discont. of Business
Further Losses ‘Make a Go Ownership (sale or Closure (%)
of It’ closure)
1. Bankruptcy 179 179 179 179 114 64
22
2. To avoid further losses 415 415 415 270 65
3. Did not make ‘a go of it’ 267 267 162 61
4. Retirement or ill health 126 37 29
5. To realize a profit 916 152 17
6a. Other – not failed 277 78 28
6b. Other – failed 34 34 23 68
179 594 895 2214 836 38
7. Unknown 166b 329 166 50
Totals 179 594 1061 2543 1002
% of Discontinuances 7 23 42 100 39
% of all businesses (n=5196) 3.4 11.4 20.4 49 19
Average annual failure rate (%) 0.7 2.3 4.1 9.4 3.9
Businesses < 5years old (%) 72 76 75 75 66
Notes:
23
a. Note that all businesses that are sold or closed due to bankruptcy, or to prevent further losses or because the owner(s) failed to ‘make a go of it’
are recorded as discontinuances. That is, the first three definitions of failure are assumed to be subsets of discontinuance. Similarly, bankruptcy
is a subset of both ‘to prevent further losses and ‘failed to “make a go of it”’ and ‘to prevent further losses’ is a subset of ‘failed to “make a go
of it”’. The last two columns of this table provide information with respect to the subset of businesses that were closed (rather than sold).
b. For the 329 businesses where the reason for their discontinuance was unknown I have assumed they failed to ‘make a go of it’ if the business
was closed (rather than sold to new owners). The direction and size of any bias caused by this assumption is unknown.
Source: Adapted from Watson and Everitt (1996, Tables 3 and 7).
24 SME performance
Table 2.4 SME failure rates within first five years of start-up, 1961–90a
(%)
Definition of Failure Years Since Start-up Cum. Average

5 Year Annual
0–1 1–2 2–3 3–4 4–5
Failure Failure
Rate Rate
Bankruptcy 0.7 1.1 1.0 0.7 0.4 3.8 0.8
To prevent further 1.6 2.5 3.3 2.4 2.7 11.9 2.5
losses
Failed to ‘make a go 3.7 6.2 7.7 5.0 6.5 26.0 5.8
of it’
Discontinuance 5.9 10.5 14.4 12.2 15.1 46.3 11.2
Closure of business 2.5 3.9 6.0 4.5 5.8 20.7 4.5
Note: a. Note that firms that commenced during the period 1985–90 are excluded from
the analysis as five years of data post start-up are required for this analysis.
more detailed information, business closure might provide a useful proxy

for the aggregate rate of business failure.2
Table 2.3 also provides information concerning the percentage of
businesses that were less than five years old at the time they were sold or
closed. We can see that approximately 75% of businesses that discontin-
ued were less than five years old, and the same applies to businesses that
were placed into bankruptcy, ceased to prevent further losses or failed to
‘make a go of it’. However, this should not be misinterpreted as 75% of
businesses fail within 5 years! And yet, it is exactly this sort of misinter-
pretation that has contributed to the myth that SMEs have excessively
high failure rates. For example, Potts (1977, p.2) noted that ‘[i]n 1973,
57 percent of all failing concerns in the United States had been in opera-
tion five years or less’. Later Potts (p.9) went on to say that: ‘As has been
discussed previously, more than half of all companies fail in the first five
years of business.’ Clearly the second sentence does not follow from the
first. The first sentence is only commenting on the age of the subset of
business failures. It says nothing about the overall failure rate. The subset
of business failures might be a very small proportion of the population of
all businesses.
Table 2.4, which extends the analysis presented in Table 2.3 by focusing
on young businesses only (those within five years of start-up), will help to
further illustrate this point. Although Table 2.3 indicates that approxi-
mately 75% of businesses that fail are less than five years old, we can see
from Table 2.4 that the five-year cumulative failure rate is significantly less
than 75% for all the failure definitions being considered. For example, if
we use business closure as our definition of failure, we can see from Table
2.4 that 20.7% of Australian businesses located within a managed shop-
ping centre failed within five years, or, alternatively, almost 80% of busi-
nesses survived beyond five years. These results clearly indicate that the
overwhelming majority of Australian SMEs are likely to survive beyond
five years.
Comparing the average annual rate of business closure in Table 2.3 for
all businesses (3.9%) with the rate in Table 2.4 for businesses less than
five years old (4.5%) we can see, as expected, that the closure rates are
higher for firms in their first five years of existence. This finding confirms
the expectation that firms are most vulnerable in their early years as they
learn about their industry. ‘The efficient grow and survive while the ineffi-
cient decline and fail’ (Jovanovic 1982, p.649). There have been numerous
studies confirming this proposition (referred to as the ‘liability of newness’,
Stinchcombe 1965), for example: Freeman, Carroll and Hannan (1983);
Stewart and Gallagher (1986); Evans (1987); Bates and Nucci (1989); and
Dunne, Roberts and Samuelson (1989).
2.5 FAILURE RATE COMPARISONS
In this section I will examine a number of studies that can provide com-
parative statistics of the rate of SME bankruptcy and closure for different
periods and geographical settings. I am not aware of any comprehensive
studies that can provide failure rate comparisons using either ‘failed to
“make a go of it”’ or ‘ceased to prevent further losses’ as a definition of
failure. Also, discontinuance (sale or closure), although often used in early
studies, is now seldom considered an appropriate measure of SME failure
and, therefore, will not be examined any further.
Bankruptcy Rates
Most of the studies on business bankruptcy do not provide any infor-

mation concerning bankrupt firms as a percentage of the population
of all businesses. A notable exception is Hudson (1997), who reported
that for all UK companies during the period 1950–90 the incidence of
company liquidations (including voluntary liquidations) ranged between
0.4% and 2.5%. For the same period in the US, Hudson (1997) reported
that the rate of bankruptcy and/or loss to creditors ranged between 0.2%
and 1.3%. Similarly for Belgian businesses, Dewaelheyns and Van Hulle
(2008) reported a mean annual bankruptcy rate of 1.8%. The 0.7% average
26 SME performance
annual bankruptcy rate for Australian SMEs located in managed shop-

ping centres appears consistent with the rates reported for the US, the UK
and Belgium.
These results highlight the fact that the rate at which SMEs are forced
to exit with losses to both creditors and owners is quite low. For example,
Harada (2007) reported that only 2.3% of Japanese businesses that ceased
trading did so because of bankruptcy; Garrod and Miklius (1990) reported
that business bankruptcies in the US represented only 9% of all discontin-
ued businesses; and for Australian SMEs located within managed shop-
ping centres, business bankruptcies represented 7% of all discontinued
businesses (see Table 2.3). Therefore, as noted by Stewart and Gallagher
(1986, p.46), ‘It is clear that the majority of firms that cease trading do not
do so because they are forced out of business through liquidation or bank-
ruptcy. The majority simply choose to stop trading, the owners changing
to another activity.’
Closure Rates
As noted earlier, in the absence of detailed information at the level of the

individual firm, business closure would seem to provide an appropriate
indication of the overall rate of SME failure. I will now examine four com-
prehensive studies that focused on business closure.
The first study is a large-scale longitudinal survey of Australian employ-
ing businesses commissioned by the Australian federal government in
an attempt to remedy the shortage of reliable data on Australian firms.
The Australian Bureau of Statistics’ (ABS) Business Register was used
as the population frame for the surveys. All employing businesses in the
Australian economy were included in the scope of the survey except for
businesses in the nature of: government enterprises; libraries; museums;
parks and gardens; private households employing staff; agriculture, for-
estry and fishing; electricity, gas and water supply; communication serv-
ices; government administration and defense; education; and health and
community services. Data collection was through self-administered ques-
tionnaires distributed by the ABS.3 Because the ABS can legally enforce
compliance with its data requests (under the Census and Statistics Act
1905) response rates were very high (typically in excess of 90%).4 For con-
fidentiality reasons, information on all large businesses (those employing
more than 200 people) was excluded from the data set made available to
researchers outside the ABS.
Excluding businesses that had no income (sales or other income), 8375
SMEs were surveyed in the first year (1994–95),5 with 5030 of these busi-
nesses (representing approximately 1.25% of eligible Australian SMEs)
Table 2.5 Australian (ABS) SME closure rates 1995–98
Details of Business Closures No.

Businesses active in 1994–95 5030
Businesses closed in 1995–96 487
% of businesses closed 10
Average annual closure rate (%) 8
targeted for follow-up surveys in each of the three subsequent years.6 The
closure rates for these businesses in the subsequent three survey periods
are presented in Table 2.5, which reports an average annual closure rate
for Australian SMEs of 8%. While this rate is almost double the 3.9%
reported in Table 2.3 for Australian businesses located within managed
shopping centres, the higher rate for the broader population of Australian
SMEs is not unexpected given that SMEs operating within a managed
shopping centre would normally have been screened prior to start-up
and, in many cases, would receive ongoing advice and support.7 Again, it
should be emphasized that the 8% closure rate reported above should not
be interpreted as representing the failure rate for Australian SMEs because
many of these businesses are likely to have been successful at the time of
their closure. For example, many of them will have closed because the
owner(s) decided it was time to retire.
The second study of business closures I would like to discuss is a com-
prehensive US study by Headd (2003, p.51) that sought ‘to challenge
the widely held but often unsubstantiated belief that new firm closure rates
are high and that a closure is a negative outcome’. Headd estimates that
49.6% of US employing businesses that commenced operations during
the period 1989–98 closed within four years. This represents an average
annual closure rate of approximately 16% for firms less than four years
old. This closure rate for US employing businesses is double the rate
shown in Table 2.5 for Australian employing businesses. However, it is
important to note that the rate for the US relates only to newly formed
businesses and, therefore, it is reasonable to expect that this rate would be
28 SME performance
significantly higher than that applying to all businesses (old and young).
This issue will be explored further in the following chapter.
The third study of business closure I would like to discuss is a longitu-
dinal study undertaken by Box (2008) that included 2154 Swedish joint-
stock companies in seven birth cohorts that commenced in various years
between 1899 and 1950. Box’s results indicate that, on average, 75% of
firms survived beyond four years.8 Box also reports that the survival rates
varied significantly across the cohort groups, most likely as the result of
different environmental (economic) forces. While Box’s (2008) average
four-year survival rate is considerably higher than the 50% reported by
Headd (2003) for US firms, it should be noted that Box’s sample was of
incorporated businesses and it is reasonable to expect that they will have
higher survival rates (lower closure rates) because they are likely to have
undergone greater scrutiny prior to start-up compared to firms that begin
as sole traders or partnerships. Interestingly, the four-year survival rate of
75% reported by Box (2008) is similar to the 80% five-year survival rate for
Australian SMEs located within managed shopping centres, as shown in
Table 2.4.
The final study of business closures I would like to discuss was carried
out by Forsyth (2005), who examined a cohort of 4103 small firms that
started operations in 1992 (the beginning of a period of economic expan-
sion) in one of Washington’s 27 rural counties. Forsyth followed this
cohort of firms from their inception to 2000 (the peak of the economic
expansion), distinguishing between employing and non-employing busi-
nesses. Forsyth (2005) reported a 60% four-year survival rate on average
for all rural firms. This survival rate is reasonably similar to the 50% four-
year survival rate reported by Headd (2003), particularly given that the
period referred to by Headd included the 1990–91 recession. However,
Forsyth’s (2005) survival rate varied considerably by employment status:
from 56% for non-employers to 77% for firms with at least one employee.
2.6 SUMMARY
Each of the failure definitions reviewed in this chapter has appealing

attributes; however, no one definition stands out as being clearly superior.
It should also be noted that different users might be interested in different
measures. For example, banks might be interested in the rate of bank-
ruptcies in the SME sector. SME owners, on the other hand, might be
more concerned with the proportion of businesses that are closed or sold
because the owners failed to ‘make a go of it’.
Table 2.6 provides a summary of the bankruptcy and business closure
Table 2.6 Summary bankruptcy and closure rates from selected studies
Author(s) Country/Sample Reported Failure Rates

Bankruptcy Closure
Watson and Australia: 5196 All firms: 0.7% All firms: 3.9% p.a.
Everett (1996) SMEs located p.a.
within managed
shopping centres
ABS Data Australia: All firms: 8% p.a.
(unpublished) 5014 randomly
selected SMEs
Hudson (1997) UK: All firms Approx. 1% p.a.
from 1950–90
US: All firms Approx. 0.7%
from 1950–90 p.a.
Headd (2003) US: 12 185 new Employing firms:
firms <4yrs: 16% p.a.
(estimated)
Forsyth (2005) US: 4103 new All firms <4yrs:
small rural firms 10% p.a.
Employers: 5.75%
p.a.
Non-employers:
11% p.a.
Box (2008) Sweden: 2154 Firms <4yrs: 6.25%
joint-stock p.a.
companies
Dewaelheyns and Belgium: Approx 1.8%
Van Hulle (2008) Registered small p.a.
firms 1986–2002
rates from the selected studies discussed in this chapter. Interestingly,

there is a reasonable level of consistency in the bankruptcy rates reported
across the different countries and periods. Similarly, there is some consist-
ency in the reported closure rates, although employing firms, incorporated
firms, older firms and firms located within a managed shopping centre all
appear (not surprisingly) to have substantially lower closure rates.
This issue is examined further in the following chapter, which looks at
the possible effects of firm age, size, industry and the state of the economy
on reported failure rates, depending on which failure definition is being
used. Before encouraging governments to direct resources to the preven-
tion of SME failure, we should first be satisfied that the current rate of
30 SME performance
failure, particularly for small business, is unacceptably high. Government

assistance, whether to very small businesses or to particular industry
sectors, should not be based on incorrect conclusions resulting from
inappropriate definitions of failure.9
NOTES
1. Note that no large retail outlets were included in the sample.

2. However, I feel I must once again emphasize that business closure does not necessar-
ily indicate failure because many successful business are closed, for example, when the
owner(s) want to retire (see Headd 2003) or when the owners wish to pursue other, more
promising, opportunities (Bates 2005). At the same time, lack of closure does not neces-
sarily indicate success as the owner(s) of some marginal businesses might continue their
business because of limited alternatives (Gimeno et al. 1997).
3. Copies of the questionnaires can be obtained from the ABS.
4. A non-response normally meant the ABS was unable to locate the business proprietor
(or the business) and this was, therefore, treated as a business closure (failure).
5. Note that in Australia the tax year runs from 1 July to 30 June.
6. To ensure the sample remained representative, additional businesses were added each
year to replace businesses that had ceased operations.
7. It should be noted that the time periods and industries represented in these two studies
are not the same and this might also have contributed to their differing closure rates.
8. Note that the survival rate is simply 1 − the closure rate. It should also be noted that in
Box’s (2008) study, if a small firm merged with a larger firm the small firm was deemed to
have closed (failed).
9. For example, Lowe, McKenna and Tibbits (1991), based on bankruptcy statistics, ques-
tioned the provision of government assistance to the manufacturing sector because this
sector has a higher ‘failure’ rate compared to the retail and service sectors. However, the
higher failure rate exhibited by manufacturers could simply be a function of the defini-
tion of failure used.
3. The effects of age, size, industry and
the economy on SME failure rates
3.0 INTRODUCTION
The previous chapter reviewed some of the more commonly used measures
of SME success and failure and showed how reported failure rates can
vary substantially depending on the definition of failure used. The broader
the definition, the higher the likely failure rate; the narrower the definition,
the lower the likely failure rate. While each of the definitions has certain
strengths, none is clearly superior. Further, the results of prior studies
suggest that reported failure rates (even where the same definition is used)
can vary substantially. For example, the annual bankruptcy rates reported
in Chapter 2 range from 0.7% to 1.8%, with business closure rates ranging
from 3.9% to 16%. This large variance in reported failure rates must surely
confuse policy makers and others interested in the SME sector. This
chapter will consider some of the potential reasons for these differences, in
particular, age, size, industry and the state of the economy.
3.1 THE EFFECT OF AGE
Most studies of SME failure have found that a business is at greatest risk
in its first few years. Jovanovic (1982) argued that younger firms are more
likely to fail because they face greater variability in their cost functions
while they learn about their industry and management capabilities. The
efficient grow and survive; the inefficient decline and fail. For this reason,
younger firms are less likely to survive than older ones. There have been
numerous studies confirming this proposition: Stewart and Gallagher
(1986); Evans (1987); Bates and Nucci (1989); and Dunne, Roberts and
Samuelson (1989). I will now look at how failure rates can be expected to
vary with firm age.
For each failure definition, Figure 3.1 plots the average annual failure
rate by age of business for my sample of 5196 business start-ups operating
within 51 Australian managed shopping centres over the period 1961–90,
as described in Chapter 2. As can be seen from this figure, the failure rates
31
32 SME performance
18
16
Discontinuance of ownership
14
Probability of failure (%)
12
10
8
Failed to ‘make a go of it’
6
Discontinuance of business
4
To prevent further losses
2
Bankruptcy
0
0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5
Age (years)
Source: Everett and Watson (1998, p.381).
Figure 3.1 Probability of failure as a function of age
generally reach a maximum at around three years of age and then gradu-
ally decline. Brüderl, Preisendörfer and Ziegler (1992, p.234) note that
this inverted U-shaped curve can be explained by the fact that ‘[m]ortality
rates are low immediately after starting a business because organizations
can survive on initial resources, increase to a maximum, and decline after-
wards’. This mortality process has been labelled the ‘liability of adoles-
cence’, in contrast to ‘liability of newness’, which depicts monotonically
declining failure rates (Brüderl and Schussler 1990). It should be noted
that Box (2008), Cressy (2006) and Ganguly (1985) also report failure rates
conforming to a bell-shaped distribution.
Also note that the failure rate distributions reported in Figure 3.1 for
‘discontinuance of business (closure)’ and ‘failed to “make a go of it”’ are
very similar. This finding further supports the observation in Chapter 2
that (at an aggregate level) where an objective measure of SME failure is
required, business closure might be an appropriate proxy. However, as
explained in Chapter 2, this measure is likely to overstate the ‘true’ SME
failure rate because many businesses close while successful (Headd 2003).
Given that failure rates appear to peak during the first five years of
a firm’s existence, Table 3.1 reports average annual failure rates for the
Effects of age, size, industry and the economy 33
Table 3.1 Average annual failure rates by age classification (%)
Firm Age (years)

0–5 5–10 10+ Av.
Bankruptcy 0.8* 0.6 0.3** 0.7
Prevent further losses 2.9** 1.8** 1.0** 2.4
Failed to ‘make a go of it’ 5.0** 3.2** 2.0** 4.2
Discontinuance of ownership 10.4 9.8 6.9** 10.0
Discontinuance (closure) of business 4.3* 4.0 2.6** 4.0
Notes:
* Significantly different from average failure rate at 5% using a one-tailed test.
** Significantly different from average failure rate at 1% using a one-tailed test.
Source: Watson (1995, Table 6.8).
following three age classifications: less than five years, five to ten years
and over ten years. As expected, the results generally indicate significantly
higher than average failure rates for the cohort of firms less than 5 years
of age, with lower failure rates being reported for the two older age groups
and particularly for those firms over 10 years of age.
Turning again to the representative sample of 5030 Australian SMEs
surveyed by the Australian Bureau of Statistics (ABS) in the four consecu-
tive years from 1995 to 1998 (as described in Chapter 2) Table 3.2 also
presents the annual closure rates for three age groups: less than five years;
five to ten years; and over ten years. From Table 3.2 it appears that age has
an even more pronounced effect on the closure rates for the ABS sample,
compared to those reported in Table 3.1 for businesses located within a
managed shopping centre. The results presented in Table 3.2 indicate that
the average annual closure rate for Australian SMEs less than five years
old (16%) is over three times the closure rate for SMEs over ten years old
(5%). For businesses located within a managed shopping centre, however,
the average annual closure rate for businesses less than five years old
(4.3%) is less than double the closure rate for businesses over ten years
old (2.6%). It would seem reasonable to suggest, therefore, that the fact
that businesses located within a managed shopping centre are typically
screened and receive ongoing advice and support might have a significant
impact both on their overall survival rate and, more particularly, on the
survival rate of younger businesses. This finding has implications for gov-
ernment policy and highlights the important role service providers, such as
accountants, might play in potentially minimizing early stage failure rates
within the SME sector.
34 SME performance
Table 3.2 Australian SME closure rates 1995–1998
Details of Business Closures Firm Age

<5 yrs 5–10 yrs >10 yrs Totals
a
Businesses active in 1994–95 1563 1224 2243 5030
Businesses closed in 1995–96 204 124 159 487
% of businesses closed 13** 10 7** 10
% of businesses closed 16** 6 3** 7
% of businesses closed 22** 5** 4** 8
Average annual closure rate (%) 16** 7** 5** 8
Notes:
a. Note that in Australia the financial/tax year runs from 1 July to 30 June.
** Significantly different from overall (total) closure rate at 1% using a two-tailed test.
The findings presented in Table 3.2 emphasize the need (when reporting,
interpreting and comparing failure rates) to provide information about the
age distribution of the SMEs under investigation and to control for age in
any analysis being conducted.
3.2 THE EFFECT OF SIZE
Much of the SME literature assumes that the probability of failure

increases as the size of a business decreases, and that SME failure rates
are unacceptably high. However, Jovanovic (1982) argued that the inverse
relationship between firm size and the rate of SME failure might more
accurately be characterized as an inverse relationship between age of
business and the rate of failure. He argued that firms learn about their
efficiency as they operate in an industry. The efficient grow and survive;
the inefficient decline and fail. For this reason, older businesses are more
likely to survive and to be larger businesses. Therefore, size might simply
be a proxy for age and there might be little or no relationship between the
size of a business and its propensity to fail, after controlling for age.
While some studies subsequent to Jovanovic have confirmed the
expected positive association between failure and age (see, for example,
Stewart and Gallagher 1986; Evans 1987; Bates and Nucci 1989; Dunne
Table 3.3 Australian SME closure rates 1995–96 to 1997–98 by size (%)
Firm Age Firm Size (employees)

1–20 >20
< 5 years 16 16
5–10 years 7 7
> 10 years 5 4*
Average annual closure rate 9 7**
Notes:
* Significantly less than small business closure rate at 5% using a one-tailed test.
** Significantly less than small business closure rate at 1% using a one-tailed test.
et al. 1989), they have generally found that a size effect persists even after
controlling for age. Using the ABS data to further explore the relation-
ship between business closure and both size and age, Table 3.3 reports
separately the average annual closure rates for Australian small (20 or
fewer employees) and medium-sized businesses (more than 20 employees).
From the results presented in this table it is apparent that age of business
has a far greater impact on average annual closure rates than does size of
business (see the significance results for age reported in Table 3.2).
The results presented in Table 3.3 provide some support for Jovanovic’s
(1982) argument because for the two younger age groups there is no differ-
ence in the average annual closure rates of the small versus the medium-
sized firms. It is only in the oldest age group that there is a significant
difference in the closure rate by size of business. The fact that the overall
closure rate is significantly higher for the smaller firms is largely a function
of there being disproportionately more small firms in the youngest age
group, where the closure rate is much higher than average.
The fact that for SMEs over ten years old the closure rate appears higher
for smaller firms could be the result of closure rates being biased against
smaller firms and, therefore, the adoption of an alternative (more appropri-
ate) definition of failure could result in a finding that supports Jovanovic’s
(1982) proposition for all age groups. Similarly, the fact that many prior
studies have reported significant differences in failure rates by size of busi-
ness could also be a function of the failure definitions used by those studies.
For example, using either discontinuance of ownership or business
closure as a definition of failure is biased against smaller firms. The reason
for this is that, compared to larger businesses, smaller businesses typi-
cally have lower start-up and closure costs, and a greater dependency on
the life cycle of their owners.1 As noted by Hutchinson, Hutchinson and
36 SME performance
Newcomer (1938), enterprises requiring little capital are likely to have

the highest discontinuance rates. They argue that when large amounts of
capital are at stake, the owners are likely to make a more thorough inves-
tigation of the prospects for the new enterprise. It should also be noted
that corporate transfers of ownership are typically treated differently from
transfers of ownership by sole traders or partnerships. Whenever a sole
trader or partnership sells a business it is generally treated as the discon-
tinuance of one business and the start-up of another; on the other hand,
a transfer of shares in a company (even if all the shares are transferred) is
generally not treated as a business discontinuance (Star and Massel 1981).
This inconsistency of treatment can lead to a serious bias in which sole
traders or partnerships appear to discontinue (and by implication fail) at a
higher rate than corporate entities.
Alternatively, using bankruptcy as a definition of failure should result
in higher reported failure rates for larger businesses because, on average,
they are likely to have relatively larger commitments and greater tangible
assets. Creditors are more likely to pursue bankruptcy proceedings where
the amounts owed are relatively large and where tangible assets exist.2
Therefore, this definition of failure is likely to be biased in favour of small
businesses and against larger businesses.
For the remaining two definitions of failure (‘to prevent further losses’
and ‘failed to “make a go of it”’) there is no reason to believe that either
would be biased for or against smaller firms. It should be noted that the
definition of failure used by researchers has, to a large extent, been justi-
fied based on the available data rather than any theoretical foundation.
Consistent with the expectations discussed above, Table 3.4 shows
that businesses located within larger managed shopping centres are more
likely to declare bankruptcy, but less likely to discontinue or close, than
businesses located within smaller shopping centres. For the other two
definitions of failure (‘to prevent further losses’ and ‘failed to “make a
go of it”’) there is no difference by size of shopping centre. Assuming
that the size of a shopping centre is a reasonable proxy for the size of the
businesses operating within the centre, this finding supports Jovanovic’s
(1982) proposition that age rather than size is the key determinant of SME
failure, provided failure is measured appropriately.
The implication from Table 3.4 is that researchers need to be mindful
of any biases inherent in their choice of failure definition and should con-
sider controlling for both the age and size of business. Similarly, policy
makers (and others with an interest in the SME sector) need to be wary of
studies that do not control for age and size of business, particularly where
the definitions of failure and success are likely to be biased for or against
smaller businesses.
Table 3.4 Comparing failure rates for SMEs located in larger and smaller
shopping centres (%)
Shopping Centre Size

Larger Smaller
Bankruptcy 0.8 0.6*
To prevent further losses 2.4 2.2
Failed to ‘make a go of it’ 4.1 4.1
Discontinuance 9.0 10.2*
Closure of business 3.7 4.3*
Notes: * Significantly different from average failure rate in larger shopping centres at 5%
using a two-tailed test.
Source: Adapted from Watson and Everett (1996b, p.279).
3.3 THE EFFECT OF INDUSTRY
Many studies examining the incidence of SME failure have reported sig-
nificant variations in failure rates between industry sectors. Furthermore,
the results from some studies are in direct conflict. For example, Lowe,
McKenna and Tibbits (1991) and Fredland and Morris (1976) examined
SME bankruptcies and reported that the manufacturing sector had the
highest failure rate, while Brüderl, Preisendörfer and Ziegler (1992) and
Phillips and Kirchoff (1989) looked at closure rates and found that the
manufacturing sector had the lowest failure rate. The significant variations
in reported failure rates and the apparent conflict between the findings of
some studies must surely be a source of confusion for policy makers and
others with an interest in the SME sector. To try to better understand the
conflicting results referred to above, I will now explore the likely associa-
tion between industry and reported failure rates for various definitions of
SME failure.
Stewart and Gallagher (1986) suggested that firms in sectors which
involve high capital costs are likely to have higher levels of liquidations
and bankruptcies and will find it harder to cease trading and simply
switch to something else. These comments suggest that using bankruptcy
as a measure of failure is likely to be biased against industries with high
start-up costs (capital requirements) and in favour of businesses with
lower start-up costs. Conversely, using closure as a definition of failure
is likely to be biased in the opposite direction, namely, in favour of busi-
nesses with high start-up costs and against businesses with low start-up
38 SME performance
High
Bankruptcy
Failure rate
Failed to ‘make
a go of it’
Business closure
Low
Low Start-up costs High
Source: Adapted from Watson and Everett (1999, p.34).
Figure 3.2 Expected relationship between failure rates and start-up costs
costs. Hutchinson, Hutchinson and Newcomer (1938) argued that in

those industries that require little initial capital, we can expect greater
competition and, therefore, higher closure rates. Using a more appropri-
ate definition of failure, such as ‘failed to “make a go of it”’, is less likely
to be influenced by a firm’s start-up costs. Figure 3.2 depicts the suggested
relationship between SME failure rates and the level of start-up costs, for
these three definitions of failure.
The implication from Figure 3.2 is that researchers need to be mindful
of any biases inherent in their choice of failure definition with respect
to different industry sectors. Further, where SMEs across a range of
industries are being examined, it will be important to control for industry
differences.
3.4 THE EFFECT OF THE ECONOMY ON SME

FAILURE RATES
There are three types of risk that can ultimately impact a venture’s chances
of success: those that are unique to the firm (for example, the experience of
the owner); those that relate to the industry in which the firm operates (for
Table 3.5 Significant relationships between macro-economic indicators

and SME failure
Interest Employment Retail Retail Unemployment

rate rate (lagged sales sales rate (lagged 6
(current) 6 months) (current) (lagged 6 months)
months)
Bankruptcy Positive
To prevent further Positive
losses
Failed to ‘make a Positive Negative
go of it’
Discontinuance Positive Negative Positive
Business closure Positive Negative
Source: Adapted from Everett and Watson (1998, p.386).
example, whether the industry is in decline); and those that relate to the
state of the economy (for example, unemployment rates). There is little an
SME owner can do to protect a business from economic downturns and,
as noted by Fredland and Morris (1976, p.9), during ‘cyclical downturns
the marginal firm is more likely to fail’.
Following the call by Shailer (1989) to pay more attention to external
(exogenous) variables, such as interest rates and various other economic
indicators, Everett and Watson (1998) modelled SME failure (using
various failure definitions) to determine just how significant various
macro-economic factors are to SME mortality. Evidence from the US
reported by Sharpe (1981) suggests that market-based risk (as opposed to
firm- or industry-based risk) represents approximately 25% of the total risk
associated with listed companies. Similarly, Foster (1986, p.199) reported
that on average, in the US, external factors (industry and economy)
explain about 43% of the variation in business net income.
The results of modelling failure rates against various macro-economic
variables for my sample of 5196 business start-ups operating within 51
Australian managed shopping centres over the period 1961–90 are sum-
marized in Table 3.5. The table reports only those macro-economic vari-
ables that were found to be significant and, for each significant variable,
whether the relationship with SME failure was positive or negative. As can
be seen from the table, retail sales for the current six-month period and
retail sales lagged six months are the only macro-economic indicators to
feature in more than one model (failed to ‘make a go of it’; discontinuance
40 SME performance
of ownership; and business closure). This is not surprising since retailers

comprised the majority of businesses in the sample.
When bankruptcy is used as the definition of failure, the current interest
rate is the only significant macro-economic variable in the model. Given
that many of the businesses in the sample would probably have required
substantial borrowings to locate within a managed shopping centre, the
positive association between interest rates and SME bankruptcy was
expected. For these businesses, interest rates would have a major effect on
operating costs and, therefore, on their chances of survival. Where busi-
nesses with substantial borrowings do fail, they have a high probability
of being placed into bankruptcy. The positive association between inter-
est rates and bankruptcy supports the earlier findings by Hall (1986) and
Wadhwani (1986).
The six-month lagged employment rate is the only significant macro-
economic variable associated with businesses that are sold (or closed) to
prevent further losses. However, the direction of the association (posi-
tive) is not as one might expect. It appears that the probability of failure
increases with increases in the lagged employment rate; that is, as employ-
ment opportunities improve, more SME owners exit their ventures. This
suggests that a significant number of SME owners remain in marginal
businesses until the opportunity to exit improves, because the alternative
might be unemployment.
A similar conclusion can be drawn from the results for the remaining
three definitions of failure. In each case, current period retail sales are sig-
nificantly positively related to failure. This again suggests that owners of
marginal businesses might delay exiting their business until a time when
macro-economic factors improve, thereby maximizing both the opportu-
nity to sell their business and the price for which the business can be sold.
Note that the negative sign for retail sales lagged one period only arises
when both retail sales and retail sales lagged are included in the failure pre-
diction model. Individually, both variables are positively related to failure.
However, when both variables are included, the sign for retail sales lagged
becomes negative as the model seeks to find the best fit to the failure data.
This can be interpreted as follows: for a given level of current sales, if past
sales have been poor there is a greater chance of failure in the current period.
This indicates that while poor past sales are positively related to failure (and,
therefore, could be seen as the cause of failure) good current sales appear to
be the trigger for marginal businesses to be either sold or closed.
When failure is defined as the sale or closure of a business for any
reason (discontinuance of ownership) the rate of unemployment lagged
six months also enters the model as a significant explanatory variable. The
lagged unemployment rate is positively associated with the rate of failure.
There are at least two possible causes for this relationship. First, a high
unemployment rate might indicate problems in the economy which, in
turn, could lead to an increase in SME failure. Second, a high unemploy-
ment rate could result in an increase in the demand for self-employment
and, therefore, greater opportunities to sell both marginal and successful
businesses.
In summary, the findings presented by Everett and Watson (1998) indi-
cate that, on average, macro-economic factors appear to be associated
with about 30% to 50% of SME failures, depending on the definition of
failure used. As expected, failure was positively associated with interest
rates (where failure was defined as bankruptcy) and the rate of unem-
ployment (where failure was defined as discontinuance of ownership).
However, failure was also positively associated with lagged employment
rates (where failure was defined as to prevent further losses) and with retail
sales (where failure was defined as: failed to ‘make a go of it’; discontinu-
ance of ownership; or discontinuance of business). These findings are con-
sistent with Bhattacharjee et al. (2009), who reported, for a large sample
of UK companies, that bankruptcies were associated with economic
downturns and acquisitions with economic upturns.
These results suggest that many businesses are sold, or cease, volun-
tarily, with their proprietors able to time their exit to take advantage of
prevailing economic conditions. That is, some of the macro-economic
variables found to be significant in this study cannot be viewed as causing
failure, but rather they seem to provide the trigger for SME owners to take
the opportunity to sell or close their businesses. Thus, depending on the
definition of failure adopted, a positive economic outlook might be asso-
ciated with an increase in the rate of SME failure. Policy decisions made
in the absence of a sound understanding of how various macro-economic
variables are likely to impact SME failure rates (under various definitions
of failure) could, therefore, be suspect. Further, without a clear under-
standing of the relationship between key economic indicators and the
various definitions of failure, accurate evaluations of government policies
and programs designed to help SMEs will be problematic.
3.5 SUMMARY
Chapter 2 examined various definitions of SME failure, and from the

available evidence it is clear that reported failure rates vary considerably
across studies even where the same failure definition is used. This chapter
considered some of the more likely reasons for such variations, in particu-
lar: age, size, industry and the state of the economy.
42 SME performance
From the available evidence, there appears to be a compelling case to

suggest that failure rates peak at around three years of age, irrespective of the
definition being used. It is important, therefore, when comparing and analys-
ing SME failure rates, that researchers control for the age of the business.
In terms of the size of the business, the results indicate that some failure
definitions can be biased either for or against smaller, compared to larger
businesses. In particular, bankruptcy appears to be biased against larger
businesses, while discontinuance is biased against smaller firms. This sug-
gests that, when comparing and analysing SME failure rates, researchers
should control for both age and size of business and should make clear
any limitations associated with the failure definition being used. Given
that the majority of past studies have used discontinuance as their measure
of failure, this is the most likely reason why the myth that SMEs have
unacceptably high failure rates has become established as part of the folk-
lore on this subject.
Similarly, it would appear that certain failure definitions are likely to be
biased either for or against certain industry sectors. In particular, indus-
tries with significant start-up costs are likely to report higher bankruptcy
rates but lower discontinuance rates. Conversely, industries with relatively
small start-up costs are likely to report lower bankruptcy rates but higher
discontinuance rates.
Finally, in terms of the effect of macro-economic factors on the rate of
SME failure, the evidence again suggests some potentially confounding
signals depending on the definition of failure being used. For example,
as expected, the rate of SME bankruptcies is positively related to interest
rates. However, for all the other failure definitions examined, the evidence
suggests that an improvement in the economy can provide the trigger for
SME owners to move out of self-employment. This suggests that policy
makers need to exercise considerable care in assessing the impact of any
measures introduced to stimulate the economy as a means of promoting
business and minimizing SME failure.
NOTES
1. Ang (1992, p.187) noted that ‘[s]mall businesses can terminate due to the departure or
demise of a single individual or the dissolution of a partnership’.
2. Storey et al. (1987, p.42) suggested that ‘manufacturers are more likely to be placed into
liquidation’ because they are more likely to have purchased fixed assets in order to operate
their business. Also, Garrod and Miklius (1990) reported that bankruptcies represented
11% of discontinuances for manufacturers but only 6% for retailers. Similarly, Stewart and
Gallagher (1986, p.46) noted that: ‘[s]ectors which have high capital costs are likely to have
higher levels of liquidations and bankruptcies. The firms will be more likely to be in debt
and will find it harder to cease trading and simply switch to something else.’
PART III
Comparing the performance of female- and

male-controlled SMEs
In this section I intend to dispel the myth that female-controlled SMEs

underperform male-controlled SMEs. As noted by Fischer, Reuber and
Dyke (1993, p.151), ‘With the rising number of women-owned businesses
has come a considerable amount of research, and even more specula-
tion, on differences between male and female entrepreneurs and their
businesses.’ Studies examining and comparing the performance of male-
and female-controlled businesses have generally reported that female-
controlled SMEs underperform male-controlled SMEs on a variety of
measures such as revenue, profit, growth and closure rates (Du Rietz and
Henrekson 2000); however, the results are by no means unanimous. ‘Thus
public policy-makers have had little guidance on such difficult issues as
whether or not unique training and support programs should be designed
for women versus men’ (Fischer et al. 1993, p.151).
Liberal feminist theory (Fischer et al. 1993) suggests that SMEs run by
women will exhibit poorer performance because women are overtly dis-
criminated against (by lenders, for example) and/or because of other sys-
tematic factors that deprive women of important resources (for example:
business education and experience). By way of contrast, social feminist
theory (Fischer et al. 1993) suggests that men and women are inherently
different by nature. These differences do not imply that women will be
less effective in business than men, but only that they might adopt differ-
ent approaches which might, or might not, be equally as effective as the
approaches adopted by men.
The majority of past research appears to have adopted a liberal
feminist theory perspective in the sense that researchers have attempted
to explain the apparent underperformance of female-owned businesses
44 SME performance
by referring to potential discrimination (for example, by bankers) and/

or by examining key demographic differences between male and female
entrepreneurs and their businesses (for example, age, size and industry
differences). The assumption in these studies is that if certain biases
against female entrepreneurs are removed and/or key demographic dif-
ferences are controlled, there should be no significant difference in the
relative performances of male- and female-owned businesses (Anna et
al. 2000).
For example, Anna et al. (p.279) suggested that one possible explana-
tion for any systematic differences in SME performance by gender might
be because ‘female business ownership is concentrated primarily in the
retail and service industries where businesses are relatively smaller in
terms of employment and revenue as opposed to high technology, con-
struction, and manufacturing’. Similarly, Rosa, Carter and Hamilton
(1996) argued that at least some of the gender difference in business per-
formance might be related to industry differences, because women tend
to start businesses in sectors that have low returns. Also, Hutchinson,
Hutchinson and Newcomer (1938) noted that enterprises requiring
little capital are likely to have the highest closure rates because when
large amounts of capital are at stake, the owners are likely to make a
more thorough investigation of the prospects for the new enterprise.
Given the comparatively lower hurdle (in terms of capital requirements)
for establishing ventures in the retail and services sector (Brush and
Chaganti 1999), these sectors could be expected to have higher closure
rates. Further, service businesses typically have a greater reliance on their
founder and may well cease when that person retires or decides to pursue
another activity. Therefore, if females are more likely to establish ventures
in the retail and services sector, and less likely to establish businesses in
the manufacturing sector, using closure of business as the definition of
failure is likely to be biased against female SME owners and, therefore,
controlling for industry is essential.
Besides industry differences, there are a number of other potential
systematic differences between male and female business owners that
might explain (at least partially) why female-controlled businesses appear
to underperform male-controlled businesses. First, is age of business.
Female-controlled businesses (on average) may be younger than male-
controlled businesses (Rosa et al. 1996), and Chapter 3 highlighted the fact
that businesses are most at risk in their early years. Second, is size of busi-
ness. Although Jovanovic (1982) argues that business failure should not
be associated with size of business (provided age of business is controlled)
the evidence provided in Chapter 3 suggests that for some failure defini-
tions there might still be a size effect after controlling for age. Therefore,
Comparing performance of female- and male-controlled SMEs 45
it would be prudent to control for size in any comparison of male- and

female-controlled SMEs. Third, is hours worked in the business. Due to
family commitments, female business owners (on average) might have less
time available for their businesses than male owners (Birley 1989; Fasci
and Valdez 1998). Fourth, is the level of human and financial capital.
Female SME owners might (on average) have less borrowing capacity
(or less access to capital) and might not have the same levels of education
and prior experience, compared to male owners (Brush and Hisrich 1991;
Cooper, Gimeno-Gascon and Woo 1994). Fifth, is risk-taking propensity.
Female owners (on average) may be more risk-averse than male owners
(Cooper 1993; Anna et al. 2000). Finally, there might be differences in
motivations by gender. For example, female owners may (on average) be
less concerned with financial rewards and more concerned with time flex-
ibility (particularly if they have family responsibilities) than male owners
(Brush 1992; Rosa et al. 1994).
Unfortunately, even after controlling for certain demographic differ-
ences, the majority of prior research has still found that female-owned
businesses underperform relative to male-owned businesses (Kalleberg
and Leicht 1991; Cooper et al. 1994; Rosa et al. 1996; Fasci and Valdez
1998; Du Rietz and Henrekson 2000).1 However, given the discussion in
Part II, it is conceivable that the performance measures used by previous
studies might have contributed to this finding. It is also possible, due to
data limitations, that many previous studies were unable to adequately
control for all important demographic differences, particularly age, size
and industry.
The following three chapters will compare the relative performances
of male- and female-controlled SMEs, incorporating as many of these
systematic differences as is possible with the available data. Chapter 4
will examine failure rates, where failure is defined as the closure of the
business; Chapter 5 will focus on return measures such as return on
equity (ROE) and return on assets (ROA); and Chapter 6 will explore a
risk-adjusted measure that might be used to compare male- and female-
controlled SMEs. I trust that by the end of these chapters, the reader will
be convinced that female-controlled SMEs do not under-perform male-
controlled SMEs, where appropriate performance and control measures
are used in the analysis.
NOTE
1. Du Rietz and Henrekson (2000) found that after controlling for industry the underper-
formance of female entrepreneurs disappeared for three variables (increased profitability;
46 SME performance
increased number of employees; and increased number of orders) but not for a fourth
(increased sales). Chell and Baines (1998), in a UK study of 104 micro-businesses, found
that after controlling for industry there was no significant difference in sales between
male- and female-owned businesses.
4. Failure rates
4.0 INTRODUCTION
In this chapter I aim to dispel the myth that female-owned SMEs are less
likely to survive (more likely to close) than male-owned SMEs. Carter,
Williams and Reynolds (1997) surveyed a sample of 203 retail firms from
two midwestern states of America in 1986 and then again in 1992. They
found that 34% of women-owned businesses but only 22% of men-owned
businesses ceased operations over the six-year period of their study. Boden
and Nucci (2000), in a large study of US sole proprietorships in the retail
and service industries that commenced operations in two different time
periods, found that the mean survival rate for male-owned businesses was
4–6% higher than for female-owned businesses. The findings of these two
studies suggest that female-owned retail and service businesses have higher
odds of closure than those owned by males.
However, Cooper et al. (1994) analysed a longitudinal study of 1053
new ventures (representative of all industry sectors and geographical
regions) in an attempt to predict the performance of new ventures based
on factors that could be observed at the time of start-up. Indicators of
initial human and financial capital were examined to determine how they
affected the probability of three possible performance outcomes: failure,
marginal survival or high growth. Cooper et al. (1994) argued that general
human capital (represented by the entrepreneur’s education, gender and
race) might reflect the extent to which the entrepreneur has had the oppor-
tunity to develop relevant skills and contacts. The results presented by
Cooper et al. (1994) suggest that while women-owned ventures are less
likely to grow, they are no more likely to close.
Kalleberg and Leicht (1991) tested several hypotheses concerning the way
the survival (and success) of small businesses headed by men and women was
related to industry differences, organizational structures and the attributes
of their owner-operators. Their analysis was based on businesses in south
central Indiana across three industries (food and drink, computer sales
and software, and health) for the period 1985–1987. As with Cooper et al.
(1994), Kalleberg and Leicht’s (1991) results also suggest that businesses
headed by women are no more likely to close than those headed by men.
47
48 SME performance
It would seem, therefore, that there are some conflicting results from
prior research into the relative performances of male- and female-owned
businesses. However, these past studies have generally been based on
limited samples (which might explain the conflicting results), making it
difficult to generalize from their findings. The closure rates presented and
discussed in this chapter are based on a large and representative data set
compiled by the Australian Bureau of Statistics (ABS) covering the four-
year period 1995–1998, as discussed in Chapter 2.
4.1 COMPARING CLOSURE RATES FOR MALE-

AND FEMALE-CONTROLLED SMES
Of the 5030 SMEs selected by the ABS for an annual survey every year
for four consecutive years, there were 3046 SMEs for which the sex of
the major decision maker could be determined (this person was deemed
to control the business even though they might not have been the major
owner). Of these businesses, 2868 were male-controlled and 178 were
female-controlled. Table 4.1 presents the annual closure rates for these
SMEs.
Table 4.1 Annual closure rates for male- and female-controlled SMEs
Details of Business Closures Male Female Total

Businesses active in 1994–95 2868 178 3046
Businesses closed in 1995–96 252 18 270
% of businesses closed 9 10 9
% of businesses closed 7 13 7*
% of businesses closed 7 6 7
Average annual closure rate 8 10 8*
Note: * Female closure rate significantly higher than male closure rate at 5% using a one-
tailed test.
Source: Adapted from Watson (2003, Table 4).

Failure rates 49
From Table 4.1 it can be seen that there is some evidence to suggest that
the closure rate for female-controlled SMEs is higher than that for male-
controlled SMEs. At face value, this finding is consistent with Carter et
al. (1997) and Boden and Nucci (2000). However, the results presented in
Table 4.1 do not control for other potentially important variables such as
age, size and industry.
4.2 AGE, SIZE AND INDUSTRY DIFFERENCES
It is possible that the higher closure rate for female-controlled SMEs is the
result of key demographic differences: in particular age, size and industry.
For the female- and male-controlled SMEs in the ABS sample, Table 4.2
presents their age and size details and Table 4.3 reports the closure rates by
industry sector, together with the percentage of female-controlled SMEs
in each of those sectors.
Table 4.2 indicates, as expected, significant differences between the
male- and female-controlled SMEs in terms of both age and size of
business. In particular, the female-controlled SMEs are significantly
overrepresented in the two youngest age categories and significantly
underrepresented in the oldest age category. As discussed in Chapter 3,
SME failure rates appear to peak during the first five years and, therefore,
Table 4.2 Age and size demographics for the female- and male-controlled
SMEs
Male Female
Controlled Controlled
n = 2868 n = 178
Firm age (%)
Less than 2 years 13 22**
2 years to less than 5 years 15 22**
5 years to less than 10 years 23 21
10 years to less than 20 years 27 24
20 years and older 21 11**
Total 100 100
Firm size in 1994–95
Number of employees 27 13**
Note: ** Significantly different to male-controlled SMEs at 1% using a two-tailed test.
Source: ABS.
50 SME performance
Table 4.3 Female-controlled SMEs by industry with industry closure

rates (%)
Industry % Female Closure

Controlled Rate
Construction 1 22
Wholesale trade 3 17
Mining 4 38
Manufacturing 5 21
Finance and insurance 7 27
Property and business services 8 25
Retail trade 9 28
Transport and storage 10 21
Cultural and recreational services 13 28
Accommodation, cafes and restaurants 17 31
Personal and other services 33 29
Average 7 23
Source: Adapted from Watson (2003, Tables 6 and 7).
if female-controlled SMEs are overrepresented in this age group it is little

wonder that they appear to have higher failure rates than male-controlled
SMEs. Similarly, the female-controlled SMEs were significantly smaller
than the male-controlled SMEs. Without controlling for these signifi-
cant age and size differences it is not possible to draw informed conclu-
sions concerning the relative failure rates of male- and female-controlled
SMEs.
In Table 4.3 the industries have been arranged in ascending order based
on the representation of female-controlled SMEs in each industry sector.
As can be seen from the table, female-controlled SMEs are underrepre-
sented in construction, wholesale trade, mining and manufacturing and
are overrepresented in most of the retail and service sectors. Table 4.3
also indicates that, with only two exceptions, the industry sectors in which
the female-controlled SMEs are overrepresented (underrepresented) have
higher (lower) closure rates. The first exception is transport and storage,
where women tend to be relatively overrepresented (10% compared to an
average of 7%) and where the closure rate is below average (21% com-
pared to 23%). The other exception is mining, where women tend to be
relatively underrepresented (4% compared to 7%) and where the closure
rate is above average (38% compared to 23%). The high closure rate in
the mining sector highlights the problem with using closure as a defini-
tion of failure. Ultimately all mines close as their resource deposits are
Failure rates 51
exhausted – does this mean that all mining ventures are failures? Clearly
not! Further, many SMEs that were initiated to provide services to a
particular mine site will also almost certainly close when the mine closes
– are all of these ventures also failures? Again, the answer is clearly not!
As noted in Chapter 2, there is no generally accepted definition of failure
that meets the needs of all users (and for which data is readily available)
and it is important, therefore, that researchers understand (and explain)
the potential limitations with respect to the particular definition they have
adopted.
Given the industry differences noted above, it is important that indus-
try, together with age and size of business, is controlled in any comparison
of failure rates for male- and female-controlled SMEs. In the following
section I will control for age, size and industry in further examining the
closure rates reported in Table 4.1 for the 2868 male-controlled and 178
female-controlled SMEs from the ABS sample.
4.3 CONTROLLING FOR AGE, SIZE AND

INDUSTRY
To examine the association between SME closure and the sex of the major
decision maker, a logistic regression analysis was undertaken incorporat-
ing age, size and industry as control variables. The results of the analysis
are reported in Table 4.4. As expected, they indicate that the effect of both
age and industry on the probability of SME closure (failure) is highly sig-
nificant. However, consistent with the arguments advanced by Jovanovic
(1982), size of the business is unrelated to its survival prospects. After
controlling for age, size and industry, the sex of the person responsible for
the major decision making within the business is also not significant, that
is, there is no association between SME closure and the sex of the major
decision maker.
4.4 SUMMARY
In support of the suggestion by Anna et al. (2000), Rosa et al. (1996)

and Hutchinson et al. (1938), the results provided in Table 4.4 (for a
large highly representative longitudinal sample of Australian SMEs)
clearly demonstrate that female-controlled SMEs do not close (fail) at a
higher rate than male-controlled SMEs after allowing for age, size and
industry differences. Chapter 5 will now examine a number of alternative
performance measures to test the robustness of this finding.
52 SME performance
Table 4.4 Results of logistic regression examining SME closure against,

size, industry and sex of major decision maker
B S.E. Wald df Sig. Exp(B)

Sex of major decision maker −0.02 0.20 0.01 1 0.94 0.99
Firm age 392.91 4 0.00
Less than 2 years 2.54 0.16 237.96 1 0.00 12.61
2 years to less than 5 years 0.70 0.17 16.45 1 0.00 2.02
Firm Size 0.00 0.00 0.04 1 0.84 1.00
Industry 23.63 10 0.01
Mining 0.49 0.64 0.60 1 0.44 1.64
Manufacturing 0.01 0.35 0.00 1 0.97 1.01
Construction −0.18 0.39 0.21 1 0.65 0.84
Wholesale trade −0.36 0.37 0.97 1 0.33 0.70
Retail trade 0.10 0.37 0.07 1 0.79 1.10
Accommodation, cafes & 0.36 0.44 0.68 1 0.41 1.43
restaurants
Transport & storage −0.67 0.47 2.09 1 0.15 0.51
Finance & insurance 0.57 0.40 2.08 1 0.15 1.78
Property & business services −0.09 0.36 0.06 1 0.80 0.91
Cultural & recreational services −0.51 0.53 0.94 1 0.33 0.60
Constant −1.98 0.38 26.35 1 0.00 0.14
Note: In running the logistic regression the last category was used as the reference point
for each categorical variable and, therefore, the last category is not shown in the table. For
firm age the last category is ‘20 years or older’ and for industry the last category is ‘Personal
and other services’.
5. Relating outputs to inputs
5.0 INTRODUCTION
Chapter 4 analysed closure rates for male- and female-controlled SMEs

and found no difference after controlling for age, size and industry.
However, as discussed in Chapter 2, using closure rates as a performance
measure has a number of limitations. For example, many service busi-
nesses will close when the proprietor reaches retirement age, and the vast
majority of these would not be considered failures. Therefore, to further
compare the performances of male- and female-controlled SMEs, this
chapter will examine a number of other potential performance indicators
that relate various outputs (such as sales and profit) to various inputs
(such as the number of employees and the owner’s investment in the
business).
While previous studies have also examined some of these output and
input variables, few have related the output measures to the input meas-
ures. For example, Rosa, Carter and Hamilton (1996) reported that busi-
nesses owned by women were found to underperform on a number of
quantitative measures, such as: number of employees, sales turnover and
value of capital assets. I would argue that these are not appropriate meas-
ures of performance; they are simply measures of size.
For example, let us assume that we have two investors (A and B) and
that A invests $100 000 in the stock market and B invests $50 000. After
twelve months both investors liquidate their stocks with the result that
A achieved a profit of $4500 and B $3000. Which investor did best? A
had a higher profit, but then A also invested (and risked) significantly
more than B. In fact, A’s return was only 4.5%, while B’s return was 6%.
Alternatively, let us assume that we have two investors (A and B), each
with $50 000 to invest in the stock market. Further, let us assume that
A borrows an additional $50 000 that is also invested. At the end of the
investment period both A and B sell their investments, resulting in a profit
to A of $3500 (after paying $1000 interest on the borrowed funds) and a
profit to B of $3000. Did A outperform B? Certainly A earned more profit
than B ($3500 compared to $3000), A also earned a higher return than B
(7% compared to 6%); but then A also took on more risk than B.
53
54 SME performance
5.1 MEASURING PERFORMANCE: RELATING

OUTPUTS TO INPUTS
As noted by Palepu and Healy (2008, ch. 5, p.6) the ‘starting point for a
systematic analysis of a firm’s performance is its return on equity (ROE)’,
which is calculated as follows:
profit
ROE 5
owner’s investment
Further, ROE can be disaggregated to examine how profitably a company
employs its assets, that is, its return on assets (ROA) and its financial lever-
age, as shown below:
profit assets
ROE 5 3
assets owner’s investment
5 ROA 3 Leverage
Borrowing allows a firm to have an asset base that is bigger than could be
provided using just the owner’s investment (equity). This, in turn, allows
the firm to potentially earn higher profits, but it also increases the risk to
the firm. Therefore, when analysing a firm’s ROE, it is important to sepa-
rate out the effects of leverage to eliminate the potentially confounding
effects of differing debt policies and attitudes to risk. In the next chapter,
the issue of risk, when comparing the performance of male- and female-
controlled SMEs, will be explored further.
5.2 COMPARING MALE- AND FEMALE-

CONTROLLED SMES
Again based on the data set compiled by the Australian Bureau of

Statistics (ABS), as described in Chapter 2, Table 5.1 presents the mean
results for two key output measures (total income and profit) and two key
input measures (total assets and owner’s equity) for both the male- and
female-controlled SMEs in the sample. Note that the results presented in
this chapter are based on pooling the data available from each of the four
years that the ABS conducted its survey. This includes firms that were
not surveyed beyond the first year, together with all new firms added to
the sample in each of the subsequent three years. This resulted in a total
of 14 426 annual observations (13 551 for male-controlled SMEs and 875
for female-controlled SMEs). As can be seen from the results presented
in Table 5.1, the male-controlled SMEs have significantly higher total
Relating outputs to inputs 55
Table 5.1 Comparing the mean annual outputs and inputs for male- and
female-controlled SMEs
Output / Input Mean Sig.

Measures ($) Level
Male Female
Output measures
Total income 6 158 168 1 661 729 0.000
Profit 354 181 22 314 0.000
Input measures
Total assets 4 432 502 734 070 0.000
Owner’s equity 1 509 569 158 866 0.000
income and profits (output measures) than the female-controlled SMEs.

However, the male-controlled SMEs also have significantly higher total
assets and investment by the owners (input measures) than the female-
controlled SMEs. While these results indicate that, on average, male-
controlled SMEs are larger than female-controlled SMEs, they do not
indicate that the former outperform the latter. Before we can appropri-
ately compare the performance of male- and female-controlled SMEs, we
need to relate the output measures to the input measures, to see if the extra
amounts invested (risked) by the male SME owners pays off.
To determine whether the extra investment in the male-controlled
SMEs pays off, Table 5.2 presents the results of examining the relationship
between: total income and total assets (TITTA); profit and total assets
(ROA); and profit and the owner’s investment (ROE), for both the male-
and female-controlled SMEs. However, an examination of the histograms
for each of these performance measures (with a normal curve fitted) sug-
gests that they are not normally distributed. For this reason the log of each
of these performance measures is also examined. Given that there were
1164 firms (1071 male controlled and 93 female controlled) with reported
losses, for the ROE and ROA measures the sample firms were split into
two groups: profitable and unprofitable. Businesses with zero profit (122)
were excluded. For the unprofitable group the absolute amount of the loss
was examined (in relation to total assets and total equity).
For all firms, the results in Table 5.2 show no significant difference in the
relative performances of the male- and female-controlled SMEs as meas-
ured by TITTA, ROA, ROE or the log of TITTA. However, for profitable
businesses, the female-controlled SMEs significantly outperformed the
male-controlled SMEs on three of the four performance indicators: ROA,
56 SME performance
Table 5.2 Relating mean annual outputs to inputs in comparing male- and
female-controlled SMEs
Performance Measure Mean Sig.

Level
Male Female
All firms
Total income to total assets (TITTA) 10.5392 10.1048 0.960
Profit to total assets (ROA) 0.3744 0.5478 0.197
Profit to total equity (ROE) 0.7824 1.8938 0.334
Log of total income to total assets (Log 1.1583 1.2406 0.142
TITTA)
Profitable firms
Profit to total assets (ROA) 0.5398 0.8339 0.048
Log of profit to total assets (Log ROA) −1.6369 −1.3604 0.000
Profit to total equity (ROE) 1.0139 2.5461 0.321
Log of profit to total equity (Log ROE) −0.6930 −0.5015 0.005
Loss making firms
Absolute loss to total assets (ABROA) 0.5902 0.6373 0.920
Log of absolute loss to total assets (Log −2.0283 −1.9877 0.759
ABROA)
Absolute loss to total equity (ABROE) 1.7095 1.3829 0.674
Log of absolute loss to total equity (Log −0.9210 −1.0729 0.248
ABROE)
Source: Watson (2002, Table 3).
the log of ROA and the log of ROE. For unprofitable businesses there was
no significant difference between the male- and female-controlled SMEs.
Therefore, there is some evidence to suggest, contrary to popular belief,
that female-controlled SMEs might actually outperform male-controlled
SMEs in terms of ROE and ROA.
However, while the results presented in Table 5.2 suggest that female-
controlled SMEs might outperform male-controlled SMEs, the results do
not control for a number of potentially important confounding variables
that would be expected to impact SME performance. For example, Chapter
3 demonstrated how SME performance could vary significantly across
industries and by age and size of business. Also, Fasci and Valdez (1998)
found that hours dedicated to a business on a weekly basis, a measure of
the owner’s labour input to the business (as opposed to financial input),
contributed significantly to earnings, and this is therefore another variable
that should be controlled in any evaluation of firm performance. In the
Relating outputs to inputs 57
Table 5.3 ANOVA of log ROA and log ROE and various business
demographics
Source of Sum of DF Mean F Sig.

Variation Squares Square of F
Log of ROA
Industry 348.40 10 34.84 14.96 0.00
Firm age 390.96 4 97.74 41.97 0.00
Days business 28.94 1 28.94 12.43 0.00
operated
Sex of owner 10.97 1 10.97 4.71 0.03
Log of ROE
Industry 143.24 10 14.32 7.06 0.00
Firm age 494.23 4 123.56 60.90 0.00
Days business 0.59 1 0.59 0.29 0.59
operated
Sex of owner 1.67 1 1.67 0.82 0.36
Source: Adapted from Watson (2002, Tables 5 and 6).
following analysis of firm performance, the SMEs in the sample were clas-
sified into two groups – those operating less than five days per week and
those operating five or more days per week – and this was included as an
additional control variable (together with age and industry). Note that size
is not included as a control variable because the performance measures
themselves control for size by relating outputs to inputs.

CONTROLLED SMES: CONTROLLING FOR
INDUSTRY, AGE AND DAYS BUSINESS
OPERATED
Table 5.3 presents the results of analysing the variance (ANOVA) in Log
ROA and Log ROE, for profitable businesses only, to see if the significant
differences between the male- and female-controlled SMEs reported in
Table 5.2 remain after controlling for industry, age of business and the
number of days the business operated. As expected, industry and age
of business are highly significant in explaining the variation in perform-
ance across SMEs. Similarly, the number of days the business operated
was significant in explaining variations in Log ROA, but not Log ROE.
Presumably the reason why the number of days the business operated was
58 SME performance
not significant in explaining variations in Log ROE is that a significant

driver of ROE is leverage, and this is related to the amount of security
available and the willingness of the owner to borrow, rather than to the
number of days the business is operated.
If a 1% level of significance is adopted as the cut-off (given the large
sample size), the results presented in Table 5.3 suggest that, after control-
ling for various demographic variables, there is no difference in the per-
formance of male- and female-controlled SMEs. Interestingly, if we use
a 5% cut-off, the results suggest that the female-controlled SMEs outper-
formed the male-controlled SMEs in terms of ROA. As noted in section
5.1, leverage (debt) can increase a firm’s ROE. Given that, on average,
female-controlled SMEs have a lower debt to asset ratio compared to
male-controlled SMEs (Watson 2006), we would expect ROE to be higher
for male-controlled SMEs. Further, if we assume that male- and female-
controlled SMEs perform equally well, we would not expect there to be a
significant difference in the ROA measure for the two groups. Put another
way, if male- and female-controlled SMEs perform equally well, but the
male-controlled firms have relatively more debt, we would expect the two
groups to have a similar ROA, but the male-controlled firms should have
a higher ROE. However, this is not what we find. What we find is that the
female-controlled SMEs have a higher ROA measure (better asset usage)
and there is no difference between the two groups in terms of ROE. This
indicates that asset usage might be better managed in female-controlled
SMEs and that this compensates for the potentially higher leverage gains
achieved by male-controlled SMEs.
5.4 SUMMARY
Again, the findings presented in this chapter clearly indicate that female-
controlled SMEs do not underperform male-controlled SMEs. Indeed,
there is some evidence to suggest that, with respect to ROA (asset man-
agement), female-controlled SMEs might outperform male-controlled
SMEs.
6. Adjusting for risk
6.0 INTRODUCTION
In this chapter I continue to explore how the performance measures used

in most previous research are incomplete and probably biased against
female SME owners. The focus in this chapter is on risk. While risk is
important for all firms, it is particularly important in SMEs because there
is little separation between business and personal risk in an SME. For
example, the personal assets of SME owners are often used to secure bank
loans for their businesses. In large incorporated businesses, the liability
of owners (shareholders) is limited to the value of the shares they have
bought and does not extend to their private assets.
Forlani and Mullins (2000) note that although risk plays a central role
in most entrepreneurial decision making, there has been little empirical
research explicitly examining how the elements of risk, risk perceptions
and an entrepreneur’s propensity to take risks, influence choices among
potentially risky entrepreneurial ventures. Ballantine, Cleveland and
Koeller (1993, p.87) argue that ‘modestly higher average profit rates in
favor of large firms have been overemphasized while the much more sub-
stantial variations in profit experienced by all firms, and particularly small
firms, have been largely ignored’. Similarly, Fischer (1992) urges future
researchers to explore the possible differences in the risk-taking propensi-
ties of men and women. Therefore, the aim of this chapter is to examine
the performances of male- and female-controlled SMEs, adjusting for
differences in risk.
6.1 CONTROLLING FOR RISK IN MEASURING

SME PERFORMANCE
‘A common human failing is the desire for simple answers to difficult

questions’ (Sharpe 1975, p.29). Sharpe notes that this applies particularly
to performance measurement, as most people want answers provided in
the form of a single unambiguous number. Research into SME perform-
ance has tended to focus on sales and/or profit (or growth in sales and/or
59
60 SME performance
profit) as the single most important number, without any explicit control
for risk. However, we know ‘there is risk in the world, and that inves-
tors generally dislike it’ (Sharpe 1975, p.29). While there is no doubting
the importance of sales and profit to a business, it is equally important
to explicitly relate these performance indicators to the underlying risks
involved in the business.
It should be noted that, although the ‘economics literature typically
defines risk as variability’ (Forlani and Mullins 2000, p.309), it can mean
different things to different people. Forlani and Mullins, in an experi-
mental study, asked a sample of entrepreneurs leading America’s fastest
growing firms to make choices among a series of hypothetical new ven-
tures. They found that while their subjects ‘tended to shun high levels
of variability . . . they appeared willing to accept a considerable degree
of hazard, or possible downside . . . presumably in pursuit of potentially
significant gains’ (2000, p.305). This finding indicates that variability is
central to the entrepreneur’s notion of risk.
Sharpe (1975) suggests that the reward-to-variability ratio (Sharpe
ratio) is an appropriate unambiguous measure of performance that
controls for risk. ‘The reward-to-variability ratio is simply the ratio of
reward (which is good) to variability (which is bad)’ (Sharpe 1975, p.29).
Other things being equal, the higher the ratio, the better the performance.
Although Sharpe discussed the reward-to-variability ratio in the context
of comparing the performances of individual securities and portfolios, it
would appear to be an appropriate measure for assessing the perform-
ance of SMEs. Taggart (1996, p.276) notes that because the Sharpe ratio
‘adjusts for total risk, it can be useful for assessing the performance of
a portfolio that is less than fully diversified’. Given that the majority of
SME owners have a less than fully diversified investment portfolio (with
the majority of their wealth being tied up in their business) the Sharpe
ratio would seem to be particularly well suited to assessing and comparing
SME performance.
In applying the Sharpe ratio, it is normal to measure ‘reward’ in terms
of stock price returns and ‘variability’ as the standard deviation in those
returns.1 However, many SMEs are not listed, and stock price information
is therefore not available. In the absence of stock price returns it seems rea-
sonable to suggest that profit might be an appropriate alternative measure
of reward and the standard deviation in profit an appropriate alternative
measure of variability (risk). In support of using profit (rather than stock
market returns) as the reward measure, the following two points should be
noted: first, for SME owners, the profit earned by their ventures is clearly
a significant reward; and second, stock prices are largely driven by profits
(particularly future expected profits).2
Adjusting for risk 61
6.2 COMPARING MALE AND FEMALE ATTITUDES

TO RISK
Barber and Odean (2001) found that for different risk measures (portfo-
lio volatility, individual stock volatility, beta and size) men consistently
invested in riskier positions than women. Similarly, Powell and Ansic
(1997) noted that a lower preference for risk amongst females is one
of the gender differences that is persistently found in both the general
and business-specific literature. For example, Jianakoplos and Bernasek
(1998) examined a US sample of household holdings of risky assets and
found that single women exhibited relatively more risk aversion in finan-
cial decision making than single men. Indeed, ‘[r]oughly 60% of the female
respondents said they were not willing to accept any risk, while only 40%
of the men said they were unwilling to take risks’ (p.620).
In a study comparing 105 female business owners (ranked in the top 10%
with respect to sales and number of employees) with similar male business
owners, Sexton and Bowman-Upton (1990) found that the females scored
significantly lower on traits related to risk taking. The scores indicate
that female entrepreneurs are ‘less willing than their male counterparts to
become involved in situations with uncertain outcomes’ (p.29). Similarly,
for a sample of male and female SME owners, Watson and Newby (2005)
reported that the males had a significantly higher risk-taking propensity
score. Powell and Ansic (1997) noted that females tend to focus on strate-
gies that avoid the worst situation to gain security. Finally, based on 229
interviews with small-business owners in the Greater Vancouver area, Cliff
(1998, p.523) suggested that female entrepreneurs ‘seem to be more con-
cerned than male entrepreneurs about the risks of fast-paced growth and
tend to deliberately adopt a slow and steady rate of expansion’.
There would appear, therefore, to be considerable evidence to suggest
that females might be more risk averse than males. If this is so, and if
this higher risk aversion affects the business strategies adopted by female
SME owners, it might help explain the apparently lower profits earned by
female-controlled businesses. Traditional investment theory suggests that
the higher the risk, the higher the expected return, and conversely, the
lower the risk, the lower the expected return (Anna et al. 2000). Therefore,
it is reasonable to expect that if we relate risk to returns (using the Sharpe
ratio) there might not be any significant difference in the performances for
male- and female-controlled SMEs. In other words, while females might
take a different approach to business (for example, they might be more
cautious in terms of the resources they commit to their ventures and in
growing their businesses) they are likely to be just as effective as males,
provided performance is measured in risk-adjusted terms.
62 SME performance

CONTROLLED SMES USING THE SHARPE
RATIO
As with the previous chapter, the data referred to in this chapter comes
from the four-year longitudinal study conducted by the Australian Bureau
of Statistics (ABS), as described in Chapter 2. Unlike the previous chapter,
however, this chapter only considers those SMEs that were in operation
for the entire four-year period covered by the ABS data. Any firm that
ceased operations during the ABS data collection period (or any new firm
added) was eliminated from the analysis because it was important to have
a full four years of data available to adequately compute a firm’s Sharpe
ratio (Sharpe 1975).
For the reduced sample of 2236 male-controlled and 131 female-
controlled SMEs, Table 6.1 reports three performance indicators: average
annual profit over the four-year period; the standard deviation in those
annual profits; and the Sharpe ratio. Because an examination of the histo-
grams for each of these measures (with a normal curve fitted) suggests that
they are not normally distributed, the log of each of these performance
measures is also reported.
As expected from the results presented in Chapter 5, the average
annual profit (and the log of average annual profit) was significantly
lower for the female-controlled SMEs. However, as expected from the
discussion in section 6.2, the standard deviation in annual profits (and
the log of the standard deviation of annual profits) was also significantly
lower for the female-controlled SMEs. When average annual profits
Table 6.1 Comparing male- and female-controlled SMEs
Performance Measure Male Female

N=2236 N=131
Average annual profit (’000) 448 63**
Log of average annual profit 4.4 3.3**
Standard deviation of annual profit (’000) 356 83**
Log of standard deviation of annual profit 4.2 3.0**
Sharpe ratio 1.7 1.6
Log of Sharpe ratio 0.2 0.2
Note: ** Female-controlled SMEs significantly lower than male-controlled SMEs at 1%

using a one-tailed test.
Source: Watson and Robinson (2003, Table 3).

are related to the standard deviation in those profits, using the Sharpe
ratio (Sharpe 1975), no difference exists between the male- and female-
controlled SMEs. This suggests that, after adjusting for risk, there is no
difference in the performance of male- and female-controlled SMEs. Put
another way, males take more risks and, as a result, on average their
firms earn higher profits.

CONTROLLED SMES USING THE SHARPE
RATIO: CONTROLLING FOR AGE, INDUSTRY
AND SIZE OF BUSINESS
The results presented in Table 6.1 do not control for industry effects,
the age of the business or the size of the business. Chapter 5 indicated
that each of these variables was potentially highly significant in explain-
ing variations in SME performance by gender. Therefore, to check
the robustness of the results presented in Table 6.1, Table 6.2 presents
the results of examining the log of average annual profit the log of the
Table 6.2 ANOVA of performance measures controlling for age, industry

and size
Source of Variation Sum of DF Mean F Sig

Squares Square of F
Log of average annual profit
Firm age 25.68 4 6.42 2.96 0.02
Industry 144.88 10 14.49 6.69 0.00
No. of employees 1963.80 3 654.60 302.18 0.00
Sex of owner 24.22 1 24.22 11.18 0.00
Log of st. dev. of annual profits
Firm age 8.94 4 2.24 1.39 0.24
Industry 197.46 10 19.75 12.26 0.00
No. of employees 2327.81 3 775.94 481.64 0.00
Sex of owner 21.70 1 21.70 13.47 0.00
Log of Sharpe ratio
Firm age 4.91 4 1.23 0.99 0.41
Industry 23.04 10 2.30 1.87 0.05
No. of employees 6.69 3 2.23 1.81 0.14
Sex of owner 0.36 1 0.36 0.29 0.59
Source: Watson and Robinson (2003, Table 4).

64 SME performance
standard deviation of annual profits and the log of the Sharpe ratio
using analysis of variance (ANOVA) to see if the sex of the person in
control of the business has any significant impact after controlling for
industry, age and size of business. The industry categories used are as
described in Chapter 3. The age categories used are: zero to less than
two years old; two to less than five years old; five to less than ten years
old; ten to less than 20 years old; and 20 or more years old. The number
of employees in the business at the time the business was first surveyed
is used to measure firm size. The SMEs were classified into the following
size groups: those with one to four employees; those with five to nine
employees; those with ten to 19 employees; and those with 20 or more
employees.3
As can be seen from Table 6.2, after controlling for industry, age and
size of business, the sex of the person in control of the business still has
an impact on the profits earned, that is, male-controlled firms typically
have higher profits than female-controlled firms. However, the table also
indicates that, after controlling for industry, age and size of business,
the sex of the person in control of the business also has an impact on
the standard deviation in annual profit, that is, male-controlled firms are
typically more risky than female-controlled firms. However, if the Sharpe
ratio is used as the basis of comparison, then the sex of the person in
control of the business has no impact. In this case industry is the only
variable associated with firm performance, but only at a 5% level of
significance.
The results presented in Table 6.2 confirm the robustness of the find-
ings reported in Table 6.1 and again support the proposition that there
is no significant difference in the performances of male- and female-
controlled SMEs, provided performance is measured appropriately. In
other words, although females may adopt different strategies in found-
ing, running and growing their businesses, they are likely to be no less
effective than males in terms of the risk-adjusted rewards earned by their
firms.
6.5 SUMMARY
Once again, the findings presented in this chapter clearly indicate that
female-controlled SMEs do not underperform male-controlled SMEs. I
trust that the evidence provided in Chapters 4, 5 and 6 will have convinced
the reader, once and for all, that provided performance is measured appro-
priately, and adequate controls are included in the analysis, male- and
female-controlled SMEs perform equally well.
NOTES
1. Stock price returns are usually calculated as: (the gain in stock price over a period +
dividends during the period)/beginning of period stock price.
2. It should also be noted that the Sharpe ratio (1975) is often adjusted by the risk-free
rate to allow comparison across periods where the underlying risk-free rate differs. In
this chapter, however, no such adjustment is made because the analysis covers the same
period for all firms.
3. Although these categories are somewhat arbitrary, they are similar to those used in previ-
ous studies. Also, the definition of a medium-sized business in Australia is one with 20 or
more employees and this is captured in the last category. This variable was also entered
into the ANOVA as a covariate (continuous) variable to see if it improved the analysis,
but it did not.
PART IV
Growth financing for SMEs
Part IV is aimed at dispelling the myth that SME growth, particularly for
female-controlled SMEs, is constrained by an inability to access appropri-
ate levels of external (bank) funding. The available literature suggests a
strong link between the availability of finance and SME growth, and this
has led to the notion of a ‘finance gap’, implying that ‘there may be major
“barriers” preventing an owner-manager’s access to equity’ (Hutchinson
1995, p.231). This notion of a ‘finance gap’ within the SME sector has been
supported by a number of researchers (see, for example, Berger and Udell
1998; Pissarides 1999; Becchetti and Trovato 2002; Carter et al. 2003). It
has also been suggested that the ‘barriers’ to finance might be even more
acute for female-owned SMEs, as there is a perception that financial insti-
tutions (banks) discriminate against female business owners (Riding and
Swift 1990; Breen, Calvert and Oliver 1995; Brush et al. 2001).
Given that SMEs are responsible for significant levels of employ-
ment, innovation and productivity, it is important that policy makers
and advisers are well-informed about the determinants of SME growth,
including the various demand-side issues surrounding the provision of
growth funding for this sector (Becchetti and Trovato 2002). Carpenter
and Petersen (2002) examined more than 1600 US small manufacturing
firms and found that the growth of these firms appeared to be constrained
by a lack of (internal) finance. Similarly, Bruno and Tyebjee (1985) found
that ventures that had received external capital achieved statistically sig-
nificantly higher sales and employment growth (compared to ventures
without external capital). With respect to women-owned businesses,
Carter and Allen (1997) noted that the availability of financial resources is
a major influence on their growth.
While there is no doubting that firms need finance to grow, it is also
the case that not all firms have the capacity, or desire, to grow. Simply
68 SME performance
looking at associations between growth/no growth SMEs and their levels

of external funding is likely to confuse cause and effect. That is, a study
of firm growth and external funding might well show a strong positive
relationship between these two variables. However, it is not possible to
conclude from such a study that firms without significant levels of exter-
nal funding have both the capacity and desire to grow and that it is only
a lack of funding that is holding them back. Given that the majority of
prior research concerned with the financing of SMEs has concentrated on
supply-side issues, Fried and Hisrich (1994) suggested that future research
should focus on demand-side issues where the available evidence and,
therefore, our level of understanding is far more limited.
In the following two chapters I will attempt to provide the reader with a
better understanding of the possible relationship between firm growth and
the availability of external funding, for both male- and female-controlled
SMEs. Based on focus group and survey results, Chapter 7 examines
(qualitatively) various demand-side issues surrounding the provision of
external funding to both male- and female-controlled SMEs. Chapter 8
then discusses a quantitative analysis of the longitudinal data provided by
the Australian Bureau of Statistics (ABS).
7. A qualitative analysis
7.0 INTRODUCTION
Winborg and Landstrom (2001) argue that financial problems (lack of

funds) constrain the development and growth of SMEs because many
are unable to access the same kinds of growth funding normally available
to large businesses. This notion of a ‘finance gap’ within the SME sector
has been supported by a number of researchers, for example: Berger and
Udell (1998); Pissarides (1999); Becchetti and Trovato (2002); and Carter
et al. (2003). It has also been suggested that the ‘finance gap’ might be
even more acute for female-owned SMEs (Riding and Swift 1990; Breen
et al. 1995; Brush et al. 2001). However, Riding and Swift (1990, p.327),
note that while ‘[p]revious work reveals a pervasive perception that there
is discrimination by bankers against women business owners’, this belief
appears to be based on subjective perceptions without any ‘real statistical
evidence’.
Further, these ‘barriers’ to SME growth are generally believed to result
from deficiencies in capital markets and include instances where SME
owners’ loan applications are rejected together with instances where
owners are ‘discouraged’ from applying for funding from a bank because
they believe their application will be rejected (Kon and Storey 2003). For
example, Levenson and Willard (2000) found that about 2% of US firms
did not obtain the funding for which they had applied and approximately
4% of firms were ‘discouraged’ from applying for funding because they
expected their request to be turned down. It seems, therefore, that twice as
many US firms are ‘discouraged’ from applying for funding as are denied
funding. However, it is also possible that many SME owners might con-
sciously decide they do not want to access funding from a bank given the
risks involved and/or the potential for them to lose control of their firms
(Barton and Matthews 1989; Cressy 1995; Hamilton and Fox 1998).
Based on a large sample of New Zealand SMEs, Hamilton and Fox
(1998) concluded that debt levels in small firms reflected demand-side
decisions and were not just the result of supply-side deficiencies. They
argued that managerial beliefs and desires played an important role in
determining the capital structure of SMEs and that a deeper appreciation
69
70 SME performance
of these issues would lead to a better understanding of the capital struc-

ture policies of individual SMEs. Chaganti, DeCarolis and Deeds (1996)
also found that the major determinant of SME capital structure was
owner goals, as these assisted in predicting debt versus equity and internal
versus external funding. These findings might be particularly relevant for
female-controlled SMEs because, as Buttner and Moore (1997, p.34) note,
female entrepreneurs measure success in terms of ‘self-fulfillment and goal
achievement’ and while profits and business growth are also important
they are considered ‘less substantial measures of . . . success’.
Hutchinson (1995) suggested that when owner-managers are risk averse
and have a desire to retain control of their firms, they may actively place
limits on the use of external sources of funding. This could result in some
SME owner-managers deliberately choosing low or no growth options
and might be particularly relevant for female-owned SMEs because, as
noted by Cliff (1998, p.523):
[F]emale entrepreneurs are more likely to establish maximum business size

thresholds beyond which they would prefer not to expand, and . . . these thresh-
olds are smaller than those set by their male counterparts. Female entrepre-
neurs also seem to be more concerned than male entrepreneurs about the risks
of fast-paced growth and tend to deliberately adopt a slow and steady rate of
expansion.
Similarly, for a sample of Swedish SMEs, Berggren, Olofsson and Silver

(2000) sought to determine the relative impact on a firm’s decision to apply
for a bank loan of the following five factors: the size of the firm; its degree
of technological development; the perceived need to grow to survive; the
amount of internally generated funds; and the owner-operator’s aversion
to losing control. They found that the strength of the owner-operators’
desire to maintain control of their firms was the principal determinant in
their decision to (not to) apply for bank finance.
Given the view that multiple research methodologies might allow a
more complete understanding of the processes that determine SME per-
formance (Cooper 1993), this chapter reviews two studies designed to help
better understand the various demand-side issues relating to the financ-
ing of SMEs (Watson, Newby and Mahuka 2006, 2009). The first study
involved a series of focus groups; the second surveyed SME owners.
7.1 FOCUS GROUP STUDY
A market research company was used to recruit a sample of 30 Western

Australian SME owner-operators who had considered (within the prior
A qualitative analysis 71
two to five years) a major expansion of their business that required

significant external funding. The plan was to run three separate focus
groups with ten participants in each, comprising ‘successful’ borrowers,
‘unsuccessful’ borrowers and ‘discouraged’ borrowers.
However, the market research company had great difficulty recruiting
participants for the second focus group, and this group therefore ended
up with only two ‘unsuccessful’ participants (both males) and seven
‘successful’ participants. Given the often-argued existence of a ‘finance
gap’, this was somewhat of a surprise, but confirmed the results reported
by Levenson and Willard (2000) indicating that the number of credit-
constrained firms might be quite small. This suggests that the so-called
‘finance gap’ might be more ‘myth’ than ‘reality’. Given that the second
group comprised mainly ‘successful’ participants, the responses from
this group have been combined with the responses from the ‘successful’
group.
The first issue the focus group participants were asked to consider
related to major obstacles SME owners might potentially encounter if
they wanted to substantially grow their business. Table 7.1 summarizes
the participants’ responses based on whether they were ‘successful’ or
‘discouraged’ SME owners. The results indicate that the majority (53%)
of obstacles identified related to operational issues (such as the lack of
appropriate staff, facilities, time and expertise). Next were concerns about
funding (24%), government regulation (15%), fear of failure (7%) and
other issues (1%).
Compared to the ‘successful’ borrowers, the ‘discouraged’ borrowers
were much more likely to list government regulation as a major obstacle
and much less likely to acknowledge potential operational problems.
There were no significant differences between the two groups in terms of
the frequency with which the remaining concerns were listed, including
financial issues.
Table 7.2 summarizes the responses from the focus group participants
when asked to list the reasons why SME owners might not seek external
funding. As expected, the most common reason given was that there were
sufficient internal funds available. Other major reasons included: the risk
involved; not wanting the burden of having to service additional debt;
and the terms of the funding being unacceptable. The contrast between
the ‘discouraged’ group and the ‘successful’ group is again of interest. The
‘discouraged’ group was much more concerned about the work and/or
hassles involved with expansion than were the members of the ‘successful’
group. This suggests that the members of the ‘discouraged’ group might
have been discouraged for primarily internal reasons rather than because
they thought their loan application would be turned down.
72 SME performance
Table 7.1 Potential obstacles to growth
Obstacles Faced Successful Discouraged Total

N=18 N=8 N=26
No. (%) No. (%) No. (%)
Operational issues – lack of 60 (59) 18 (39) 78 (53)**
appropriate staff, facilities, time,
supplies and expertise/knowledge
in assessing strategy, market size,
competition and how to go about
growing the business
Funding issues – lack of 24 (24) 11 (24) 35 (24)
internal cash flow or external funds
at reasonable rates and terms
Government regulation 7 (7) 15 (33) 22 (15)**
intrusion, lack of support, fees and
taxes
Fear of failure/unsure of 9 (9) 1 (2) 10 (7)
benefits
Other 1 (1) 1 (2) 2 (1)
Total 101 (100) 46 (100) 147 (100)
Note: ** ‘Discouraged’ group significantly different from ‘successful’ group at 1% using

a two-tailed test.
Source: Watson, Newby and Mahuka (2006, Table 9.6).
The groups were then asked to specifically consider the disadvantages

of debt funding and their responses are presented in Table 7.3. Consistent
with the views expressed by a number of authors, as noted in the intro-
duction, both groups agreed that the major issue centred on the risks
involved. These included the obvious concern about the potential for inter-
est rate rises, but many participants also raised a (perhaps less obvious)
concern that having easy access to funds (particularly credit cards) might
cause the business owner to spend unnecessarily (for example, on new
equipment). For many of the participants, the additional burden and
stress involved with repaying the debt was also of concern. Comparing
the ‘discouraged’ and ‘successful’ groups indicates that the potential for
loss of control seemed to be of much greater concern to the ‘discouraged’
group. This finding is consistent with Cressy’s (1995) argument that where
Table 7.2 Reasons SME owners might choose not to seek external
funding
Reason Successful Discouraged Total

N=18 N=8 N=26
No. (%) No. (%) No. (%)
Adequate internal funds 18 (28) 4 (15) 22 (24)
Risk/uncertainty about future, 11 (17) 5 (19) 16 (17)
potential to lose control, fear
of failure
Burden (don’t want) – can’t 9 (14) 5 (19) 14 (15)
service more debt
Terms unacceptable – interest 10 (15) 3 (11) 13 (14)
rates, security, etc.
Expand (don’t want to) – can’t 1 (2) 7 (26) 8 (9)**
cope, too many hassles, lack
confidence
Time and aggravation of trying 6 (9) 2 (7) 8 (9)
to get a bank loan
Unavailable because of poor/no 4 (6) 1 (4) 5 (5)
credit rating
Other 6 (9) 0 0 6 (7)
Total 65 (100) 27 (100) 92 (100)
Notes:
Figures do not always add up to 100% because of rounding.
** ‘Discouraged’ group significantly different from ‘successful’ group at 1% using a two-
tailed test.
owners’ desire for control is strong enough their businesses will be entirely
self-funded. Again, this indicates the importance of internal factors when
SME owners are considering growth options that require significant levels
of external funding.
Table 7.4 sets out the participants’ responses when asked to list the
reasons why a bank might refuse a loan application. The majority of
reasons provided by the participants can be grouped under two broad
headings: perceived inadequacies in the owner’s business acumen (busi-
ness plan and track record); and the perceived risks (including lack of
74 SME performance
Table 7.3 Main disadvantages of debt funding

N=18 N=8 N=26
No. (%) No. (%) No. (%)
Risk – interest rate changes, can’t 26 (47) 12 (43) 38 (46)
repay debt, spend too much
Burden – worry, work, stress, 12 (22) 4 (14) 16 (19)
paperwork
Dislike of financial institutions – 4 (7) 4 (14) 8 (10)
terms, fees, disclosures
Control may be lost – dependence 2 (4) 4 (14) 6 (7)**
on others is increased
Costs 5 (9) 0 (0) 5 (6)
Other 6 (11) 4 (14) 10 (12)
Total 55 (100) 28 (100) 83 (100)
Notes:
** ‘Discouraged’ group significantly different from ‘successful’ group at 1% using a two-
tailed test.
equity/security). This suggests that business owners need to ensure they

have a credible business plan that is easily understood so that bank loan
officers can properly assess the risks involved. In this regard there might
be an important role for external advisers, particularly for inexperienced
owners and/or owners without the necessary skills to complete the task.
For this question there were no significant differences between the ‘dis-
couraged’ and ‘successful’ groups. There was also nothing in the responses
to suggest that the banks routinely made ‘screening errors’ (Kon and
Storey 2003, p.37) and, therefore, the focus group results suggest that
the ‘discouraged’ borrower syndrome, as described by Kon and Storey,
might be relatively trivial in the Australian context. This is not to say that
owners are not discouraged from borrowing funds for growth, but rather
the causes of their discouragement might have more to do with the owners
themselves (for example, their desire to maintain control and their risk
aversion) rather than with deficiencies in the banking sector or the exist-
ence of a ‘finance gap’.
Table 7.4 Reasons for bank refusing a loan application

N=18 N=8 N=26
No. (%) No. (%) No. (%)
Business plan – don’t have 23 (29) 12 (31) 35 (29)
one or not convincing or
bank doesn’t understand
Track record – poor credit 19 (24) 9 (23) 28 (24)
rating or don’t have one
Risk – too high 11 (14) 8 (21) 19 (16)
Lack of equity/security 15 (19) 4 (10) 19 (16)
Other 12 (15) 6 (15) 18 (15)
Total 80 (100) 39 (100) 119 (100)
Note: Figures do not always add up to 100% because of rounding.
Table 7.5 confirms just how important maintaining control was for the
majority of the focus group participants. In the ‘discouraged’ group, all
participants rated the importance of maintaining control as a ‘1’ (very
important). In the ‘successful’ group there was a little more dispersion in
the ratings, but the majority of the group still attached a very high level of
importance to maintaining control. The one exception was a female partic-
ipant who indicated that, for her, maintaining control was unimportant.
Table 7.6 reports the views of the group with respect to the risk-taking
propensities of men and women. The responses for the male and female
focus group participants are shown separately. Interestingly, over half
the participants (mainly the males) did not express a view on this issue.
The consensus of those who spoke was that women were likely to be more
conservative (risk averse) than men, although a significant number of par-
ticipants believed either that there was no difference, or that it depended
on the personality of the individual owner, rather than the sex.1
Finally, Table 7.7 sets out the key thoughts/ideas expressed by the
participants concerning the factors (issues) that might influence an SME
owner’s decision to seek external funding. Two main thoughts emerged
from this discussion. First, that growth for growth’s sake, without a growth
in profit, was not worthwhile. Second, owners who were planning to exit
76 SME performance
Table 7.5 The importance of maintaining control
Rating Successful Discouraged Total

N=18 N=8 N=26
No. (%) No. (%) No. (%)
1 14 (78) 8 (100) 22 (85)
2 2 (11) – – 2 (8)
3 1 (6) – – 1 (4)
4 – – – – – –
5 – – – – – –
6 – – – – – –
7 1 (6) – – 1 (4)
Mean 1.63 1.00 1.38
Std Dev 1.54 0.00 1.24
Notes:
(1 = very important, 7 = unimportant)
Table 7.6 Comparing male and female SME owners’ attitudes to risk
Attitude to Risk Males Females Total

Women more conservative 4 3 7
Men more conservative 0 1 1
No difference 0 2 2
Depends on personality 2 1 3
Total 6 7 13
Note: Where a participant spoke on more than one occasion, his/her view was only
recorded once.
the firm were unlikely to want to raise additional funding for fear of over-
capitalizing their business. In terms of funding sources, it is interesting
to note that the few negative comments made about banks (for example:
banks will only lend you money when you don’t need/want it; and banks
won’t lend to businesses they don’t understand) came from the ‘discour-
aged’ group. It is difficult to know on what basis they formed their views
(perhaps it was from their previous experiences or, alternatively, from
Table 7.7 Factors likely to influence an SME owner’s decision to access

external funding
Key Thoughts (Ideas) No. of times

mentioned
Growth:
Without profit it isn’t worth it 4
Can’t just decide to grow – there are competitors 2
Business conditions:
Less likely to borrow if business is volatile and control could be 2
lost
If firm is failing you might borrow to try and save the business 2
Exiting the firm:
Additional funding isn’t needed if you are planning to exit the 4
firm
Unless you want to build the business up ready for sale 1
Or have someone (whom you trust) willing to buy in 1
Although finding that trustworthy person isn’t easy 2
I would prefer to sell the business to my staff 1
I would be prepared to offer key personnel a stake in the 1
business
I am planning to hand the business over to my kids 1
Accessing debt funding:
As you get older it is harder to get a loan (repayments) 1
As you get older it is easier to get a loan (security) 1
As you get older you may not want to borrow 1
Age makes no difference 2
Being married helps when you borrow 1
But for women the husband must often co-sign 1
Banks want security 2
Banks don’t lend to businesses they don’t understand 2
Banks will lend you money when you don’t need/want it 1
Bigger firms find it easier to borrow 1
Bigger loans attract lower interest rates 1
More likely to borrow to modernize equipment 1
But over-capitalizing isn’t helpful 1
Total 37

78 SME performance
hearing about the experiences of others), but clearly the participants in the
‘successful’ group did not share their views. However, the comment that
banks don’t lend to businesses they don’t understand raises two important
issues. First, it suggests that banks should not discount the importance of
relationship (compared to transactions-based) lending for the SME sector.
‘Relationship lending is generally associated with the collection of “soft”
information over time through relationships with the firm, the owner, and
the local community’ (Berger and Udell 2002, p.F38). By way of contrast,
transactions-based lending is ‘generally associated with the use of “hard”
information’ (Berger and Udell 2002, p.F38), such as financial ratios, and
may not be appropriate for many SMEs, particularly for non-traditional
businesses. Second, it is important that SME owners seeking external
funding ensure that they have a clearly articulated business plan that
makes it as easy as possible for a loan officer to understand the nature of
their business and the risks involved. SME owners with limited expertise in
this area should consider obtaining professional help. This is an area that
government policy makers might consider investigating further.
In summary, a number of interesting findings emerged from the focus
group sessions. First, it seems that the majority of SME owners are acutely
aware of the various risks associated with business ownership, and this is
therefore foremost in their minds when they consider the merits of seeking
external funding. Related to this notion of risk is the issue of control, the
second major theme that seemed to be at the ‘top of the mind’ of virtu-
ally all the focus group participants. It seems that many SME owners are
unlikely to consider external funding if there is a reasonable likelihood
that they could lose control of their business. The focus group results
also suggest that (compared to male owners) female owners are more
risk averse, and therefore less inclined to access external sources of funds.
Based on the results of these focus groups, it appears that the so-called
‘finance gap’ (believed by many to be a major barrier inhibiting SME
growth) is more ‘myth’ than ‘reality’. This issue is pursued further in the
next section, which looks at the findings of a survey designed to further
explore the issues raised by the focus group participants.
7.2 SURVEY OF SME OWNERS
This second study was designed to further investigate the various demand-
side issues that arose from the focus group discussions. Based on the focus
group results and a review of the literature, a mail survey was sent to 534
SME owners. Excluding the 69 ‘dead letters’ that were returned ‘address
unknown’, the 123 usable responses represented a response rate in excess
Table 7.8 Application rate and success of application by sex of owner
Sex of Owner
Female Male
Applied for funding in last 3 years? Yes 13 54% 55 56%
No 11 46% 43 44%
Total 24 98
Pearson c2 = 0.030, p value = 0.863
Application for funding successful? Yes 11 85% 46 88%
No 2 15% 6 12%
Total 13 52
Pearson c2 = 0.143, p value = 0.706
Note: One respondent failed to specify sex and three male respondents failed to indicate
whether their applications had been successful.
Source: Watson, Newby and Mahuka (2009, Table 2).
of 25%. This was a good response rate for a target group of this type (given
the length of the questionnaire – 16 pages) and, no doubt, was aided by the
offer of a A$30 payment in return for a completed questionnaire (Newby,
Watson and Woodliff 2003).
The results presented in Table 7.8 indicate that a little over 50% of
the SME owners had applied for funding in the last three years, with
the application rates about the same for both females and males. This
finding suggests that female SME owners are not being ‘discouraged’ from
applying for bank funding by perceptions of bank discrimination against
women. Further, the results indicate that about 12% of SME owners have
their applications for funding denied, again with no significant difference
between the female and male SME owners. The fact that there was no
significant difference in the rejection rates for the female and male SME
owners suggests that the banking sector does not routinely discriminate
against women. It should also be noted that the 12% rejection rate for this
Australian sample is similar to the 10% reported by Levenson and Willard
(2000) for the US and the 11% reported by Fraser (2006) for the UK.
Given that Levenson and Willard (2000, p.83) conclude that the ‘extent
of true credit rationing appears quite small’ and Fraser (2006, p.123) con-
cludes ‘that most SMEs are getting the finance they want’, this suggests
that it is also unlikely (given the similar rejection rates) that any substan-
tial supply-side ‘finance gap’ exists in Australia.
To further explore the ‘finance gap’ issue, Table 7.9 presents the
main reasons why external (debt) funding had not been sought by the
80 SME performance
Table 7.9 Reasons for not applying for external funding in the last three
years
7 point scale: Mean scores for 2-tailed

1 = Strongly disagree to p value
Females Males
7 = Strongly agree
The firm had sufficient funds 6.75 5.77 0.08
under its existing
arrangements
The firm did not need 6.10 5.73 0.48
additional funds
Procedures to obtain funding 2.38 3.82 0.11
from a bank were too
complicated
Interest rates were too high 1.50 2.87 0.01
It was unlikely the bank would 1.00 2.53 0.00
provide the full amount
A previous loan application 1.00 1.42 0.41
was rejected
respondents during the last three years. The relevant respondents were
given a list of items and asked to rate each using a seven-point scale
ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (7).
The results presented in Table 7.9 indicate two primary reasons why
SME owners do not apply for funding from financial institutions: first,
because they have access to sufficient funding under their current arrange-
ments; and second, because they do not require additional funding. These
results are consistent with Fraser’s (2006) finding that 95% of businesses
that had not sought new finance reported that they did not need additional
funding. Interestingly, the results presented in Table 7.9 suggest that, com-
pared to the male SME owners, the female SME owners were more likely
to believe they already had sufficient funds under existing arrangements
and, therefore, had no need to apply for additional external funding. Also
of interest were the responses to the last two items in Table 7.9, which
indicate that the females strongly disagreed with the premise that the
reason they had not applied for external funding in the last three years was
because they had been rejected previously or because they were unlikely to
get the full amount requested. Taken together, and consistent with Fraser’s
(2006) finding in the UK, these results strongly suggest that female SME
owners are not discriminated against by Australian financial institutions,
Table 7.10 Funding terms
Mean scores for 2-tailed

p value
Females Males
How long did it take the bank to approve 19 21 0.85
the loan (days)
Term that applied to funding (months) 119 56 0.22
Interest rate at inception (per annum) 7.41% 8.35% 0.50
Interest rate currently (per annum) 8.88% 8.50% 0.76
Table 7.11 Level of satisfaction with lending institution

1 =Totally dissatisfied to p value
Females Males
7 = Totally satisfied
The amount granted by the 6.64 6.48 0.97
bank relative to the
amount requested
Customer service 4.73 5.22 0.41
The time taken to process the 5.00 4.96 0.95
application
The interest rate charged 4.55 4.60 0.94
The guarantees required 5.00 4.16 0.19
The security required 4.82 4.15 0.31
The service fees charged 3.82 4.04 0.72
and they are not therefore discouraged from applying for external funding
because of any perceptions of bias against them. They might, however, be
less inclined to access external funding for other reasons, such as a desire
to maintain control and/or to minimize the risks they are exposed to.
Tables 7.8 and 7.9 suggest that financial institutions in Australia do
not discriminate against female SME owners and female SME owners
are no more likely to be discouraged from applying for external funding
than their male counterparts. Tables 7.10, 7.11 and 7.12 present further
evidence to support this conclusion.
For the female and male SME owners in the sample, Table 7.10 pro-
vides a comparison of the time taken to approve a loan, loan duration
82 SME performance
Table 7.12 SME owner’s perceptions of discrimination by financial

institutions

1 = Strongly Disagree to p value
Females Males
7 = Strongly Agree
Male applicants are more likely 3.83 3.84 0.99
to have their application
approved
Female applicants are more 3.46 3.72 0.58
likely to have to provide
security
likely to need loan
guarantors
likely to incur higher interests
rates
and the interest rate charged. As can be seen from the results, there are
no significant differences between the female and male SME owners with
respect to these factors. This provides further evidence that Australian
financial institutions do not routinely discriminate against female SME
owners.
Table 7.11 provides a comparison of the levels of satisfaction experi-
enced by both the females and males with respect to their lending institu-
tion in terms of issues such as time taken to process an application and
the interest rate charged. As can be seen from the results, there are no
significant differences in the average satisfaction levels for the female and
male SME owners and for all but the last item (the service fees charged)
the female and male SME owners rated their level of satisfaction above the
mid-point on the scale. This again suggests that financial institutions do
not routinely discriminate against female SME owners and that the major-
ity of SME owners (both female and male) are reasonably satisfied with
their financial institution.
Finally, Table 7.12 provides a comparison of the responses given by
both the female and male SME owners concerning their views on potential
differences in the treatment men and women might receive when apply-
ing for funding from a financial institution. These responses suggest that
neither the female nor the male respondents believed women suffer any
Table 7.13 Comparing the personal characteristics of SME owners

1 = Low to 7 = High p value
Females Males
(Non-Applicants) (Applicants)
Risk-taking propensity (av. of 2.85 3.37 0.01
10 items)a (3.00) (3.47) 0.00
Internal locus of control (av. of 5.01 5.15 0.47
8 items)b (5.08) (5.16) 0.60
Notes:
a Risk-taking propensity was assessed using statements taken from the risk sub-scale
of the Jackson Personality Inventory (1976).
b Internal locus of control was assessed using statements taken from Levenson
(1974).
systematic discrimination when applying for bank funding. The mean

response for both the females and males was close to the mid-point for all
statements except the last, where there was general disagreement (from
both the female and male SME owners) with the statement that females
are more likely to incur higher interest rates (than males).
With the results so far providing strong support for the proposition that
Australian SMEs do not face a ‘finance gap’, the remainder of this chapter
will look at various demand-side issues to see if they might provide a better
explanation for any observed differences in the level of external funding
across SMEs. Table 7.13 summarizes the mean scores for a number of
items (questions) designed to capture an SME owner’s risk aversion and
desire to maintain control and compares the results for females and males
and for non-applicants and applicants.
The results presented in Table 7.13 suggest that both female SME
owners and non-applicants have lower levels of risk-taking propensity
compared to male SME owners and applicants. This conclusion is con-
sistent with the findings reported in Chapter 6, indicating that female-
controlled SMEs appear to be significantly less risky than male-controlled
SMEs. In terms of desire to maintain control, however, the findings
reported in Table 7.13 indicate no significant difference between either
females and males or non-applicants and applicants.
To further explore both these issues (risk aversion and desire to main-
tain control), the respondents were asked to assess the relevance of various
factors in potentially discouraging them from applying for external
84 SME performance
Table 7.14 Factors discouraging SME owners from applying for funding
in the future

1 = Not relevant to p value
Females Males
7 = Very relevant
Likelihood of unreasonable 5.13 5.40 0.54
terms and conditions (5.42) (5.28) 0.70
Risk of not being able to repay 5.08 4.39 0.12
the loan (4.87) (4.08) 0.04
Potential to lose control of the 4.65 3.75 0.07
business (4.15) (3.60) 0.17
Amount of documentation 3.96 4.18 0.62
required (3.97) (4.34) 0.28
Probability of not getting the 3.13 2.56 0.19
loan (2.93) (2.30) 0.07
Apprehension over having the 2.91 2.05 0.02
loan application rejected (2.46) (1.89) 0.05
funding in the future. From the results presented in Table 7.14, it seems
that for all groups of SME owners (that is, females, males, non-applicants
and applicants) the prospect of unreasonable terms and conditions is likely
to be the most relevant factor deterring them from applying for external
funding in the future. The second most relevant factor for all groups is the
risk of not being able to repay the loan, with the non-applicants ranking
the relevance of this factor significantly higher than the applicants. The
potential to lose control of their business was the third most important
factor for the females and non-applicants and the fourth most important
factor for the males and applicants. Interestingly, the females ranked this
concern significantly higher than their male counterparts, suggesting,
contrary to the internal locus of control results reported in Table 7.13,
that (compared to males) female SME owners might indeed have a greater
desire to maintain control over their businesses.
It should also be noted in Table 7.14 that the two least relevant factors
likely to discourage SME owners from applying for external funding in the
future are the probability of not getting the loan and apprehension over
having their loan application rejected. These two factors were rated as the
Table 7.15 Most likely use of surplus funds

1 = Highly Unlikely to p value
Females Males
7 = Highly Likely
Repay debt 5.88 5.20 0.05
(5.71) (4.89) 0.03
Grow the current business 4.71 4.60 0.80
(5.09) (4.06) 0.00
Improve lifestyle 4.67 4.13 0.23
(4.16) (4.34) 0.63
Invest in other ventures 4.17 4.36 0.70
(4.45) (4.15) 0.44
least relevant for all four groups. In summary, the evidence provided in
Tables 7.13 and 7.14 supports the proposition that it is more likely that
individual owner characteristics (such as desire to maintain control and
risk aversion) are responsible for variations in the level of external funding
across SMEs rather than these variations being the result of systematic
discrimination by financial institutions.
Finally, Table 7.15 provides the mean scores given by the SME owners
(female/male and non-applicants/applicants) when asked how they would
most likely use any surplus funds generated by their business. As can be
seen from the table, the repayment of debt was the priority for all owners,
and particularly for the female owners and the non-applicants. This indi-
cates that SME owners have an aversion to borrowing funds from finan-
cial institutions and would, therefore, prefer to repay debt rather than use
surplus funds for other purposes.
7.3 SUMMARY
In summary, the findings reported in this chapter, particularly for female-

controlled SMEs, support the view expressed by Hamilton and Fox (1998,
p.239) that the ‘supposed gaps in the supply of finance to small firms might
be in part a consequence rather than the cause of financing decisions of
the business owners’. As such, the results do not support the findings of
Becchetti and Trovato (2002, p.298), suggesting that the ‘availability of
external finance is a constraint to small firm . . . growth’.
86 SME performance
Further, there is no evidence of any perceived, or actual, bank discrimi-

nation against female SME owners. These issues will be explored further
in the following chapter.
NOTE
1. Note that this view is consistent with the argument advanced by Watson and Newby
(2005) that biological sex might not be an appropriate discriminator when examining
differences in the psychological attributes of SME owners. Instead, they suggest that the
use of masculine and feminine traits might prove more useful in future research.
8. A quantitative analysis
8.0 INTRODUCTION
This chapter builds on the key findings of the previous chapter suggesting
that banks do not routinely discriminate against female SME owners and
that the notion of a ‘finance gap’ might be more ‘myth’ than ‘reality’. In
particular, this chapter will contrast the implications that arise from Myers’
Pecking Order Theory (1984) with those that might be expected if, indeed,
there is a ‘finance gap’ negatively impacting growth in the SME sector.
8.1 MYERS’ PECKING ORDER THEORY
Based on the notion of asymmetric information and the costs of financial

distress, Pecking Order Theory (Myers 1984) implies that SME owners
will prefer to use internal sources of finance before external sources.
According to this theory, SME owners will only consider external financ-
ing options where insufficient internal funds are available to take advan-
tage of value adding opportunities. This suggests that younger firms that
have not yet had the opportunity to generate significant levels of internal
funds are more likely (than older firms) to require external funding. Based
on Pecking Order Theory, we would expect a firm’s relative level of exter-
nal debt to fall over time as the firm replaces external debt with internally
generated funds. In other words, Pecking Order Theory suggests that,
other things being equal, older firms will have lower levels of external debt
than younger firms.
8.2 THE NOTION OF A ‘FINANCE GAP’
However, there are no guarantees that external finance providers will

always be willing to lend to young SMEs with positive net present value
projects. As noted by Berger and Udell (1998), it is difficult for providers of
external finance to readily verify that a young SME has access to a quality
project (adverse selection problem) or to ensure that the funds provided
87
88 SME performance
High
Finance gap
Relative
level of
external
funding
Pecking Order Theory
Low
Younger Firm age Older
Source: Watson (2006, Figure 1).
Figure 8.1 Pecking Order Theory versus the notion of a ‘finance gap’
will not be diverted to alternative projects (moral hazard problem). This

has led to the belief that there is a significant ‘finance gap’ within the SME
sector. Given that younger firms are more ‘informationally opaque’ (than
older firms), if there is a ‘finance gap’ it is more likely to be a problem for
younger firms.
The above discussion leads to two competing propositions. If Pecking
Order Theory prevails, we would expect younger firms to have relatively
higher external debt levels than older firms because, over time, older firms will
replace external debt with internally generated sources of funds. However, if
the notion of a ‘finance gap’ prevails then we would expect younger firms
to have relatively lower external debt levels than older firms, because older
firms will find it easier to access external debt as they become less ‘informa-
tionally opaque’. These competing propositions are depicted in Figure 8.1.
8.3 FEMALE-CONTROLLED SMES AND BANK

DISCRIMINATION
As noted earlier, there is a perception that banks routinely discriminate

against female SME owners (Riding and Swift 1990). If this were true, we
A quantitative analysis 89
would expect to find relatively lower levels of external funding in female-

owned SMEs compared to male-owned SMEs, other things being equal.
However, we might also expect female-controlled SMEs to have relatively
lower levels of external funding for other reasons, regardless of any bank
discrimination. For example, part of the reason owners prefer to use inter-
nal (rather than external) funds has to do with the risk of default. Powell
and Ansic (1997) note that a lower preference for risk amongst females
is one of the gender differences which is persistently found in both the
general and business-specific literature. For example, in a US study com-
paring 105 female business founders (ranked in the top 10% with respect
to sales and number of employees) with similar male business owners,
Sexton and Bowman-Upton (1990) found that the females scored signifi-
cantly lower on traits related to risk taking. The scores that they reported
(p.29) indicate that female entrepreneurs are ‘less willing than their male
counterparts to become involved in situations with uncertain outcomes
(risk taking)’. Powell and Ansic (1997) also note that females tend to focus
on strategies that will avoid the worst situation to gain security. Given the
extensive body of literature suggesting that females are more risk averse
than males (Watson and Robinson 2003), it seems reasonable to suggest
that female-controlled firms (compared to male-controlled firms) might
have lower levels of external funding by choice, and not because of any
bank discrimination. That is, Pecking Order Theory might apply more
strongly to female than to male SME owners, causing female SME owners
to have a stronger preference for internal (rather than external) sources of
funds.
Further, Cressy (1995, p.291) notes that: ‘Loan capital is productive
and increases the firm’s revenue, but brings the business under the control
of the bank. Profits generated by borrowing are a good increasing owner
utility, whereas control is a bad, reducing it.’1 Also, Mukhtar (2002)
found that women had a significantly greater need (compared to men) to
be in control of all aspects of their business. If female SME owners are
(on average) more concerned than male SME owners about the prospect
of losing control of their businesses, this should also cause female SME
owners to have a stronger preference for using internal (rather than exter-
nal) sources of funding and might, therefore, also cause Pecking Order
Theory to apply more strongly to female-controlled SMEs.
In summary, the above discussion suggests that (other things being
equal) female-controlled SMEs will have lower levels of external funding
than male-controlled SMEs, either because Pecking Order Theory applies
more strongly to females or because banks routinely discriminate against
females. In the latter case, we would expect young female-controlled
SMEs to have relatively lower levels of bank debt compared to young
90 SME performance
male-controlled SMEs, with this gap narrowing as firms get older (and
become less ‘informationally opaque’). Alternatively, if Pecking Order
Theory applies more strongly to females than males (and banks don’t
routinely discriminate against females) we would expect both male- and
female-controlled SMEs to have relatively similar debt levels in their early
years, but with a more noticeable reduction in external debt over time for
the female-controlled SMEs. These competing propositions will be tested
in section 8.5, but before that the relationship between external funding
and firm growth is discussed in the next section.
8.4 THE RELATIONSHIP BETWEEN EXTERNAL

(BANK) FUNDING AND GROWTH
As noted earlier, the literature suggests a strong relationship between

external (debt) funding and SME growth. To test this proposition, I
will again analyse the reduced sample of 2236 male-controlled and 131
female-controlled SMEs that existed for the entire four-year period of the
Australian Bureau of Statistics (ABS) data set (as discussed in Chapter 6,
section 6.3). Growth is measured as (total income in 1997–98/total income
in 1994–95) − 1. Total income is used as the measure of growth because
Delmar, Davidsson and Gartner (2003, p.194) note that there ‘seems to
be an emerging consensus that if only one indicator is to be chosen as a
measure of firm growth, the most preferred measure should be sales’. The
SMEs are divided into two groups: those with high (above the median)
growth – coded 1; and those with low (below the median) growth – coded
0. Bank loans and bank overdrafts are classified as external (bank) debt.
In any comparison of external debt levels across firms it is important
to control for firm size because larger firms can be expected to have more
assets available to offer as security (collateral) and more profits available
to service debt. This will be particularly important in any comparison of
female- and male-controlled SMEs, because female-controlled SMEs are
generally smaller (in terms of both assets and profits) than male-controlled
SMEs (Loscocco et al. 1991; Cliff 1998; Watson 2002). Firm size is control-
led by examining a firm’s bank debt relative to its total assets, that is, by
examining a firm’s debt-to-asset ratio (dtar), rather than its absolute level
of debt. For purposes of analysis, the firms are divided into two groups:
those with a high (above the median) dtar – coded 1; and those with a low
(below the median) dtar – coded 0.
The analysis also controls for a number of other variables that could
potentially confound the relationship between firm growth and external
funding. First, the natural log of a firm’s average annual profit over the
four-year period is included in the analysis as a continuous independent

(control) variable. Other things being equal, more profitable firms will
not only be able to service larger loans but will also have more internal
funds available to substitute for external sources of funds and to facili-
tate firm growth. Second, firm age is included as a categorical variable
because previous studies have shown that growth is inversely related to
firm age and also because, other things being equal, younger firms are
likely to find it harder to obtain external financing because they are more
‘informationally opaque’ than older firms. Firms less than ten years old
were classified as younger firms and were coded 1, while firms ten or more
years old were classified as older firms and coded 2. Third, the sex of the
owner is included because previous research suggests that women are more
risk averse (Watson and Robinson 2003) and prefer steady rather than
rapid growth (Cliff 1998). Female-controlled firms are coded 1 and male-
controlled firms are coded 2. Fourth, industry is included because, again,
previous research indicates significant variation in firm performance across
different industries. Fifth, type of legal organization has been included as
a control variable because Becchetti and Trovato (2002, p.292) note that
the ‘higher personal wealth at risk in unlimited liability firms reduces
incentives to invest in risky opportunities which may foster firm growth’.
Limited liability (incorporated) firms are coded 1 with unlimited liability
(unincorporated) firms coded 2. Finally, Cressy (1996) argued that much
of the prior research showing a relationship between capital availability
and SME success was flawed, because these prior studies had typically
not controlled for the owner’s human capital. That is, entrepreneurs with
greater levels of human capital are likely to be both more successful and
better able to secure external funding. Therefore, two additional variables
have been included to control for the owner’s human capital: education
and years of experience. The ABS database provided details concerning
the highest level of education achieved by the principal decision maker
according to the following four categories: school – coded 1; trade – coded
2; tertiary (non-business) – coded 3; and tertiary (business) – coded 4. The
owner’s years of experience was provided as a continuous variable.
Table 8.1 provides the descriptive statistics for the high/low dtar firms
and the high/low growth firms. From the table it can be seen that, except
for years of experience, there appears to be a significant relationship
between each of the independent (control) variables and whether the firm
has a high or low dtar. In terms of firm growth, however, only average
profits appears to be relevant. Somewhat surprisingly, dtar does not
appear to be related to firm growth.
Table 8.2 presents the results of running two binary logistic regression
models for firm growth. Model 1 includes all the independent (control)
92 SME performance
Table 8.1 Descriptive statistics for firms with high/low dtar and high/low
growth
Variables Debt-to-asset ratio Growth

High Low High Low
Average profits ’000 (median) 36 62** 69 30**
Debt-to-asset ratio (%)
High (above the median) 51 49
Low (below the median) 49 51
Age of firm (%) *
Less than 10 years old (younger) 47 53 51 49
10 or more years old (older) 52 48 49 51
Industry (%) **
Mining 9 91 9 91
Manufacturing 49 51 48 52
Construction 55 45 54 46
Wholesale trade 49 51 50 50
Retail trade 61 39 51 49
Accomodation, cafes & restaurants 62 38 40 60
Transport & storage 62 38 54 46
Finance & insurance 41 59 54 46
Property & business services 44 56 50 50
Cultural & recreational services 44 56 55 45
Personal & other services 54 46 60 40
Type of legal organization (%) **
Incorporated (limited liability) 47 53 51 49
Unincorporated (unlimited liability) 58 42 48 52
Sex of owner (%) *
Female 40 60 56 44
Male 51 49 50 50
Education (%) **
School 55 45 50 50
Trade 54 46 49 51
Tertiary (non-business) 43 57 52 48
Tertiary (business) 44 56 49 51
Years of experience (median) 14 13 14 14
Note: * Significantly different at 5%; ** Significantly different at 1%.

Table 8.2 Logistic regression models for SMEs with high/low growth
Variables Model 1 Model 2

Wald Sig Exp(B) Wald Sig Exp(B)
Debt-to-asset ratio (dtar) 2.54 0.11 0.86
Average profits (logged) 31.30 0.00 1.16 32.54 0.00 1.17
Firm age 2.99 0.08 1.19 3.31 0.07 1.20
Sex of owner 2.73 0.10 1.42 3.10 0.08 1.46
Industry 17.89 0.06 17.48 0.06
Mining 8.08 0.00 0.04 7.86 0.01 0.04
Manufacturing 2.35 0.13 0.60 2.34 0.13 0.60
Construction 0.47 0.49 0.77 0.47 0.49 0.77
Wholesale trade 2.60 0.11 0.57 2.59 0.11 0.57
Retail trade 0.85 0.36 0.72 0.95 0.33 0.71
Accom., cafes & restaurants 4.02 0.05 0.38 4.16 0.04 0.38
Transport & storage 0.01 0.92 0.96 0.03 0.86 0.93
Finance & insurance 0.13 0.72 0.86 0.09 0.76 0.88
Property & business services 1.22 0.27 0.68 1.13 0.29 0.69
Cultural & recreation services 1.17 0.28 0.60 0.89 0.35 0.63
Type of legal organization 3.17 0.08 1.20 3.44 0.06 1.21
Owner education 3.18 0.37 3.43 0.33
School 1.32 0.25 1.17 1.25 0.27 1.16
Trade 0.94 0.33 1.15 0.90 0.34 1.15
Tertiary (non-business) 3.14 0.08 1.30 3.42 0.06 1.31
Years of experience 0.03 0.87 1.00 0.04 0.84 1.00
Nagelkerke R Square 0.039 0.040
Note: In running the logistic regression, the last category was used as the reference point
for each categorical variable and, therefore, is not shown in the table. For industry the last
category is ‘Personal and other services’ and for education the last category is ‘Tertiary
(business)’.
variables, except for a firm’s dtar, which is added in model 2. The results
in model 2 (consistent with the descriptive statistics provided in Table
8.1) suggest that a firm’s dtar is not a significant factor in explaining firm
growth. Running a separate analysis for the female- and male-controlled
SMEs confirmed the results for the whole group.
The logistic regression results presented in Table 8.2 indicate that, rather
than being associated with a firm’s dtar, firm growth appears to be highly
associated with a firm’s average profits and, to a lesser extent, with firm
age, industry and type of legal organization. Firm growth also appears
94 SME performance
to be associated with the owner’s sex (but only weakly), but not with the
owner’s human capital (education or experience). The positive associa-
tion between growth and average profits is expected because (other things
being equal) profitable firms will have more funds available to fund growth
opportunities. This result is also consistent with Carpenter and Petersen’s
(2002) finding that firm growth, for a sample of 1600 US small manufac-
turing firms, appeared to be constrained by the lack of internal funds. The
association between firm growth and firm age is also as expected, with
younger SMEs being more likely to be in the high-growth group.
Somewhat surprisingly, given that most of the literature suggests that
the growth of female-controlled SMEs is constrained by a lack of external
finance, the findings reported in Table 8.2 indicate that female-controlled
SMEs are more likely (than male-controlled SMEs) to be in the high-
growth group (although this finding is only significant at the 10% level).
Also somewhat unexpectedly, incorporated (limited liability) firms
appeared less likely to have a high dtar compared to unincorporated
(unlimited liability) firms. This result suggests that for unincorporated
firms the banks are probably requiring additional security over the
owner’s personal assets.
8.5 PECKING ORDER THEORY VERSUS ‘FINANCE

GAP’
Table 8.3 presents the results of examining the mean and median dtar for
younger and older female- and male-controlled SMEs to determine the
possible impact of age on the relative debt levels of these two groups. The
results in this table indicate that while male-controlled SMEs have a sig-
nificantly higher dtar, overall, there is no significant difference in the dtar
Table 8.3 DTAR for female- and male-controlled SMEs by age
Firm Age Number Mean (%) Median (%)

Female Male Female Male Female Male
Younger (< 10 years) 76 1027 16 25 10 13
Older (≥ 10 years) 54 1204 12 26 5 16**
All firms 130 2231 15 26 8 15**
Note: ** Significantly different at 1% using a two-tailed test.

Table 8.4 Four-year growth rates for female- and male-controlled SMEs
by age
Firm age Number Mean (%) Median (%)

Female Male Female Male Female Male
Younger (< 10 years) 77 1028 25 59 10 11
Older (≥ 10 Years) 54 1208 41 26 18 8*
All firms 131 2236 32 41 17 9
Note: * Significantly different at 5% using a two-tailed test.
for the younger male- and female-controlled SMEs. This finding suggests
that banks do not routinely discriminate against females but, rather, it is
likely that female SME owners prefer to have lower debt levels and will,
therefore, take the opportunity to repay debt over time from the cash flows
generated by their firms. This conclusion is consistent with the qualitative
findings presented in Chapter 7.
Table 8.4 presents the results of examining mean and median four-year
growth rates for younger and older female- and male-controlled SMEs to
determine the possible impact of firm age on firm growth for these two
groups. The results indicate that, for all firms, there is no significant differ-
ence in the growth rates for female- and male-controlled SMEs. However,
for older firms, it would appear that female-controlled SMEs experience
higher growth rates than male-controlled SMEs. Further, this result
appears to be driven by a decline in growth rates for older male-controlled
SMEs (which was significant at p = 0.05, not reported) rather than an
increase in growth rates for older female-controlled SMEs (which was not
significant).
This finding of a decline in growth rates over time for male-controlled
SMEs is consistent with most of the previous literature suggesting a
negative relationship between firm growth and age (Becchetti and Trovato
2002). However, the finding of no significant change in growth rates over
time for the female-controlled SMEs is also consistent with Cliff’s (1998)
conclusion that female SME owners prefer steady rather than fast-paced
growth. It is also possible that many female SME owners are unable to
devote the necessary time and energy required to achieve rapid growth in
the early years of their business because of family responsibilities. As these
diminish, female SME owners are more likely to have the time, energy and
financial resources needed to develop and expand their ventures.
96 SME performance
8.6 SUMMARY
The findings reported in this chapter, and Chapter 7, indicate that Pecking
Order Theory, rather than bank discrimination, might provide a better
explanation for any observed differences in the level of external funding
between male- and female-controlled SMEs. Because female SME owners
are, on average, more risk averse and have a greater need to feel in control
of their businesses (than male SME owners) they will be less inclined to
access external funding. The fact that the relatively lower levels of external
funding in female-controlled SMEs is most noticeable in older firms (with
established track records) is consistent with the proposition that lower
levels of external funding in female-controlled SMEs is the result of per-
sonal choice (Pecking Order Theory) rather than bank discrimination.
Contrary to some prior research, the findings reported in this chapter
indicate that growth is not significantly associated with a firm’s relative
level of external (bank) funding, but is associated with a number of other
firm-level variables, in particular, firm profitability. Further, the findings
suggest that there is no significant difference in the overall growth rates for
female- and male-controlled SMEs in Australia. However, for older firms,
female-controlled SMEs appear to have significantly higher growth rates
than male-controlled SMEs. For younger firms, there appears to be no
difference in the growth rates for the female- and male-controlled SMEs.
This result is somewhat puzzling, given that the older female-controlled
SMEs also had relatively lower levels of external (bank) debt available to
fund firm growth. The implication is that, compared to male-controlled
SMEs, female-controlled SMEs have more internally generated funds.
Given that female- and male-controlled firms have similar levels of relative
profitability (see Chapter 5), the higher level of internally generated funds
available to female-controlled SMEs can only be the result of female SME
owners withdrawing less funds from their businesses (for example, in the
form of wages, fees, dividends or drawings). As the ABS data does not
include any information on amounts paid to (withdrawn by) the owner,
this is an issue that future research might usefully investigate.
Finally, while the findings reported in this and the previous chapter are
limited to external (bank) debt, it is conceivable that they could equally
apply to external equity funding. That is, female SME owners might be
less inclined to access external equity funding because of the risks involved
and the potential loss of control that might follow.
Given that there has only been a limited amount of research focusing
‘on small, growing entrepreneurial companies and the factors affecting the
capital structure of these firms’ (Michaelas, Chittenden and Poutziouris
1999, p.114), particularly female-controlled SMEs (Brush et al. 2001), the
results provided in these two chapters should help to better inform finance
providers, business advisers and policy makers about the determinants of
SME growth and the various demand-side issues surrounding the provi-
sion of growth funding for this sector.
I trust the material presented in this and the previous chapter will have
convinced the reader that banks do not routinely discriminate against
female SME owners and that the notion of a ‘finance gap’ might be more
‘myth’ than ‘reality’.
NOTE
1. Note that Cressy (1995) argues that for owners to experience a loss of control does not
require the bank to have an equity stake in the firm; it might, for example, take the form
of monitoring activities imposed on the firm.
PART V
Networking and SME performance
In this section I intend to examine the relationship between networking

and SME performance and, in particular, to demonstrate that female
SME owners are not disadvantaged, relative to male SME owners, in
terms of their networking activities.
Network theory suggests that the ability of owners to gain access to
resources not under their control in a cost-effective way through network-
ing can influence the success of business ventures (Zhao and Aram 1995).
Florin, Lubatkin and Schulze (2003) argue that networking can provide
value to members by allowing them access to the social resources embed-
ded within a network; that is, networking can provide the means by which
SME owners can tap needed resources that are ‘external’ to the firm (Jarillo
1989). Julien (1993) notes that this form of cooperation can facilitate the
achievement of economies of scale in small firms without producing the
diseconomies caused by large size. Using networks can, therefore, poten-
tially lower a firm’s risk of ‘failure’ and increase its chances of ‘success’.
It has also been suggested that there might be significant differences
between males and females in terms of their network use. For example,
Cromie and Birley (1992) argue that networks are the product of personal
drive and historical experiences, and the social structure and domestic
duties of many women might result in female entrepreneurs having fewer
networks than their male counterparts. Aldrich (1989) notes that these
differences in network use could have a significant impact on the rate at
which women (compared to men) start new ventures and the performance
of those ventures.
However, although the arguments in favour of networking appear com-
pelling, and most of the existing literature is premised on the belief that
networking is beneficial (Havnes and Senneseth 2001), there has been little
100 SME performance
empirical evidence to date of an association between firm performance

and the owner’s use of networks, particularly for established businesses.
Indeed, Aldrich and Reese (1993) were unable to find any evidence linking
an entrepreneur’s use of networks to business survival or performance
and, similarly, Cooper, Gimeno-Gascon and Woo (1994) were unable to
find a significant relationship between the use of professional advisers and
firm survival. Similarly, although there has been considerable conjecture
about the possible networking differences between men and women, few
empirical studies exist examining gender differences in networking.
The aim of this section, therefore, is to investigate the possible asso-
ciation between the networking activities of SME owners and the per-
formance of their firms. Chapter 9 examines the relationship between
networking and SME performance generally, and then Chapter 10 specifi-
cally focuses on gender differences in networking and the possible impact
of any such differences on the performance of female- and male-controlled
SMEs. Consistent with most prior empirical studies on networking, the
material presented in the next two chapters focuses on the personal net-
works of the SME owner, rather than on the organizational networks of
the business (Brüderl and Preisendörfer 1998).
9. The association between networking
and performance
9.0 INTRODUCTION
While there are many factors that can influence the success of a venture
and there are various risk reduction strategies that can be employed to
increase a firm’s chances of survival (see, for example, Duchesneau and
Gartner 1990; Cooper 1993; Cooper et al. 1994; Robson and Bennett
2000; Shepherd, Ettenson and Crouch 2000; Larsson, Hedelin and Garling
2003), only recently have researchers begun to highlight the potential
significance to SME performance of an owner-manager’s networking
involvement. Coleman (1988) notes that information is important to deci-
sion making but is costly to obtain and that networks provide a means
by which important information can potentially be acquired in a cost-
effective manner. Therefore, networking can enhance an SME owner’s
social capital (Coleman 1988) because it provides access to information
embedded within the networks accessed. Further, Granovetter (1983)
argues that individuals whose networks (and, therefore, main source of
information) comprise primarily family and friends (strong ties) are likely
to have access to less information than individuals whose networks include
many acquaintances (weak ties). Presumably for this reason, Fischer and
Reuber (2003) suggest that owners of high-growth firms need to develop
ties beyond their personal circle of contacts and local communities.
Similarly, innovation theory suggests that networks (particularly those
comprised of many weak ties) are important in diffusing innovations and,
therefore, SMEs whose owners are heavily involved in networking should
outperform SMEs whose owners make limited (or no) use of networks
(Havnes and Senneseth 2001).
In support of the foregoing propositions (and despite Aldrich and Reese
(1993) and Cooper et al. (1994) being unable to find a significant relation-
ship between networking and firm performance), there have been a limited
number of studies that have documented a positive association between
networking and various aspects of firm performance. For example,
Duchesneau and Gartner (1990) found that successful firms were more
likely to have used professional advice. Potts (1977) noted that successful
101
102 SME performance
companies relied more heavily on accountants’ information and advice

than did unsuccessful companies. Kent (1994) found that the financial per-
formance of a group of small pharmacy businesses was positively related
to using external management advisory services. Donckels and Lambrecht
(1995) found that network development, particularly at the national and
international level, was positively associated with firm growth. Lerner,
Brush and Hisrich (1997) found that network affiliation was significantly
related to profitability, and that the use of outside advisers was related to
revenue. Larsson et al. (2003) found that a lack of contacts with outside
expert advisers was an obstacle to the expansion of small businesses.
Hustedde and Pulver (1992) found that entrepreneurs who failed to seek
assistance were less successful in acquiring equity capital and, similarly,
Carter et al. (2003) reported that the more varied the group of business
advisers a women business owner consulted, especially professional advis-
ers, the more likely she was to succeed in securing equity financing.
9.1 THE NATURE OF NETWORKS AND NETWORK

ACCESS
Seibert, Kraimer and Liden (2001) provide a useful summary and discus-
sion of the three conceptualizations of social capital found in the litera-
ture. First, there is weak tie theory as proposed by Granovetter (1973).
Here the focus is on the strength of social ties and it is argued that net-
works comprising strong ties (such as family and friends) are more likely
to be the source of redundant information than would be the case where
networks comprise weak ties (such as acquaintances). Second, is Burt’s
(1992) notion of structural holes. A structural hole is deemed to exist
where two individuals are not connected in any way. Here the focus is not
on the direct ties between SME owners and individual members of their
network but, rather, on the relationships between the various members in
an SME owner’s network. An SME owner whose network contains many
structural holes (that is, few of the other members of the network are con-
nected) is likely to have ‘more unique and timely access to information’
(Seibert et al. 2001, p.221). Third, is social resource theory (Lin, Ensel
and Vaughn 1981), which focuses on the nature of the resources embed-
ded within a network, rather than on the strength of ties or the existence
of structural holes. So, while weak tie theory and structural hole theory
examine the links between members of a network, social resource theory
is concerned with the nature of information (social resources) held by
individual members of the network.
A variety of terms can be found in the literature to describe the
Networking and performance 103
important properties of personal networks. For example, Munch,

McPherson and Smith-Lovin (1997) refer to network size, contact
volume and composition; Moore (1990) refers to network range and the
volume and diversity of contacts; Zhao and Aram (1995) refer to network
range and intensity; and Ibarra (1992) refers to network composition,
homophily, tie strength, range, density and the distinction between formal
and informal networks.
The focus of this and the following chapter is on the number of net-
works SME owners use and their frequency (volume) of contact. Network
composition will also be examined using Ibarra’s (1992) classification
of networks as either formal or informal, with formal networks likely to
comprise more weak ties and structural holes (and, therefore, to be more
beneficial) than informal networks. Littunen (2000) suggests that formal
networks would include the likes of accountants, banks, lawyers and trade
associations, while informal networks would comprise; for example, busi-
ness contacts, family and personal relationships. The distinction between
formal and informal networks raises an interesting question: namely,
are formal networks (which typically cost more to access but generally
comprise weaker ties) of greater benefit than informal networks (which
typically cost less to access but generally comprise stronger ties)? Given
that formal networks result in more weak ties than informal networks,
and Granovetter’s (1983) argument that weak ties are likely to result in
the transfer of more information than strong ties, this suggests that formal
networks should be more important to firm performance than informal
networks.
As noted above, Zhao and Aram (1995) suggest that networking can
be understood in terms of range (the number of different networks that
owners are involved with) and intensity (the frequency with which owners
access those networks). This distinction raises a further interesting ques-
tion, namely, should SME owners attempt to have a very broad support
network that they access on a limited basis, or should they have fewer sup-
ports that they access on a more regular basis? In other words, is network
range more important to firm performance than network intensity?
Further, do the answers to these questions change over time as SME
owners gain more business and industry experience and/or as the owner’s
primary objective changes? For example, the relative benefits of accessing
a broad range of networks (infrequently) compared to accessing a nar-
rower set of networks (with more intensity) might well depend on the key
objectives of the SME owner. If the primary objective is to grow rapidly,
the SME owner might be best advised to develop and pursue a broad
range of networks because this is likely to result in more weak ties (and,
therefore, more information) than having a narrow range of networks. If,
104 SME performance
however, an SME owner has only recently started in business, and survival
is of utmost importance, the owner might require more intense help from
a reduced set of networks, particularly from formal network sources such
as external accountants and lawyers.
Again, the analysis presented in this (and the following) chapter is based
on the data set compiled by the Australian Bureau of Statistics (ABS), as
described in Chapter 2. The ABS data from the 1995–96 (second) survey
contained information relating to the frequency (never – coded 0; between
one and three times – coded 1; or more than three times – coded 2) with
which owners had sought advice during the year from seven formal sources
(banks; business consultants; external accountants; industry associations;
the Small Business Development Corporation (SBDC); solicitors/lawyers;
and the tax office) and three informal sources (family and friends; local
businesses; and others in the industry). There were 5027 responses to the
1995–96 (second) survey. However, on examining the data, it was found
that 13 businesses had no income (sales or other income). Therefore, they
were excluded from the analysis on the assumption that they were not
active businesses. This left 5014 firms (representing approximately 1.25%
Table 9.1 Range and intensity of formal and informal network access (%)
Range of Networks Accessed Intensity of Access

Never 1–3 times >3 times
Formal networks
Bank 38 35 27
Business consultant 72 18 10
External accountant 20 35 45
Industry associations 60 21 19
SBDC 84 13 3
Solicitor/lawyer 43 34 23
Tax office 59 31 10
Average for formal networks 54 27 19
Informal networks
Family & friends 65 19 16
Local businesses 74 16 10
Others in the industry 45 29 25
Average for informal networks 61 22 17
Average for all networks 56 25 19

of eligible Australian SMEs) that could be examined over the three-year

period 1995–96 to 1997–98.
Table 9.1 summarizes the frequency (intensity) with which the SME
owners accessed the various formal and informal network sources for
advice. Consistent with Cooper, Woo and Dunkelberg (1989) and Robson
and Bennett (2000), the owners in this study accessed information from
a number of different sources (both formal and informal); with external
accountants, banks, others in the industry, solicitors, industry associa-
tions and family and friends being the most frequently accessed sources of
advice. For example, 45% of SME owners accessed an external account-
ant on more than three occasions during the 1995–96 year. This finding is
consistent with Robson and Bennett (2000), who reported that, from the
private sector, external accountants were the most widely accessed source
of advice, followed by banks and lawyers. However, unlike Birley (1985),
who found that entrepreneurs relied heavily on informal networks (but
seldom tapped into formal networks), the results presented in Table 9.1
suggest that Australian SME owners make extensive use of both formal
and informal networks.
9.2 THE ASSOCIATION BETWEEN NETWORKING

AND PERFORMANCE
Given the arguments advanced in favour of networking, and the balance

of available evidence, it would be reasonable to expect that the owners of
SMEs that survive and prosper are likely to be more involved in network-
ing than the owners of SMEs that fail, or are less prosperous. However,
Watson (2007) notes that the relationship between networking and firm
performance (measured as either survival, income growth or return on
equity) is unlikely to be linear. While it is reasonable to expect that some
level of networking will be beneficial, it is also plausible to suggest, con-
sistent with the law of diminishing returns, that excessive networking is
likely to be counterproductive. Economists have long argued that time is
the scarcest economic resource and how individuals allocate their time can
have profound economic effects (Uzzi 1997). Therefore, it is improbable
that an SME owner could spend excessive amounts of time networking
and still have the time necessary to run a sustainable business. Beyond
some limit, it is likely that the marginal benefit from further networking
will be more than offset by the negative impact of the owner’s lack of
available time to attend to other business matters. If this is true, we should
observe an inverted U-shaped relationship between firm performance and
the level of networking undertaken by SME owners.
106 SME performance
1.1
1.0
Probability of Survival
0.9
0.8
0.7
Linear
Quadratic
0.6
–10 0 10 20 30
Networking Score
Source: Watson (2007, Figure 1).
Figure 9.1 Probability of survival fitted against an SME owner’s

networking score
Figure 9.1 depicts the results of estimating the relationship between firm
survival and networking using both a linear and quadratic model. The
networking score variable depicted in Figure 9.1 can vary from 0 to 20 and
is the product of network range (which can vary from 0 to 10, based on
the total number of formal and informal networks) and network intensity
(which can vary from 0 to 2, based on the frequency of network contact –
with no contact coded 0, contact between one and three times coded 1 and
contact more than three times coded 2).
The figure indicates that an inverted U-shaped function might indeed
best represent the relationship between firm survival and networking.
Watson (2007) notes that the probability of firm survival peaks when
the SME owner is involved in about six networks beyond this level the
probability of survival declines.
Similarly, Figures 9.2 and 9.3 depict the results of estimating the rela-
tionship between networking and both income growth and return on
equity (ROE), respectively. To assess the relationship between network-
ing and both firm growth and ROE, the analysis focuses on firms in the
upper and lower quartiles for these two performance measures. Firms in
0.58
0.56
0.54
Probability of High Growth
0.52
0.50
0.48
0.46
0.44
0.42 Linear
Quadratic
0.40
–10 0 10 20 30
Networking score
Source: ABS.
Figure 9.2 Probability of high growth fitted against an SME owner’s

networking score
the upper quartile are coded 1 and those in the lower quartile are coded 0.
It should be noted that, unlike the analysis of firm survival, the analysis
for growth and ROE is restricted to only those firms that survived to the
last year of the ABS’s four-year longitudinal study. The results depicted
in Figure 9.2 indicate that the same inverted U-shaped relationship that
applies to survival and networking also applies to growth and networking.
However, as can be seen from Figure 9.3, the same cannot be said for ROE
and networking.
The findings presented in Figures 9.1 and 9.2 suggest that both the sur-
vival and growth of SMEs can be enhanced by owners being involved, up
to a limit, in a range of networks. However, the same relationship does
not appear to exist between ROE and networking. It seems that the costs
involved with networking (particularly in terms of the SME owner’s time)
might have a negative impact on overall firm profitability. This issue will
be examined further in the following analysis.
Given the findings above, the SME owners’ networking scores were
entered into logistic regression models as both first and second order vari-
ables. Table 9.2 provides the results of examining the relationship between
108 SME performance
0.54
0.52
Probability of High ROE
0.50
0.48
0.46
0.44
Linear
Quadratic
0.42
–10 0 10 20 30
Networking score
Source: ABS.
Figure 9.3 Probability of high ROE fitted against an SME owner’s

networking score
firm performance (survival, growth and ROE) and the level of networking
activity undertaken by SME owners. In the first model, only the demo-
graphic variables (age, industry and size) are included. In the second
model, networking is added. As can be seen from the table, the first order
networking variable is significantly positively related to the probability of
firm survival and, to a lesser extent, growth, but not ROE. The results also
indicate that the second order networking variable is significantly nega-
tively associated with survival and growth, but not ROE. These results
add further support to the proposition that the relationship between SME
performance and networking resembles an inverted U-shaped function
for both survival and growth, but not ROE (where there was simply no
significant relationship between networking and performance). Perhaps
the additional revenues gained through networking help the firm survive
and grow but any additional profit earned is offset by the additional costs
involved with networking (both time and financial).
In Table 9.3, the overall networking score variable is replaced by its
various constituent parts. First, the overall networking score is broken
down into a formal and informal network variable (model 3) and, second,
Table 9.2 Logistic regression models of survival, growth and ROE against networking
Variables Survival Growth ROE

Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
B Exp(B) B Exp(B) B Exp(B) B Exp(B) B Exp(B) B Exp(B)
Firm age ** ** ** ** * *
Less than 2 years −3.21 0.04** −3.34 0.04** 0.96 2.62** 0.97 2.63** 0.47 1.60* 0.47 1.59*
old
2 years to less −0.43 0.65* −0.44 0.64* 0.41 1.50* 0.40 1.50* 0.40 1.49* 0.39 1.47*
than 5
5 years to less than −0.35 0.70* −0.28 0.76 0.28 1.33* 0.28 1.32* 0.34 1.40* 0.33 1.39*
109
10
10 years to less 0.05 1.05 0.10 1.10 0.22 1.24 0.23 1.26 0.37 1.45** 0.36 1.43*
than 20
Industry * * ** **
Mining −0.24 0.79 −0.20 0.82 −0.47 0.63 −0.43 0.65 −1.02 0.36 −1.06 0.35*
Manufacturing 0.02 1.02 −0.04 0.96 −0.25 0.78 −0.28 0.76 −0.70 0.50* −0.70 0.50*
Construction 0.28 1.33 0.14 1.15 0.45 1.57 0.44 1.55 −0.15 0.86 −0.14 0.87
Wholesale trade 0.45 1.56 0.38 1.46 −0.18 0.83 −0.21 0.81 −0.56 0.57 −0.55 0.58
Retail trade 0.11 1.12 0.06 1.06 −0.22 0.81 −0.23 0.80 −0.63 0.53* −0.64 0.53*
Accom., cafes, −0.10 0.90 −0.11 0.90 −0.66 0.52 −0.66 0.52 −0.69 0.50 −0.71 0.49*
rest’s
Table 9.2 (continued)

Transport & 0.24 1.27 0.20 1.22 −0.24 0.78 −0.24 0.78 −0.95 0.39** −0.96 0.38**
storage
Finance & −0.17 0.85 −0.22 0.80 −0.10 0.90 −0.11 0.90 −0.51 0.60 −0.51 0.60
insurance
Property & bus. 0.23 1.25 0.16 1.17 0.03 1.03 0.03 1.03 −0.32 0.73 −0.32 0.72
serv.
Cultural & rec. 0.16 1.17 0.19 1.20 −0.19 0.82 −0.18 0.84 −1.20 0.30** −1.22 0.30**
110
serv.
Firm size 0.00 1.00 0.00 1.00 0.00 1.00 0.00 1.00 0.00 1.00 0.00 1.00
Networking 0.37 1.45** 0.10 1.10** −0.02 0.98
Networking2 −0.02 0.98** −0.01 1.00* 0.00 1.00
Nagelkerke R Square 0.327 0.392 0.032 0.039 0.024 0.027
Notes:
* Significant at 5%; ** Significant at 1%.
In running the logistic regression, the last category was used as the reference point for each categorical variable and, therefore, the last category is
not shown in the table. For firm age the last category is ‘20 years or older’ and for industry the last category is ‘Personal and other services’.

Table 9.3 Logistic regression models of survival, growth and ROE against formal and informal networks and network
range and intensity

Firm age ** ** ** ** * *
Less than 2 −3.33 0.04** −3.33 0.04** 0.98 2.67** 0.97 2.64** 0.46 1.59* 0.47 1.60*
years old
2 years to less −0.43 0.65* −0.46 0.63* 0.43 1.54** 0.40 1.49** 0.38 1.47* 0.40 1.48*
111
than 5
5 years to less −0.27 0.76 −0.26 0.77 0.30 1.35* 0.27 1.31* 0.33 1.39* 0.33 1.39*
than 10
10 years to less 0.09 1.09 0.07 1.08 0.23 1.25 0.22 1.25 0.36 1.43* 0.36 1.43*
than 20
Industry ** ** ** **
Mining −0.25 0.78 −0.22 0.81 −0.46 0.63 −0.42 0.66 −1.06 0.35* −1.06 0.35*
Manufacturing −0.08 0.92 −0.03 0.97 −0.29 0.75 −0.29 0.75 −0.70 0.50* −0.69 0.50*
Construction 0.11 1.12 0.06 1.07 0.45 1.56 0.43 1.54 −0.15 0.86 −0.13 0.88
Wholesale 0.33 1.39 0.40 1.50 −0.23 0.80 −0.23 0.80 −0.55 0.58 −0.56 0.57
trade

Retail trade 0.04 1.04 0.06 1.06 −0.22 0.81 −0.24 0.79 −0.64 0.53* −0.63 0.53*
Accom., cafes, −0.13 0.88 −0.18 0.83 −0.65 0.52 −0.68 0.51 −0.71 0.49 −0.71 0.49
rest’s
Transport & 0.18 1.20 0.20 1.22 −0.25 0.78 −0.25 0.78 −0.96 0.38* −0.97 0.38**
112
storage
Finance & −0.24 0.79 −0.25 0.78 −0.11 0.90 −0.11 0.90 −0.51 0.60 −0.52 0.60
insurance
Property & 0.11 1.11 0.10 1.10 0.02 1.02 0.02 1.02 −0.32 0.73 −0.33 0.72
bus. serv.
Cultural & rec. 0.19 1.21 0.23 1.26 −0.15 0.86 −0.17 0.84 −1.22 0.29** −1.23 0.29**
serv.
Firm size 0.00 1.00 0.00 1.00 0.00 1.00 0.00 1.00 0.00 1.00 0.00 1.00
Formal networks 0.46 1.59** 0.13 1.14** −0.03 0.98
Formal networks2 −0.03 0.97** −0.01 0.99* 0.00 1.00
Informal 0.19 1.20* −0.07 0.94 0.01 1.01
networks
Informal −0.02 0.98 0.01 1.01 −0.01 1.00
networks2
Network range 0.04 1.04 0.23 1.26** −0.06 0.95
Network range2 0.00 1.00 −0.02 0.98* 0.00 1.00
Network intensity 2.19 8.92** −0.25 0.78 −0.07 0.93
Network intensity2 −0.65 0.52** 0.09 1.09 0.06 1.06
113
Nagelkerke R 0.392 0.406 0.041 0.041 0.027 0.027
Square
Notes:
* Significant at 5%; ** Significant at 1%.
In running the logistic regression, the last category was used as the reference point for each categorical variable and, therefore, the last category is
not shown in the table. For firm age the last category is ‘20 years or older’ and for industry the last category is ‘Personal and other services’.

114 SME performance
into a network range and intensity variable (model 4). Consistent with
expectations, the results in Table 9.3 indicate that a firm’s survival and
growth (but not ROE) are more strongly associated with an owner’s
involvement in formal rather than informal networks. This finding sup-
ports the argument that weak ties are likely to be more important than
strong ties in the dissemination of information and, therefore, firm
performance (Granovetter 1983). Also, as expected, firm survival was
significantly associated with network intensity (but not network range),
while firm growth was significantly associated with network range (but not
network intensity).
The results presented in Table 9.3 indicate that SME owners need to be
strategic in terms of the nature of their networking involvement. If growth
is of paramount concern, SME owners would be well advised to consider
developing a broad range of networks, although the significance of the
second order network range variable suggests that there is a limit beyond
which further networking involvement is likely to be counterproductive.
Alternatively, if survival is of paramount concern (as might be the case
early in the life of a new venture), SME owners would be well advised
to develop closer ties with a smaller range of networks. Again, however,
SME owners should carefully monitor the time and cost associated with
networking because the results indicate that very high levels of network
intensity can be counterproductive and are unlikely to benefit overall
profitability (ROE).
9.3 SUMMARY
The findings reported in this chapter indicate that (after allowing for
age, industry and size of business) networking appears to be significantly
positively associated with firm survival and, to a lesser extent, growth
(consistent with the results of Brüderl and Preisendörfer 1998 for newly
established firms). This finding confirms the importance of social capital
in providing SME owners with information critical to the success of their
ventures. However, there appears to be no significant association between
networking and ROE (profitability). Further, the findings with respect
to both survival and growth suggest that there might be some optimum
level of resources (both time and financial) that an owner should allocate
to networking. For example, accessing more than six networks during a
year is likely to be counterproductive. Similarly, accessing any individual
network on more than three occasions during a year is also likely to be
counterproductive. Therefore, given that business failure generally results
in heavy personal loss (Bannock 1981), owners need to seriously consider
the range and intensity with which they access various potential networks
(formal and informal).
The results also indicate that both formal and informal networks are
associated with firm survival, but that only formal networks are associated
with growth (and neither formal nor informal networks are associated
with ROE). The finding with respect to formal networks highlights the
particular importance of weak ties (Granovetter 1983) in building an SME
owner’s social capital.
Further, the results show that network intensity is more critical to firm
survival than network range. Conversely, network range is more critical to
firm growth than network intensity, again confirming the importance of
weak ties (Granovetter 1983) in disseminating information, and provid-
ing support for the assertion by Fischer and Reuber (2003) that owners of
rapid-growth firms should be interested in (and should support) govern-
ment policy aimed at developing a network-based approach to facilitating
firm growth.
Having explored the relationship between networking and SME per-
formance for all firms, the following chapter will specifically look at
possible networking differences in male- and female-controlled SMEs.
10. Networking: comparing female-
and male-controlled SMEs
10.0 INTRODUCTION
In the previous chapter I explored the relationship between networking

and firm performance. Following Ibarra’s (1992) call for further empirical
evidence to clarify the way men’s and women’s networks differ, the extent
of these differences and the potential consequences of any such differences,
this chapter has three primary objectives: first, to determine whether there
are any systematic networking differences between male and female SME
owners; second, to investigate the association between networking and
firm performance for male- and female-controlled SMEs, separately; and
third, to dispel the myth that female SME owners are disadvantaged as the
result of having fewer network contacts.
10.1 POTENTIAL DIFFERENCES IN THE

NETWORKS OF MALE AND FEMALE SME
OWNERS
Cromie and Birley (1992) argue that because the majority of women enter
self-employment from a domestic and/or non-managerial background it is
likely that their personal network contacts will not be as extensive, or well
developed, as their male counterparts. Similarly, Munch, McPherson and
Smith-Lovin (1997) note that housework and childrearing are extremely
lonely forms of work and this isolation results in many women having
limited network contacts compared to men. Even where women move
directly from paid employment into self-employment, it is likely they will
have fewer network contacts because females typically occupy lower level
positions within the organizations they leave, compared to the typical
male (Cromie and Birley 1992).
Aldrich (1989) also argues that past research, and much of the literature,
indicates that female entrepreneurs might not only have fewer networks
than their male counterparts, but they are likely to be embedded in differ-
ent types of networks. For example, Munch et al. (1997) suggest that as a
116
Networking: female- and male-controlled SMEs 117
result of their childrearing responsibilities, women will typically rearrange

their network composition to favour kin (family and friends) over other
forms of network contacts. Consistent with this argument, Orhan (2001)
notes that past research has found that the first source of advice for male
entrepreneurs is usually professional experts (such as accountants and
lawyers) and second is their spouse, whereas the first source of advice for
female entrepreneurs is their spouse, second their friends, and third pro-
fessional experts. Similarly, Moore (1990) found that women were more
likely than men to include family members in their networks. This suggests
that male SME owners are more likely to access formal networks, while
female SME owners are more likely to access informal networks (particu-
larly family and friends).
Although the literature, and the vast majority of past research, indi-
cates that women are likely to have less well-developed networks than
men, it should be noted that Cromie and Birley (1992) found that female
SME owners were just as active in their networking relationships as their
male counterparts. Cromie and Birley (p.249) suggest that, once in busi-
ness, women might well recognize the need to have appropriate network
contacts and ‘proceed to develop them vigorously’. Further, the material
presented in Chapters 4, 5 and 6 suggest no difference in the performance
of male- and female-controlled SMEs after incorporating appropriate
controls (such as size, industry and risk). If it is accepted that, after incor-
porating appropriate controls, there is no difference in the performance of
male- and female-controlled SMEs then, despite much conjecture, there
might indeed be no significant difference in the networking activities of
male and female SME owners. Alternatively, even if men are making
greater use of networks than women, it is possible that the additional
network involvement by men is not paying off. As noted in the previous
chapter, beyond some optimum level additional networking activities can
be counterproductive, and women might therefore not be significantly
disadvantaged by their lower levels of networking, particularly if their
networking efforts are well targeted.
Of the 5014 firms surveyed by Australian Bureau of Statistics (ABS), as
described in Chapter 2, 1914 firms were excluded from the analysis in this
chapter because either they did not have a single major decision maker, or
the sex of that person was not reported. This left 2919 male-controlled and
181 female-controlled SMEs that could be examined over the three-year
period from 1 July 1995 to 30 June 1998. Table 10.1 shows the number of
network contacts (sources of information) accessed during the year by the
male and female SME owners. The results show that most SME owners
(88% of males and 84% of females) accessed at least one network during the
year, with approximately 50% of all SME owners (52% of males and 46%
118 SME performance
Table 10.1 Number of networks accessed by male and female SME

owners
Number of Male Female

Networks n = 2919 n = 181
Accessed
% Cum % % Cum %
10 3 3 2 2
9 5 8 4 6
8 8 16 5 11
7 12 28 8 19
6 11 39 14 34
5 13 52 12 46
4 12 64 12 58
3 11 75 9 67
2 7 82 11 77
1 6 88 7 84
0 12 100 16 100
Note: Chi-Square test comparing males and females not significant at 5%.
Source: ABS.
of females) accessing five or more networks during the year. This finding
is consistent with Cooper, Woo and Dunkelberg (1989) and Robson and
Bennett (2000), who reported that entrepreneurs sought information
from a variety of different sources. However, the results also indicate no
significant differences between the male and female owners in terms of the
number of networks they accessed during the year. This result appears at
odds with most of the literature on gender and networking but supports
Cromie and Birley’s (1992) finding that the personal contact networks of
women are just as diverse as those of men. Note that a separate analysis of
the sub-set of SMEs that accessed three or fewer networks during the year
also failed to find any gender difference and the same applied to the sub-set
of SMEs that accessed seven or more networks during the year.
Table 10.2 provides a summary of the frequency with which both the
male and female SME owners made contact with a variety of formal and
informal networks during the year. As expected, the male SME owners,
on average, made significantly more frequent contact with formal network
sources, particularly with banks, business consultants, industry associa-
tions and solicitors. Unexpectedly, the female SME owners (on average)
did not make significantly more frequent contact with informal network
sources in total, although they did make significantly more frequent contact
Table 10.2 Frequency of formal and informal network contact for male
and female SME owners (%)
Networks Frequency of Contact

Nil 1–3 times >3 times
Male Female Male Female Male Female
Formal
External accountant 19 20 34 36 47 44
Bank 36 44 36 39 28 18**
Solicitor 41 48 35 40 24 12**
Industry association 57 75 23 15 20 10**
Business consultant 71 82 19 13 10 5**
Tax office 58 65 32 30 10 6
SBDC 84 87 13 12 3 1
Av. formal 52 60 27 26 20 13*
Informal
Others in the industry 44 48 30 26 27 27
Family & friends 63 52 20 23 17 25**
Local businesses 73 75 17 15 10 9
Av. informal 60 58 22 21 18 20
Av. all networks 55 60 26 25 20 16
Note: *, ** Chi-Square test significantly different for males and females at 5% and 1%,
respectively.
Source: ABS.
with family and friends. These findings are consistent with Robson, Jack
and Freel (2008), who reported that male Scottish business owners were
significantly more likely to seek advice from consultants and chambers of
commerce, while female Scottish business owners were significantly more
likely to turn to friends and relatives. Shaw, Lam and Carter (2008) also
reported that female owners were significantly more likely (than male
owners) to identify a family member as their prime network contact.
When the overall frequency of contact with all network sources is
examined, the result is again contrary to expectations as there is no sig-
nificant difference between the male and female SME owners. This result,
although inconsistent with the majority of the literature, again confirms
Cromie and Birley’s (1992) finding that females are just as active in their
networking relationships as men. As noted by Cromie and Birley, once in
120 SME performance
business, women might well proceed to vigorously develop their network

contacts.
Interestingly, Table 10.2 shows that the networking group most often
contacted by both the male and female SME owners (with no significant
difference between the two groups) was external accountants (a formal
network): 47% of males and 44% of females accessed an external account-
ant on more than three occasions during the year. This finding is consist-
ent with Robson and Bennett (2000), who reported that, from the private
sector, accountants are the most widely used source of advice. The result is
also consistent with Robson et al. (2008), who found that accountants were
the most widely used source of advice for both male and female Scottish
business owners (with no significant difference by gender). Similarly, the
male and female SME owners also frequently contacted others in the
industry: 27% of both males and females accessed this informal network
source on more than three occasions during the year. Unlike Birley (1985),
who found that entrepreneurs relied heavily on informal networks but
seldom tapped into formal networks, the results presented in Table 10.2
suggest that Australian SME owners (male and female) make extensive
use of both formal and informal networks.
10.2 RELATIONSHIP BETWEEN NETWORKING

AND SME PERFORMANCE FOR MALE- AND
FEMALE-CONTROLLED SMES
Tables 10.3 and 10.4 present the results of modelling the relationship
between a firm’s networking score and its chances of surviving and achiev-
ing high growth, respectively. Note from Chapter 9 that a firm’s network-
ing score can range from zero (if no networks had been accessed during
the year) to 20 (if all ten networks had been accessed on more than three
occasions during the year). Growth is measured as the percentage increase
in total income over the three-year period being examined. Surviving firms
are coded 1, while discontinued firms are coded 0. In terms of growth, the
analysis presented in this chapter focuses on those firms in the top 25%
(upper quartile – coded 1) compared to those in the bottom 25% (lower
quartile – coded 0).
The results in Table 10.3 indicate a significant positive relationship
between networking and firm survival (and a negative relationship between
age of business and firm survival). Similarly, the results in Table 10.4 indi-
cate a significant positive relationship between networking and firm growth
(with younger businesses also more likely to achieve high growth). Note
that, consistent with previous studies that have incorporated appropriate
Table 10.3 Modelling firm survival and networking
Variables B S.E. Wald df Sig. Exp(B)

Sex of owner 0.01 0.25 0.00 1 0.97 1.01
Education 0.49 3 0.92
School −0.03 0.17 0.03 1 0.87 0.97
Trade 0.09 0.20 0.19 1 0.67 1.09
Non-business degree −0.02 0.19 0.02 1 0.90 0.98
Experience 0.00 0.01 0.10 1 0.76 1.00
Industry 14.67 10 0.14
Mining 0.58 0.80 0.53 1 0.47 1.79
Manufacturing −0.27 0.44 0.38 1 0.54 0.76
Construction 0.05 0.49 0.01 1 0.92 1.05
Wholesale trade 0.23 0.46 0.26 1 0.61 1.26
Retail trade 0.04 0.46 0.01 1 0.94 1.04
Accom., cafes & restaurants −0.35 0.53 0.44 1 0.51 0.70
Transport & storage 0.09 0.54 0.03 1 0.87 1.09
Finance & insurance −0.67 0.50 1.81 1 0.18 0.51
Property & bus. services 0.01 0.45 0.00 1 0.98 1.01
Cultural & rec. services −0.23 0.58 0.16 1 0.69 0.79
Firm age 483.34 4 0.00
Less than 2 years old −3.03 0.21 219.28 1 0.00 0.05
2 years to less than 5 −0.40 0.24 2.80 1 0.09 0.67
5 years to less than 10 −0.12 0.22 0.27 1 0.60 0.89
10 years to less than 20 0.21 0.22 0.91 1 0.34 1.24
Firm size 0.00 0.00 0.88 1 0.35 1.00
Networking score 0.16 0.02 100.85 1 0.00 1.18
Constant 0.19 0.56 0.11 1 0.74 1.21
Percentage predicted correctly
Survived/Discontinued/ 42.7 96 88.4
Overall
Nagelkerke R Square 0.36
education the last category is ‘Tertiary (business)’ for industry the last category is ‘Personal
and other services’; and for firm age the last category is ‘20 years or older’.
Source: ABS.
122 SME performance
Table 10.4 Modelling firm growth and networking

Sex of owner 0.08 0.25 0.11 1 0.74 1.09
Education 0.80 3 0.85

School 0.09 0.16 0.29 1 0.59 1.09
Trade 0.12 0.18 0.45 1 0.50 1.13
Non-business 0.00 0.18 0.00 1 0.98 1.00
degree
Experience 0.00 0.01 0.15 1 0.70 1.00
Industry 19.42 10 0.04

Mining −0.62 0.79 0.62 1 0.43 0.54
Manufacturing −0.70 0.47 2.23 1 0.14 0.50
Construction 0.11 0.50 0.05 1 0.82 1.12
Wholesale trade −0.44 0.48 0.84 1 0.36 0.64
Retail trade −0.66 0.49 1.80 1 0.18 0.52
Accom., cafes & −1.13 0.62 3.37 1 0.07 0.32
restaurants
Transport & −0.90 0.55 2.72 1 0.10 0.41
storage
Finance & −0.44 0.52 0.72 1 0.40 0.65
insurance
Property & bus. −0.43 0.48 0.80 1 0.37 0.65
services
Cultural & rec. −0.40 0.62 0.41 1 0.52 0.67
services
Firm age 16.05 4 0.00

Less than 2 years 1.03 0.26 15.81 1 0.00 2.79
old
2 years to less 0.36 0.20 3.29 1 0.07 1.43
than 5
5 years to less 0.36 0.18 4.02 1 0.05 1.43
than 10
10 years to less 0.26 0.17 2.25 1 0.13 1.29
than 20
Firm size 0.00 0.00 0.00 1 0.97 1.00
Networking score 0.03 0.01 5.96 1 0.02 1.03
Constant −0.17 0.54 0.09 1 0.76 0.85


Percentage predicted
correctly
Low/High 63.2 48.3 55.7
growth/Overall
Nagelkerke R Square 0.04
education the last category is ‘Tertiary (business)’ for industry the last category is ‘Personal
and other services’; and for firm age the last category is ‘20 years or older’.
Source: ABS.
controls, there is no relationship between gender and firm performance

(either survival or growth). Also note that when firms were classified as
high/low growth based on whether their growth rate was above/below the
median result (rather than being based on the upper and lower quartiles),
the findings were qualitatively the same as those reported in Table 10.4,
but the explanatory power of the model (Nagelkerke R Square value) was
reduced.
Finally, Tables 10.5 and 10.6 present the results of separately model-
ling for male- and female-controlled SMEs the relationship between
an owner’s use of various formal and informal networks and both firm
survival and firm growth, respectively. Given the relatively large number
of control variables and potential networking sources, the forward step-
wise (conditional) logistic regression method was used, adopting the
SPSS default cut-off of 5% for variables entering the model and 10% for
removal. To check the robustness of the results, the logistic regressions
were also run backwards, with no significant differences found. Note that
when using stepwise logistic regression, SPSS highlights those variables
that are significant and ‘in the equation’; those variables that are not
significant are therefore not reported in Tables 10.5 and 10.6.
Table 10.5 shows that the only networking source significantly related
to the survival of both male- and female-controlled SMEs is external
accountants (a formal network). Firms that never accessed an external
accountant during the year were significantly less likely to survive com-
pared to those firms that accessed an external accountant on more than
three occasions during the year. Interestingly, there was no advantage to
accessing an external accountant on more than three occasions during the
124 SME performance
Table 10.5 Modelling firm survival and individual network contact for
male- and female-controlled SMEs
Variables in the Final Male-Controlled Female-Controlled

Models
Wald Sig. Exp(B) Wald Sig. Exp(B)
Firm age 434.74 0.00 37.84 0.00
Less than 2 years 212.94 0.00 0.05 14.55 0.00 0.01
old
2 years to less than 3.10 0.08 0.66 0.15 0.70 0.59
5
10
20
Formal networks
External accountant 77.07 0.00 10.18 0.01
Never 68.23 0.00 0.24 9.40 0.00 0.09
1–3 times 2.77 0.10 0.76 0.78 0.38 0.53
Industry association 13.27 0.00
Never 10.86 0.00 0.50
1–3 times 1.49 0.22 0.74
Informal networks
Others in the industry 9.48 0.01
Never 0.06 0.81 0.96
1–3 times 5.42 0.02 1.58
Family and friends 5.52 0.06
Never 0.47 0.49 0.60
1–3 times 2.91 0.09 4.87
Constant 201.84 0.00 32.25 11.24 0.00 77.25
Survived/Discont./ 96.6 39.4 88.5 94.5 66.7 89.0
Overall
Chi-square 0.00 0.00
significance
22 Log likelihood 1716 91
firm age the last category is ‘20 years or older’ and for each network the last category is
‘More than 3 times’.
Source: ABS.
Table 10.6 Modeling firm growth and individual network contact for
male- and female-controlled SMEs
Variables in the Final Models Male-Controlled Female-Controlled

Wald Sig. Exp(B) Wald Sig. Exp(B)
Firm age 17.06 0.00
Less than 2 years old 15.71 0.00 2.78
2 years to less than 5 2.92 0.09 1.39
Industry 21.65 0.02
Mining 0.93 0.33 0.44
Manufacturing 2.47 0.12 0.42
Construction 0.00 0.97 0.98
Wholesale trade 1.55 0.21 0.49
Retail trade 2.62 0.11 0.39
Accom., cafes & 4.03 0.05 0.22
restaurants
Transport & storage 3.26 0.07 0.32
Finance & insurance 1.16 0.28 0.52
Property & bus. services 1.22 0.27 0.53
Cultural & rec. services 0.60 0.44 0.57
External accountant 6.71 0.04 8.10 0.02
Never 6.36 0.01 0.64 3.02 0.08 0.25
1–3 times 0.08 0.78 0.96 7.25 0.01 0.24
Industry association 6.46 0.04
Never 0.09 0.76 0.95
1–3 times 3.07 0.08 1.37
Constant 0.69 0.41 1.62 3.52 0.06 2.00
High/Low/Overall 53.2 63.0 58.1 62.9 71.1 67.1
Chi-square significance 0.00 0.01
22 Log likelihood 1683 92
firm age the last category is ‘20 years or older’; for industry the last category is ‘Personal
and other services, and for each network the last category is ‘More than 3 times’.
Source: ABS.
126 SME performance
year compared to only accessing this source on between one and three
occasions. This finding suggests that there might be some optimal level of
networking with external accountants beyond which there is no additional
benefit to be gained (but nor is there any evidence that more frequent
contact does any harm). The only other formal network that showed up
in the models was industry association, but only for the male-controlled
SMEs. As was the case with external accountants, it would seem that pro-
vided male SME owners access industry associations on between one and
three occasions during the year there is no additional benefit to accessing
this network source more frequently.
The results with respect to the use of informal networks were also quite
interesting, with the males benefiting from networking with others in the
industry and the females from family and friends. In this case, however,
the results strongly suggest that that excessive networking might be coun-
terproductive. For both the male- and female-controlled SMEs, it would
appear that accessing informal networks (others in the industry for males
and family and friends for females) on between one and three occasions
during the year is significantly more likely to be associated with firm
survival than accessing such sources more frequently (or not at all). This
finding suggests that the association between accessing informal networks
and firm survival resembles an inverted U-shaped function for both male
and female SME owners.
In summary, the final model for predicting the survival of male-
controlled SMEs incorporates (along with the age of the business)
both formal networks (external accountants and industry associations)
and informal (others in the industry). Accessing other networks (the
Australian tax office, banks, business consultants, family and friends,
local businesses, the SBDC and solicitors) does not add significantly to the
explanatory power of the model. Similarly, the final model for predicting
the survival of female-controlled SMEs incorporates (along with the age
of the business) both formal networks (external accountants) and informal
(family and friends).
Consistent with Granovetter’s (1973) weak tie theory and Burt’s (1992)
notion of structural holes, for both male- and female-controlled SMEs,
the results show a stronger relationship between survival and formal
network sources than between survival and informal network sources,
although both networking sources are clearly important. This result is
contrary to Brüderl and Preisendörfer’s (1998) finding that strong ties are
more important than weak ties in explaining firm survival. However, the
results support the suggestion by Uzzi (1996) that networks consisting of
a balance of both weak and strong ties might ultimately be more valu-
able than networks that are focused on only weak or only strong ties. It
should also be noted that the model for predicting the survival of female-
controlled SMEs appears to be superior to that for male-controlled SMEs
(in terms of the Nagelkerke R Square value).
Table 10.6 provides the results of undertaking a similar analysis using
sales growth as the dependent variable. In terms of formal networks,
male-controlled high-growth SMEs appear to gain some advantage from
accessing both external accountants and industry associations. However,
the results again indicate, with respect to external accountants, that there
might be some optimum level of networking beyond which there is no
further benefit to be gained and, in the case of industry associations,
excessive networking (more than three times during a year) might be
counterproductive. That is, there is no difference (in terms of firm growth)
between accessing external accountants one to three times during the year
and more often. However, accessing industry associations between one
and three times during the year appears to be significantly more beneficial
than accessing this source more often (or not at all). This suggests that, for
growth-oriented male-controlled SMEs, accessing both external account-
ants and industry associations up to three times during a year might be
an optimal strategy; any further interaction with these formal network
sources is likely to be counterproductive (particularly with respect to
networking with industry associations).
For the female-controlled SMEs, the results indicate that accessing
an external accountant for advice on more than three occasions during
the year is significantly associated with high sales growth compared to
never accessing this source, or only doing so one to three times. Beyond
noting that accessing an external accountant on more than three occasions
during the year appears beneficial, it is not possible to indicate what the
optimum level of networking with external accountants might be for high-
growth female-controlled SMEs. These results suggest that while male
SME owners make effective use of both external accountants and industry
associations, female SME owners tend to rely more heavily on exter-
nal accountants (possibly because of problems associated with access-
ing industry associations which typically meet after hours, or perhaps
because they see little value in accessing this network). Interestingly, no
informal networks (which typically consist of stronger ties and fewer
structural holes) appear to be related to firm growth for either the male- or
female-controlled SMEs.
In summary, the final model for predicting high-growth male-controlled
SMEs incorporates (along with age and industry) two formal networks
(external accountants and industry associations) but no informal net-
works. The final model for predicting high-growth female-controlled
SMEs incorporates only one formal network (external accountants)
128 SME performance
and no informal networks. These findings are again consistent with

Granovetter’s (1973) weak tie theory and Burt’s (1992) notion of struc-
tural holes because each of these theories would predict that SME owners
are likely to derive more benefit (in terms of accessing new products and
markets) from formal, rather than informal, networking sources. The
results are also consistent with Brüderl and Preisendörfer’s (1998) finding
that strong ties were more important to firm survival than to firm growth.
It should once again be noted (as was the case for modelling firm survival)
that the model for predicting high-growth female-controlled SMEs is
superior to that for predicting high-growth male-controlled SMEs (in
terms of the Nagelkerke R Square value).
10.3 SUMMARY
Several interesting observations arise from the results presented in this

chapter. First, while male and female SME owners appear to access a
similar number of networks, male SME owners (as suggested by the
literature) appear to make more frequent use of formal networks (in par-
ticular, banks, solicitors, industry associations and business consultants).
Further, with the exception of the relationship between industry associa-
tions and survival, the formal networks that were accessed significantly
more frequently by male SME owners had no apparent impact on firm
performance. It would appear, therefore, that female-controlled SMEs are
not disadvantaged by their owners devoting fewer resources to networking
with these groups; this finding contrasts with Aldrich’s (1989) suggestion
that differences in network access could have a significant impact on the
performances of female-controlled SMEs.
Second, accessing an external accountant is the only formal network
significantly related to both firm survival and growth for both the male-
and female-controlled SMEs. Therefore, given limited time for network-
ing, it would seem that SME owners would be well advised to ensure that
they maintain regular contact with an external accountant; this would
appear to be particularly relevant for female SME owners. While this
finding is consistent with Potts (1977, p.93), who found that ‘successful
companies rely more heavily on accountants’ information and advice
than do unsuccessful companies’, it contrasts with the results of Robson
and Bennett (2000) and Cooper, Gimeno-Gascon and Woo (1994). The
former found no statistically significant relationship between accessing
advice from accountants and any of their measures of firm performance.
Similarly, Cooper et al. (1994) found that the use of professional advisers
had no significant effect on firm performance.
Third, with respect to informal networks, there does not appear to be

any significant difference in the overall frequency with which male and
female SME owners access these groups, although female owners appear
to make significantly more use of family and friends. Further, while the
evidence suggests that SME owners make frequent contact with a variety
of informal networks, none of these network sources appear to be related
to firm growth and only two appear to be related to firm survival (others in
the industry for male-controlled SMEs, and family and friends for female-
controlled SMEs). The finding that no informal networks were related to
firm growth (for either the male- or female-controlled SMEs) is somewhat
surprising given Fischer and Reuber’s (2003) observation that owners of
high-growth firms see owners of other high-growth firms as an invalu-
able source of relevant and useful advice. However, the finding supports
Nelson’s (1989) argument that owners who want to grow their firms are
best advised to make more frequent use of a limited number of networks
where they can access the particular expertise they require. The finding
also supports the argument that weak ties are more important than strong
ties for business growth and development (Granovetter 1973).
Fourth, there were fewer networks associated with firm growth than
was the case for firm survival. This again suggests that owners seeking
rapid growth for their firms might be best advised to seek more frequent
help from a smaller number of network sources that have the specific
expertise required (Nelson 1989; Zhao and Aram 1995). This result might
also help to explain the finding by Bates (1994, p.671) that the heavy use
of social support networks typified ‘the less profitable, more failure-prone
businesses’. That is, it might be important for SME owners to regularly
assess their networking activities to ensure they are accessing appropriate
networks without devoting too many resources to networking, relative
to the benefits they receive. Through a process of expanding and culling
their networks, entrepreneurs can identify those relationships that merit
‘continued development and future investment’ (Larson and Starr 1993,
p.6).
Fifth, while there are some notable differences between the male- and
female-controlled SMEs in terms of the networking sources that were
significant in the models developed to predict firm performance (survival
and growth), these differences do not appear to negatively impact the per-
formances of female-controlled SMEs relative to their male counterparts.
Indeed, there was no significant gender difference in the performances
(survival or growth) of the male- and female-controlled SMEs. Further,
it should be noted that the models developed to predict firm performance
(survival and growth) appear stronger (in terms of explanatory power –
Nagelkerke R Square) for the female- compared to the male-controlled
130 SME performance
SMEs. This result is consistent with a social feminist theory perspective

(Fischer et al. 1993) in that, although there might be some differences in
the networking activities of male and female SME owners, both groups
appear equally effective in terms of the overall economic benefits they
derive from their networking activities.
Finally, for the relatively few networks that are significantly related to
firm performance, there is some evidence to suggest that excessive net-
working (more than three times during a year) might be counterproduc-
tive. This was particularly true of the association between firm survival
and the use of certain informal networks (others in the industry for male-
controlled SMEs and family and friends for female-controlled SMEs).
In summary, although SME owners appear to access a number of
different networks, few of these networks appear to be associated with
firm performance (survival or growth). The only networks to show up
as being significantly associated with firm performance are: external
accountants (for firm survival and growth, for both male- and female-
controlled SMEs); industry associations (for the survival and growth of
male-controlled SMEs); others in the industry (for the survival of male-
controlled SMEs); and family and friends (for the survival of female-
controlled SMEs).
However, the reader should be cautioned against interpreting the results
presented in this chapter as indicating that networking with those groups
not featured in the various models has no benefit. SME owners might get
other benefits from networking, beyond the purely economic benefits that
were the focus of this chapter. For example, through networking, owners
might draw more comfort (reassurance) about their future plans and
might gain the reassurance needed to continue in difficult times (Birley
1985). Networks can also help SME owners integrate into the social life
of a community (Donckels and Lambrecht 1995). Further, the benefits
from some networking sources might be firm- and/or situation-specific
and might, therefore, not show up in a large-scale study looking at average
outcomes. For example, use of management consultants might be of sub-
stantial benefit in a few very specific cases. An analysis of a large data set
might mask, or make it difficult to detect, these benefits. This is an area
that future research could investigate further.
PART VI
Conclusions
11. Conclusions, implications and
areas for future research
11.0 INTRODUCTION
When I agreed to write this book I had two primary motivations. First,
was to summarize the key findings from the research I have been involved
with over the past 20 years. Second, drawing on those findings and the
work of other scholars, to try and dispel a number of myths that have been
allowed to perpetuate in ‘the absence of good statistical evidence’ (Scott
and Lewis 1984, p.49). Without such evidence, there is the risk that these
myths will get ‘reported by the media, perpetuated by spokespeople for
the industry and subsequently accepted by the wider public’ (Stanworth
1995, p.59). Further, policy decisions by governments and others with an
interest in SMEs are likely to be suspect if they are based on such misper-
ceptions. For instance, the assumed high risk of failure within the SME
sector has been cited as justification for the high rates of return demanded
from this sector by bankers and venture capitalists (Phillips and Kirchoff
1989).
11.1 SUMMARY OF KEY FINDINGS
Having provided a brief introduction to the book in Part I (Chapter 1),

Part II (Chapters 2 and 3) then focused on SME performance. Chapter
2 examined five definitions of failure that have been suggested or used by
SME researchers. The definitions include: bankruptcy; discontinuance
(sale or closure) of a business; business closure (that is, excluding busi-
nesses that are sold); termination of a business to prevent further losses;
and failure to ‘make a go of it’. While each of these definitions might have
appealing attributes, no one definition stands out as being clearly superior.
It should also be noted that different users might be interested in different
measures of SME performance. For example, banks might be interested
in the rate of bankruptcies in the SME sector. SME owners, on the other
hand, might be more concerned with the proportion of businesses that are
closed or sold because the owners failed to ‘make a go of it’.
133
134 SME performance
On balance, it would appear that business closure might provide the

most appropriate indicator of the rate of SME failure. However, it should
be noted that both Headd (2003) and Bates (2005), in relatively recent
studies, reported that about a third of SME owners considered their busi-
nesses to be successful at the time of closure. In many of these cases the
owners were simply retiring or had found ‘a superior alternative’ (Bates
2005, p.344). Therefore, researchers and others with an interest in reported
failure rates need to be mindful of the limitations inherent in the various
SME failure definitions found in the literature.
Ideally, SME performance should be judged on the same basis as is
applied to large businesses, namely, the financial return provided to
owners. However, there are two difficulties with this proposition. First, it
is very difficult to obtain this type of information for SMEs, as they are not
typically required to publically provide it.
Second, unlike shareholders in large corporations, SME owners can
derive utility from their firms beyond the purely financial returns available
to shareholders in large corporations. Murphy, Trailer and Hill (1996)
suggest that one approach to measuring business effectiveness is to relate
performance to organizational goals. Such an approach would seem par-
ticularly appropriate for SMEs, where the goals of the organization and
the owner are generally one and the same. This is undoubtedly the view
taken by Birley and Westhead (1990), Brush (1992) and Shuman (1975),
and is explicitly acknowledged by Bhide (1996, p.122) who stated that:
[a]n entrepreneur’s personal and business goals are inextricably linked.

Whereas the manager of a public company has a fiduciary responsibility to
maximize value for shareholders, entrepreneurs build their businesses to fulfil
personal goals.
Further, the SME literature suggests that the goals and expectations of
owner-operators impact on how they evaluate their firm’s performance.
For example, Buttner and Moore (1997, p.34) discovered that female
small-business owners measure success in terms of ‘self-fulfilment and
goal achievement. Profits and business growth, while important, were less
substantial measures of their success.’
We need to also recognize that each SME owner is likely to have a unique
set of goals related to his/her individual situation (Naffziger, Hornsby and
Kuratko 1994) and, consequently, it could be argued that the perform-
ance of an SME can only be appropriately assessed based on the extent to
which those specific goals have been (are being) met (Murphy et al. 1996).
This clearly is an area that future research could usefully explore.
Having examined the various definitions of SME failure found in the
literature and noted that reported failure rates vary considerably across
Conclusions 135
studies, even where the same failure definition is used, Chapter 3 then con-
sidered some of the more likely reasons for such variations. In particular,
the relationship between the various failure definitions and firm age, size,
industry and the state of the economy were examined.
In terms of age, the available evidence presents a compelling case to
suggest that failure rates peak at around three years of age, irrespective of
the failure definition being used. It is important, therefore, when compar-
ing and analysing SME failure rates, that researchers control for the age
of the business.
In terms of the size of the business, the results indicate that some failure
definitions can be biased either for or against larger or smaller businesses.
In particular, bankruptcy appears to be biased against larger businesses,
while discontinuance is biased against smaller firms. This suggests that,
when comparing and analysing SME failure rates, researchers should
control for both age and size of business and should make clear any limi-
tations associated with the failure definition being used. The fact that the
majority of past studies have used discontinuance (which is biased against
SMEs) as their definition of failure is the most likely reason why the myth
that SMEs have unacceptably high failure rates (compared to large busi-
nesses) has become established as part of the folklore on this subject.
Similarly, it would appear that some failure definitions are likely to be
biased either for or against certain industry sectors. In particular, indus-
tries with significant start-up costs are likely to report higher bankruptcy
rates but lower discontinuance rates. Conversely, industries with relatively
small start-up costs are likely to report lower bankruptcy rates but higher
discontinuance rates.
Finally, in terms of the effect of macro-economic factors on the rate of
SME failure, the evidence again suggests some potentially confounding
signals depending on the definition of failure being used. For example, and
as would be expected, the rate of SME bankruptcies appears to be posi-
tively related to interest rates. However, for all other failure definitions,
the evidence suggests that improvements in the economy can provide the
trigger for SME owners to move out of self-employment (resulting in an
apparent increase in failure rates). This indicates that policy makers need
to exercise considerable care, for example, in assessing the impact (on
SME failure rates) of any measures introduced to stimulate the economy
as a means of promoting business, particularly in the SME sector.
Part III was devoted to a comparison of male- and female-controlled
SMEs in terms of: failure rates (Chapter 4); various return measures
(Chapter 5); and a risk-adjusted measure (Chapter 6). In each case (and
contrary to popular belief), after controlling for key demographic differ-
ences (such as age and industry), there was no evidence to suggest that
136 SME performance
female-controlled SMEs underperform male-controlled SMEs. I believe

past studies that have found female-controlled SMEs underperform
male-controlled SMEs have, typically, not measured performance appro-
priately and/or have not controlled for key demographic differences. As a
result, most of the previous research in this area is incomplete and prob-
ably biased against female-owned SMEs, which tend to be both younger
and smaller than male-owned SMEs.
In Part IV of the book I examined the issue of financing for SMEs. The
available literature suggests a strong link between the availability of finance
and SME growth, and this has led to the notion of a ‘finance gap’, implying
that ‘there may be major “barriers” preventing an owner-manager’s access
to equity’ (Hutchinson 1995, p.231). This notion of a ‘finance gap’ within
the SME sector has been supported by a number of researchers and it has
also been suggested that the ‘barriers’ to finance might be even more acute
for female-owned SMEs, as there is a perception that financial institutions
(banks) discriminate against female business owners. While there is no
doubting that firms need finance to grow, it is also the case that not all firms
have the capacity, or desire, to grow. Simply looking at the relationship
between growth/no growth SMEs and their levels of external funding is likely
to confuse cause and effect. A study of firm growth and external funding
might well show a strong positive relationship between these two variables.
However, it is not possible to conclude from such a study that firms without
significant levels of external funding have both the capacity and desire to
grow and that it is only a lack of funding that is holding them back.
Based on both a qualitative (Chapter 7) and quantitative analysis
(Chapter 8), the findings I report indicate that individual owner prefer-
ences, rather than bank discrimination, might be the primary cause of any
observed differences in the level of external funding between firms, and
particularly between male- and female-controlled SMEs. Because female
SME owners are, on average, more risk averse and have a greater need
to feel in control of their businesses, they will be less inclined to access
external funding unless it is absolutely essential. The fact that the rela-
tively lower levels of external funding in female-controlled SMEs are most
noticeable in older firms (with established track records) is consistent with
the proposition that lower levels of external funding in female-controlled
SMEs are the result of personal choice rather than bank discrimination.
This finding is consistent with Hamilton and Fox’s (1998) conclusion that
debt levels in small firms reflect demand-side decisions and are not just the
result of supply-side deficiencies. Hamilton and Fox argue that managerial
beliefs and desires play an important role in determining the capital struc-
ture of SMEs and that a deeper appreciation of these issues will lead to a
better understanding of the capital structure policies of individual SMEs.
Conclusions 137
Further, contrary to some prior research, the findings I present indicate

that growth is not significantly associated with a firm’s relative level of
external (bank) funding but, instead, is associated with a number of other
firm-level variables, in particular, firm profitability. The findings presented
in Chapter 8 also suggest that there is no significant difference in the
overall growth rates for female- and male-controlled SMEs in Australia.
Part V of the book then examined the importance of networking to
SME performance (Chapter 9) and whether there were substantial dif-
ferences in the networking activities of male and female SME owners
(Chapter 10). A number of important conclusions flowed from the analysis
presented in these two chapters. First, while male and female SME owners
appear to access a similar number of networks, male SME owners (as
suggested by the literature) appear to make more frequent use of formal
networks (in particular, banks, solicitors, industry associations and busi-
ness consultants). However, with the exception of the relationship between
industry associations and survival, the formal networks that were accessed
significantly more frequently by male (compared to female) SME owners
(banks, solicitors and business consultants) had no apparent impact on
firm performance. It would appear, therefore, that female-controlled
SMEs are not disadvantaged by their owners devoting fewer resources to
networking with these groups; this finding contrasts with Aldrich’s (1989)
suggestion that differences in network access could have a significant
impact on the performances of female-controlled SMEs.
Second, accessing an external accountant (which male and female
owners appear equally likely to do) is the only formal network signifi-
cantly related to both firm survival and growth, for both the male- and
female-controlled SMEs. Therefore, given limited time for networking, it
would seem that SME owners would be well advised to ensure that they
maintain regular contact with an external accountant; this would appear
to be particularly relevant for female SME owners. While this finding is
consistent with Potts (1977, p.93), who found that ‘successful companies
rely more heavily on accountants’ information and advice than do unsuc-
cessful companies’, it contrasts with the results of Robson and Bennett
(2000) and Cooper et al. (1994). The former found no statistically signifi-
cant relationship between accessing advice from accountants and any of
their measures of firm performance. Similarly, Cooper et al. (1994) found
that the use of professional advisers had no significant effect on firm
performance.
Third, with respect to informal networks, there does not appear to be
any significant difference in the overall frequency with which male and
female SME owners access these groups; although the female owners did
access family and friends significantly more often than the male owners.
138 SME performance
Further, while SME owners appear to make frequent contact with a

variety of informal networks, none of these network sources appear to be
related to firm growth and only two appear to be related to firm survival
(others in the industry for male-controlled SMEs, and family and friends
for female-controlled SMEs). This finding supports Nelson’s (1989)
argument that owners who want to grow their firms are best advised to
make more frequent use of a limited number of networks where they can
access the particular expertise they require. The finding also supports the
argument that weak ties are more important than strong ties for business
growth and development (Granovetter 1973).
Fourth, there were fewer networks associated with firm growth than
was the case for firm survival. This again suggests that owners seeking
rapid growth for their firms might be best advised to seek more frequent
help from a smaller number of network sources that have the specific
expertise required (Nelson 1989; Zhao and Aram 1995). This result might
also help explain the finding by Bates (1994, p.671) that the heavy use of
social support networks typified ‘the less profitable, more failure-prone
businesses’. That is, it might be important for SME owners to regularly
assess their networking activities to ensure they are accessing appropriate
networks without devoting too many resources to networking, relative
to the benefits they receive. Through a process of expanding and culling
their networks, SME owners can identify those relationships that merit
‘continued development and future investment’ (Larson and Starr 1993,
p.6).
Fifth, while there are some notable differences between the male- and
female-controlled SMEs in terms of the networking sources that were
significant in the models developed to predict firm performance (survival
and growth), these differences do not appear to negatively impact the per-
formances of female-controlled SMEs relative to their male counterparts.
Indeed, there was no significant gender difference in the performances
(survival or growth) of the male- and female-controlled SMEs. Further,
it should be noted that the models developed to predict firm performance
(survival and growth) appear stronger (in terms of explanatory power –
Nagelkerke R Square) for the female- compared to the male-controlled
SMEs. This result is consistent with a social feminist theory perspective
(Fischer et al. 1993) in that, although there might be some notable differ-
ences in the networking activities of male and female SME owners, both
groups appear equally effective in terms of the overall economic benefits
they derive from their networking activities.
Finally, for the relatively few networks that are significantly related to
firm performance, there is some evidence to suggest that excessive network-
ing (more than three times during a year) might be counterproductive. This
Conclusions 139
was particularly true of the association between firm survival and the use
of certain informal networks (others in the industry for male-controlled
SMEs and family and friends for female-controlled SMEs).
In summary, although SME owners appear to access a number of dif-
ferent networks, few of these networks appear to be associated with firm
performance (survival or growth). The only networks to show up as being
significantly associated with firm performance are: external accountants
(for firm survival and growth, for both male- and female-controlled SMEs);
industry associations (for the survival and growth of male-controlled
SMEs); others in the industry (for the survival of male-controlled SMEs);
and family and friends (for the survival of female-controlled SMEs).
11.2 FUTURE RESEARCH
In terms of future research into SME performance, I believe much more

is needed in terms of understanding the motivations of individual SME
owners. For example, why do some owners take excessive risks, while
others might be overly cautious? Similarly, why are some owners driven by
a desire to grow their businesses rapidly while others are content to enjoy
their current lifestyle? As noted earlier, we need to recognize that each SME
owner is likely to have a unique set of goals related to his or her individual
situation (Naffziger et al. 1994) and, consequently, these goals need to be
understood before the performance of an individual SME can be appro-
priately assessed. Then, for the sub-set of SME owners who do seek rapid
growth, for example, research is needed to identify the major impediments
they face and the role (if any) that governments could or should play?
11.3 CONCLUSION
I trust that the material I have presented in this book will help dispel a
number of myths relating to SME performance that have been allowed
to perpetuate in the absence of well-constructed research. In particular,
I hope I have been able to convince the reader that: SMEs do not suffer
from excessive failure rates; female-owned SMEs do not underperform
male-owned SMEs (when appropriate performance measures are adopted
and key control variables are incorporated into the analysis); female SME
owners do not find it more difficult than male SME owners to access exter-
nal funding; SME growth is not limited by a lack of external funding; and
female SME owners are not disadvantaged, relative to male SME owners,
in terms of their networking activities.
References
Aldrich, H. (1989). Networking Among Women Entrepreneurs, in
Women-Owned Businesses. Ed. O. Hagan, C. Rivchun and D. Sexton,
New York: Praeger.
Aldrich, H. E. and Reese, P. R. (1993). Does Networking Pay Off? A
Panel Study of Entrepreneurs in the Research Triangle, in Frontiers
of Entrepreneurship Research 1993. Ed. N. C. Churchill, S. Birley, J.
Doutriaux, E. J. Gatewood, F. S. Hoy and W. E. Wetzal, Babson Park,
MA: Wellesley. 325–339.
Ang, J. S. (1992). On the Theory of Finance for Privately Held Firms. The
Journal of Small Business Finance, 1(3), 185–203.
Anna, A. L., Chandler, G. N., Jansen, E. and Mero, N. P. (2000). Women
Business Owners in Traditional and Non-Traditional Industries. Journal
of Business Venturing, 15(3), 279–303.
Baldwin, J. R. and Gorecki, P. K. (1991). Firm Entry and Exit in
the Canadian Manufacturing Sector, 1970–1982. Canadian Journal of
Economics, 24(2), 300–323.
Ballantine, J. W., Cleveland, F. W. and Koeller, T. C. (1993). Profitability,
Uncertainty, and Firm Size. Small Business Economics, 5(2), 87–100.
Bannock, G. (1981). The Economics of Small Firms: Return from the
Wilderness. Oxford, UK: Basil Blackwell.
Barber, B. M. and Odean, T. (2001). Boys Will Be Boys: Gender,
Overconfidence, and Common Stock Investment. The Quarterly Journal
of Economics, 116(1), 261–292.
Barton, S. L. and Matthews, C. H. (1989). Small Firm Financing:
Implications From a Strategic Management Perspective. Journal of
Small Business Management, 27(1), 1–7.
Bates, T. (1994). Social Resources Generated by Group Support Networks
May Not Be Beneficial to Asian Immigrant-Owned Small Businesses.
Social Forces, 72(3), 671–689.
Bates, T. (1995). Analysis of Survival Rates Among Franchise and
Independent Small Business Startups. Journal of Small Business
Management, 33(2), 26–36.
Bates, T. (2005). Analysis of Young, Small Firms That Have Closed:
Delineating Successful From Unsuccessful Closures. Journal of Business
Venturing, 20(3), 343–358.
140
References 141
Bates, T. and Nucci, A. (1989). An Analysis of Small Business Size and

Rate of Discontinuance. Journal of Small Business Management, 27(4),
1–7.
Becchetti, L. and Trovato, G. (2002). The Determinants of Growth
for Small and Medium Sized Firms: The Role of the Availability of
External Finance. Small Business Economics, 19(4), 291–306.
Berger, A. N. and Udell, G. F. (1998). The Economics of Small Business
Finance: The Roles of Private Equity and Debt Markets in the
Financial Growth Cycle. Journal of Banking and Finance, 22(6–8),
613–673.
Berger, A. N. and Udell, G. F. (2002). Small Business Credit Availability
and Relationship Lending: The Importance of Bank Organizational
Structure. The Economic Journal, 112(477), F32–F55.
Berggren, B., Olofsson, C. and Silver, L. (2000). Control Aversion and
the Search for External Financing in Swedish SMEs. Small Business
Economics, 15(3), 233–242.
Berryman, J. (1983). Small Business Failure and Bankruptcy: A Survey of
the Literature. European Small Business Journal, 1(4), 47–59.
Bhattacharjee, A., Higson, C., Holly, S. and Kattuman, P. (2009).
Macroeconomic Instability and Business Exit: Determinants of Failures
and Acquisitions of UK Firms. Economica, 76(301), 108–131.
Bhide, A. (1996). The Questions Every Entrepreneur Must Answer.
Harvard Business Review, 74(6), 120–130.
Birley, S. (1985). The Role of Networks in the Entrepreneurial Process.
Journal of Business Venturing, 1(1), 107–117.
Birley, S. (1986). The Role of New Firms: Births, Deaths and Job
Generation. Strategic Management Journal, 7(4), 361–376.
Birley, S. (1989). Female Entrepreneurs: Are They Really Any Different.
Journal of Small Business Management, 27(1), 32–37.
Birley, S. and Westhead, P. (1990). Growth and Performance Contrasts
Between ‘Types’ of Small Firms. Strategic Management Journal, 11(7),
535–557.
Boden, R. J. and Nucci, A. R. (2000). On the Survival Prospects of Men’s
and Women’s New Business Ventures. Journal of Business Venturing,
15(4), 347–362.
Box, M. (2008). The Death of Firms: Exploring the Effects of
Environment and Birth Cohort on Firm Survival in Sweden. Small
Business Economics, 31(4), 379–393.
Breen, J., Calvert, C. and Oliver, J. (1995). Female Entrepreneurs in
Australia: An Investigation of Financial and Family Issues. Journal of
Enterprising Culture, 3(4), 445–461.
Brüderl, J. and Preisendörfer, P. (1998). Network Support and the
142 SME performance
Success of Newly Founded Businesses. Small Business, Economics,

10(3), 213–225.
Brüderl, J., Preisendörfer, P. and Ziegler, R. (1992). Survival Chances of
Newly Founded Business Organizations. American Sociological Review,
57(April), 227–242.
Brüderl, J. and Schussler, R. (1990). Organizational Mortality: The Liability
of Newness and Adolescence. Administrative Science Quarterly, 35(3),
530–547.
Bruno, A. V. and Leidecker, J. K. (1988). Causes of New Venture Failure:
1960s vs. 1980s. Business Horizons, 31(6), 51–56.
Bruno, A. V. and Tyebjee, T. T. (1985). The Entrepreneur’s Search for
Capital. Journal of Business Venturing, 1(1), 61–74.
Brush, C. G. (1992). Research of Women Business Owners: Past Trends,
a New Perspective, Future Directions. Entrepreneurship Theory and
Practice, 16(4), 5–30.
Brush, C. G., Carter, N., Gatewood, E., Greene, P. G. and Hart, M.
M. (2001). An Investigation of Women-Led Firms and Venture Capital
Investment: A Report for the U.S. Small Business Administration, Office
of Advocacy, and the National Women’s Business Council, Washington,
DC: CB Associates.
Brush, C. G. and Chaganti, R. (1999). Businesses Without Glamour?
An Analysis of Resources on Performance by Size and Age in Small
Service and Retail Firms. Journal of Business Venturing, 14(3),
233–257.
Brush, C. G. and Hisrich, R. D. (1991). Antecedent Influences On Women-
Owned Businesses. Journal of Managerial Psychology, 6(2), 9–16.
Burt, R. S. (1992). Structural Holes. Cambridge, MA: Harvard University
Press.
Buttner, E. H. and Moore, D. P. (1997). Women’s Organizational Exodus
to Entrepreneurship: Self-Reported Motivations and Correlates with
Success. Journal of Small Business Management, 35(1), 34–46.
Cahill, E. (1980). Company Failure and the Vigilant Accountant.
Accountancy, 91(1045), 63–65.
Carpenter, R. E. and Petersen, B. C. (2002). Is the Growth of Small
Firms Constrained by Internal Finance? The Review of Economics and
Statistics, 84(2), 298–309.
Carter, N. M. and Allen, K. R. (1997). Size-Determinants of Women-
Owned Businesses: Choice or Barriers to Resources. Entrepreneurship
and Regional Development, 9(3), 211–220.
Carter, N. M., Brush, C. G., Gatewood, E. J., Greene, P. G. and Hart,
M. M. (2003). Financing High-Growth Enterprise: Is Gender an
Issue? In Critical Junctures in Women’s Economic Lives: A Collection
References 143
of Symposium Papers, Minneapolis, MN: The Centre for Economic

Progress. 45–51.
Carter, N. M., Williams, M. and Reynolds, P. D. (1997). Discontinuance
Among New Firms in Retail: The Influence of Initial Resources,
Strategy, and Gender. Journal of Business Venturing, 12(2), 125–145.
Chaganti, R., DeCarolis, D. and Deeds, D. (1996). Predictors of Capital
Structure in Small Ventures. Entrepreneurship Theory and Practice,
20(2), 7–18.
Chell, E. and Baines, S. (1998). Does Gender Affect Business
‘Performance’? A Study of Microbusinesses in Business Services in the
UK. Entrepreneurship and Regional Development, 10(2), 117–135.
Churchill, B. C. (1952). Survival Patterns of the Postwar Business
Population. Survey of Current Business, 32(12), 12–19.
Cliff, J. E. (1998). Does One Size Fit All? Exploring the Relationship
Between Attitudes Towards Growth, Gender, and Business Size. Journal
Cochran, A. B. (1981). Small Business Mortality Rates: A Review of The
Literature. Journal of Small Business Management, 19(4), 50–59.
Cohen, G. A. (1987). Accountants and Small Business: A Disturbing
Picture. The Chartered Accountant in Australia, 58(5), 6.
Coleman, J. S. (1988). Social Capital in the Creation of Human Capital.
The American Journal of Sociology, 94(Supplement), S95–S120.
Cooper, A. C. (1993). Challenges in Predicting New Firm Performance.
Cooper, A. C., Dunkelberg, W. C. and Woo, C. Y. (1988). Survival and
Failure: A Longitudinal Study, Frontiers of Entrepreneurship Research,
Wellesley, MA: Babson College.
Cooper, A. C., Gimeno-Gascon, J. F. and Woo, C. (1994). Initial Human
and Financial Capital as Predictors of New Venture Performance.
Cooper, A. C., Woo, C. Y. and Dunkelberg, W. C. (1989). Entrepreneurship
and the Initial Size of Firms. Journal of Business Venturing, 4(5), 317–332.
Cressy, R. (1995). Business Borrowing and Control: A Theory of
Entrepreneurial Types. Small Business Economics, 7(4), 291–300.
Cressy, R. (1996). Are Business Startups Debt-Rationed? The Economic
Journal, 106(438), 1253–1270.
Cressy, R. (2006). Why do Most Firms Die Young? Small Business
Economics, 26(2), 103–116.
Cromie, S. (1991). The Problems Experienced by Young Firms.
International Small Business Journal, 9(3), 43–61.
Cromie, S. and Birley, S. (1992). Networking by Female Business Owners
in Northern Ireland. Journal of Business Venturing, 7(3), 237–251.
144 SME performance
Dekimpe, M. G. and Morrison, D. G. (1991). A Modeling Framework for

Analysing Retail Store Durations. Journal of Retailing, 67(1), 68–92.
Delmar, F., Davidsson, P. and Gartner, W. B. (2003). Arriving at the
High-Growth Firm. Journal of Business Venturing, 18(2), 189–216.
Dewaelheyns, N. and Van Hulle, C. (2008). Legal Reform and Aggregate
Small and Micro Business Bankruptcy Rates: Evidence from the 1997
Belgian Bankruptcy Code. Small Business Economics, 31(4), 409–424.
Donckels, R. and Lambrecht, J. (1995). Networks and Small Business
Growth: An Explanatory Model. Small Business Economics, 7(4),
273–289.
Du Rietz, A. and Henrekson, M. (2000). Testing the Female
Underperformance Hypothesis. Small Business Economics, 14(1), 1–10.
Duchesneau, D. A. and Gartner, W. B. (1990). A Profile of New Venture
Success and Failure in an Emerging Industry. Journal of Business
Venturing, 5(5), 297–312.
Dun and Bradstreet (1979). The Business Failure Record. New York: Dun
and Bradstreet.
Dunne, T., Roberts, M. J. and Samuelson, L. (1989). The Growth
and Failure of U.S. Manufacturing Plants. The Quarterly Journal of
Economics, 104(4), 672–698.
Esteve-Pérez, S. and Mañez-Castillejo, J. A. (2008). The Resource-Based
Theory of the Firm and Firm Survival. Small Business Economics, 30(3),
231–249.
Evans, D. S. (1987). The Relationship Between Firm Growth, Size, and
Age: Estimates for 100 Manufacturing Industries. The Journal of
Industrial Economics, 35(4), 567–581.
Everett, J. and Watson, J. (1998). Small Business Failure and External
Risk Factors. Small Business Economics, 5(3), 371–390.
Fasci, M. A. and Valdez, J. (1998). A Performance Contrast of Male- and
Female-Owned Small Accounting Practices. Journal of Small Business
Management, 36(3), 1–7.
Fischer, E. (1992). Sex Differences and Small-Business Performance
Among Canadian Retailers and Service Providers. Journal of Small
Business and Entrepreneurship, 9(4), 2–13.
Fischer, E. and Reuber, R. A. (2003). Support for Rapid-Growth Firms:
A Comparison of the Views of Founders, Government Policymakers,
and Private Sector Resource Providers. Journal of Small Business
Management, 41(4), 346–365.
Fischer, E. M., Reuber, R. A. and Dyke, L. S. (1993). A Theoretical
Overview and Extension of Research on Sex, Gender, and
Entrepreneurship. Journal of Business Venturing, 8(2), 151–168.
Florin, J., Lubatkin, M. and Schulze, W. (2003). A Social Capital Model
References 145
of High-Growth Ventures. Academy of Management Journal, 46(3),

374–384.
Forlani, D. and Mullins, J. W. (2000). Perceived Risks and Choices in
Entrepreneurs’ New Venture Decisions. Journal of Business Venturing,
15(4), 305–322.
Forsyth, G. D. (2005). A Note on Small Business Survival Rates in Rural
Areas: The Case of Washington State. Growth and Change, 36(3),
428–440.
Foster, G. (1986). Financial Statement Analysis. Englewood Cliffs, NJ:
Prentice-Hall.
Fraser, S. (2006). Finance for Small and Medium-Sized Enterprises: A
Report on the 2004 UK Survey of SME Finances. Coventry, UK: Centre
for Small and Medium-Sized Enterprises, Warwick Business School.
Fredland, E. J. and Morris, C. E. (1976). A Cross Section Analysis of
Small Business Failure. American Journal of Small Business, 1(1), 7–18.
Freeman, J., Carroll, G. R. and Hannan, M. T. (1983). The Liability of
Newness: Age Dependence in Organizational Death Rates. American
Sociological Review, 48(5), 693–710.
Fried, V. H. and Hisrich, R. D. (1994). Toward a Model of Venture
Capital Investment Decision Making. Financial Management, 23(3),
28–37.
Ganguly, P., Ed. (1985). UK Small Business Statistics and International
Comparisons. London: Harper and Row.
Garrod, P. and Miklius, W. (1990). Modeling the Timing of Business Firm
Exits. Small Business Economics, 2(2), 141–153.
Gimeno, J., Folta, T. B., Cooper, A. C. and Woo, C. Y. (1997). Survival
of the Fittest? Entrepreneurial Human Capital and the Persistence
of Underperforming Firms. Administrative Science Quarterly, 42(4),
750–783.
Granovetter, M. S. (1973). The Strength of Weak Ties. The American
Journal of Sociology, 78(6), 1360–1380.
Granovetter, M. S. (1983). The Strength of Weak Ties: A Network Theory
Revisited. Sociological Theory, 1(1), 201–233.
Hall, G. and Young, B. (1991). Factors Associated with Insolvency
Amongst Small Firms. International Small Business Journal, 9(2),
54–63.
Hall, K. S. (1986), Business Failures: Economic Determinants and Potential
Effects on the Macroeconomy, Unpublished Doctoral Thesis, The
George Washington University.
Hamilton, B. (1984). Entry and Exit of Businesses in Scotland, in Success
and Failure in Small Business. Ed. J. Lewis, J. Stanworth and A. Gibb,
Aldershot, UK: Gower. 57–69.
146 SME performance
Hamilton, R. T. and Fox, M. A. (1998). The Financing Preferences of

Small Firm Owners. International Journal of Entrepreneurial Behaviour
and Research, 4(3), 239–248.
Harada, N. (2007). Which Firms Exit and Why? An Analysis of Small
Firm Exits in Japan. Small Business Economics, 29(4), 401–414.
Havnes, P. and Senneseth, K. (2001). A Panel Study of Firm Growth
Among SMEs in Networks. Small Business Economics, 16(4),
293–302.
Headd, B. (2003). Redefining Business Success: Distinguishing Between
Closure and Failure. Small Business Economics, 21(1), 51–61.
Hudson, J. (1997). Company Bankruptcies and Births Matter. Applied
Economics, 29(5), 647–654.
Hustedde, R. J. and Pulver, G. C. (1992). Factors Affecting Equity Capital
Acquisition: The Demand Side. Journal of Business Venturing, 7(5),
363–374.
Hutchinson, R. G., Hutchinson, A. R. and Newcomer, M. (1938). A Study
in Business Mortality. American Economic Review, 28(3), 497–514.
Hutchinson, R. W. (1995). The Capital Structure and Investment
Decisions of the Small Owner-Managed Firm: Some Exploratory
Issues. Small Business Economics, 7(3), 231–239.
Ibarra, H. (1992). Homophily and Differential Returns: Sex Differences in
Network Structure and Access in an Advertising Firm. Administrative
Science Quarterly, 37(3), 422–447.
Jackson, D. N. (1976). Personality Inventory Manual. Goshen, NY:
Research Psychologists Press.
Jarillo, C. J. (1989). Entrepreneurship and Growth: The Strategic Use of
External Resources. Journal of Business Venturing, 4(2), 133–147.
Jianakoplos, N. A. and Bernasek, A. (1998). Are Women More Risk
Averse? Economic Inquiry, 36(4), 620–630.
Jovanovic, B. (1982). Selection and the Evolution of Industry.
Econometrica, 50(3), 649–670.
Julien, P. A. (1993). Small Business as a Research Subject: Some
Reflections on Knowledge of Small Businesses and its Effects on
Economic Theory. Small Business Economics, 5(2), 157–166.
Kalleberg, A. L. and Leicht, K. T. (1991). Gender and Organizational
Performance: Determinants of Small Business Survival and Success.
Academy of Management Journal, 34(1), 136–161.
Kent, P. (1994). Management Advisory Services and the Financial
Performance of Clients. International Small Business Journal, 12(4),
45–58.
Knott, A. M. and Posen, H. E. (2005). Is Failure Good? Strategic
Management Journal, 26(7), 617–641.
References 147
Kon, Y. and Storey, D. J. (2003). A Theory of Discouraged Borrowers.

Small Business Economics, 21(1), 37–49.
Land, N. R. (1975). Too Much Emphasis on Management Assistance?
Larson, A. and Starr, J. A. (1993). A Network Model of Organization
Formation. Entrepreneurship Theory and Practice, 17(2), 5–15.
Larsson, E., Hedelin, L. and Garling, T. (2003). Influence of Expert
Advice on Expansion Goals of Small Businesses in Rural Sweden.
Lerner, M., Brush, C. and Hisrich, R. (1997). Israeli Women Entrepreneurs:
An Examination of Factors Affecting Performance. Journal of Business
Venturing, 12(4), 315–339.
Leslie, M., Magdulski, N. and Champion, N. (1985). The Role of
the Accountant in the Survival of Small Business. The Australian
Accountant, 55(1), 22–30.
Levenson, A. R. and Willard, K. L. (2000). Do Firms Get the Financing
They Want? Measuring Credit Rationing Experienced by Small
Businesses in the U.S. Small Business Economics, 14(2), 83–94.
Levenson, H. (1974). Activism and Powerful Others: Distinctions
Within the Concept of Internal-External Control. Journal of Personal
Assessment, 38, 377–383.
Lin, N., Ensel, W. M. and Vaughn, J. C. (1981). Social Resources and
Strength of Ties: Structural Factors in Occupational Status Attainment.
American Sociological Review, 46(4), 393–405.
Littunen, H. (2000). Networks and Local Environmental Characteristics
in the Survival of New Firms. Small Business Economics, 15(1), 59–71.
Loscocco, K. A., Robinson, J., Hall, R. H. and Allen, J. K. (1991).
Gender and Small Business Success: An Inquiry into Women’s Relative
Disadvantage. Social Forces, 70(1), 65–85.
Lowe, J., McKenna, J. and Tibbits, G. (1991). Small Firm Growth and
Failure: Public Policy Issues and Practical Problems. Economic Papers,
10(2), 69–81.
Massel, M. Z. (1978). It’s Easier to Slay a Dragon Than Kill a Myth.
McGrath, R. G. (1999). Falling Forward: Real Options Reasoning
and Entrepreneurial Failure. Academy of Management Review, 24(1),
13–30.
Michaelas, N., Chittenden, F. and Poutziouris, P. (1999). Financial Policy
and Capital Structure Choice in U.K. SMEs: Empirical Evidence from
Company Panel Data. Small Business Economics, 12(2), 113–130.
Moore, G. (1990). Structural Determinants of Men’s and Women’s
Personal Networks. American Sociological Review, 55(5), 726–735.
148 SME performance
Mukhtar, S.-M. (2002). Differences in Male and Female Management

Characteristics: A Study of Owner-Manager Businesses. Small Business
Economics, 18(4), 289–311.
Munch, A., McPherson, J. M. and Smith-Lovin, L. (1997). Gender,
Children, and Social Contact: The Effects of Childrearing for Men and
Women. American Sociological Review, 62(4), 509–520.
Murphy, G. B., Trailer, J. W. and Hill, R. C. (1996). Measuring
Performance in Entrepreneurship Research. Journal of Business
Research, 36(1), 15–23.
Myers, S. C. (1984). The Capital Structure Puzzle. Journal of Finance,
39(3), 575–592.
Naffziger, D. W., Hornsby, J. S. and Kuratko, D. F. (1994). A Proposed
Research Model of Entrepreneurial Motivation. Entrepreneurship
Theory and Practice, 18(3), 29–42.
Nelson, G. W. (1989). Factors of Friendship: Relevance of Significant
Others to Female Business Owners. Entrepreneurship Theory and
Practice, 13(4), 7–18.
Newby, R. R., Watson, J. and Woodliff, D. (2003). SME Survey
Methodology: Response Rates, Data Quality and Cost Effectiveness.
Entrepreneurship Theory and Practice, 28(2), 163–172.
Orhan, M. (2001). Women Business Owners in France: The Issue of
Financing Discrimination. Journal of Small Business Management,
39(1), 95–102.
Palepu, K. G. and Healy, P. M. (2008). Business Analysis and Valuation:
Using Financial Statements, 4th edn. Mason, OH: South-Western
College Publishing.
Phillips, B. D. and Kirchoff, B. A. (1989). Formation, Growth and
Survival: Small Firm Dynamics in the U.S. Economy. Small Business
Economics, 1(1), 65–74.
Pissarides, F. (1999). Is Lack of Funds the Main Obstacle to Growth?
EBRD’s Experience with Small- and Medium-Sized Businesses in Central
and Eastern Europe. Journal of Business Venturing, 14(5–6), 519–539.
Potts, A. J. (1977). A Study of the Success and Failure Rates of Small
Businesses and the Use or Non-Use of Accounting Information, Doctoral
Thesis, The George Washington University.
Powell, M. and Ansic, D. (1997). Gender Differences in Risk Behaviour
in Financial Decision-Making: An Experimental Analysis. Journal of
Economic Psychology, 18(6), 605–628.
Price, R. (1984). Small Business Management for Small Business Owner/
Managers. Melbourne: Wordsworth.
Reynolds, P. D. (1987). New Firms: Societal Contribution Versus Survival
Potential. Journal of Business Venturing, 2(3), 231–246.
References 149
Riding, A. and Swift, C. S. (1990). Women Business Owners and Terms of

Credit: Some Empirical Findings of the Canadian Experience. Journal
Robson, P. and Bennett, R. J. (2000). SME Growth: The Relationship
with Business Advice and External Collaboration. Small Business
Economics, 15(3), 193–208.
Robson, P., Jack, S. L. and Freel, M. S. (2008). Gender and the Use of
Business Advice: Evidence from Firms in the Scottish Service Sector.
Environment and Planning C: Government and Policy, 26(2), 292–314.
Rosa, P., Carter, S. and Hamilton, D. (1996). Gender as a Determinant
of Small Business Performance: Insights from a British Study. Small
Business Economics, 8(6), 463–478.
Rosa, P., Daphne, H., Sara, C. and Helen, B. (1994). The Impact of
Gender on Small Business Management: Preliminary Findings of a
British Study. International Small Business Journal, 12(3), 25–32.
Said, K. E, and Hughey, K. J. (1977). Managerial Problems of the Small
Firm. Journal of Small Business Management, 15(1), 37–42.
Schumpeter, J. (1942). Capitalism, Socialism and Democracy. New York:
Harper.
Scott, M. (1982). Mythology and Misplaced Pessimism: The Real
Failure Record of New, Small Businesses, in Stimulating Small
Firms. Ed. D. Watkins, J. Stanworth and A. Westrip, Farnham, UK:
Gower.
Scott, M. and Lewis, J. (1984). Re-thinking Entrepreneurial Failure, in
Success and Failure in Small Business. Ed. J. Lewis, J. Stanworth and A.
Gibb, Farnham, UK: Gower. 29–56.
Seibert, S. E., Kraimer, M. L. and Liden, R. C. (2001). A Social Capital
Theory of Career Success. Academy of Management Journal, 44(2),
219–237.
Sexton, D. L. and Bowman-Upton, N. (1990). Female and Male
Entrepreneurs: Psychological Characteristics and their Role in
Gender-Related Discrimination. Journal of Business Venturing, 5(1),
29–36.
Shailer, G. (1989). Research Note: The Predictability of Small Enterprise
Failures: Evidence and Issues. International Small Business Journal,
7(4), 54–58.
Sharpe, W. F. (1975). Adjusting for Risk in Portfolio Performance
Measurement. The Journal of Portfolio Management, 1(2), 29–34.
Sharpe, W. F. (1981). Investments. Englewood Cliffs, NJ: Prentice-Hall.
Shaw, E., Lam, W. and Carter, S. (2008). The Role of Entrepreneurial
Capital in Building Service Reputation. Service Industries Journal,
28(7), 899–917.
150 SME performance
Shepherd, D. A., Ettenson, R. and Crouch, A. (2000). New Venture

Strategy and Profitability: A Venture Capitalist’s Assessment. Journal
of Business Venturing, 15(5–6), 449–467.
Shuman, J. C. (1975). Corporate Planning in Small Companies: A Survey.
Long Range Planning, 8(5), 81–90.
Stanworth, J. (1995). Penetrating the Mists Surrounding Franchise Failure
Rates: Some Old Lessons for New Businesses. International Small
Business Journal, 13(2), 59–63.
Star, A. D. and Massel, M. Z. (1981). Survival Rates for Retailers. Journal
of Retailing, 57(2), 87–99.
Stewart, H. and Gallagher, C. (1986). Business Death and Firm Size in the
UK. International Small Business Journal, 4(1), 42–57.
Stinchcombe, A. L. (1965). Social Structure and Organizations, in
Handbook of Organizations. Ed. J. G. March, Chicago, IL: Rand
McNally. 153–193.
Storey, D., Keasey, K., Watson, R. and Wynarczyk, P. (1987). The
Performance of Small Firms: Profits, Jobs and Failures. London, UK:
Croom Helm.
Taggart, R. A. (1996). Quantitative Analysis for Investment Management.
Upper Saddle River, NJ: Prentice Hall.
Tauzell, J. (1982). Survival of Minnesota New Businesses: 1977–1980.
Review of Labour and Economic Conditions (Research and Statistical
Offices, Minnesota Department of Economic Security), 9(2), 10–17.
Thankappan, V. and Hammer, G. R. (1980). Small Business Observations.
American Journal of Small Business, 5(July–Sept.), 1–2.
Ulmer, M. J. and Nielsen, A. (1947). Business Turn-Over and Causes of
Failure. Survey of Current Business, 27(4), 10–16.
Uzzi, B. (1996). The Sources and Consequences of Embeddedness for
the Economic Performance of Organizations: The Network Effect.
American Sociological Review, 61(4), 674–698.
Uzzi, B. (1997). Social Structure and Competition in Interfirm Networks:
The Paradox of Embeddedness. Administrative Science Quarterly, 42(1),
35–67.
Wadhwani, S. B. (1986). Inflation, Bankruptcy, Default Premia and the
Stock Market. The Economic Journal, 96(March), 120–138.
Watson, J. (1995), Small Business Failure Rates in Managed Shopping Centres:
Definition, Measurement and the Impact of Macro-Economic Variables,
Doctor of Philosophy thesis, The University of Western Australia.
Watson, J. (2002). Comparing the Performance of Male- and Female-
Controlled Businesses: Relating Outputs to Inputs. Entrepreneurship
Theory and Practice, 26(3), 91–100.
Watson, J. (2003). Failure Rates for Female Controlled Businesses: Are
References 151
They Any Different? Journal of Small Business Management, 41(3),

262–277.
Watson, J. (2006). External Funding and Firm Growth: Comparing
Female- and Male-Controlled SMEs. Venture Capital: An International
Journal of Entrepreneurial Finance, 8(1), 33–49.
Watson, J. (2007). Modeling the Relationship Between Networking and
Firm Performance. Journal of Business Venturing, 22(6), 852–874.
Watson, J. and Everett, J. E. (1993). Defining Small Business Failure.
International Small Business Journal, 11(3), 35–48.
Watson, J. and Everett, J. E. (1996a). Do Small Businesses Have Higher
Failure Rates: Evidence from Australian Retailers. Journal of Small
Business Management, 34(4), 45–62.
Watson, J. and Everett, J. E. (1996b). Small Business Failure Rates:
Choice of Definition and the Size Effect. The Journal of Entrepreneurial
and Small Business Finance, 5(3), 271–285.
Watson, J. and Everett, J. E. (1999). Small Business Failure Rates: Choice
of Definition and Industry Effects. International Small Business Journal,
17(2), 33–49.
Watson, J. and Newby, R. (2005). Biological Sex, Stereotypical Sex-
Roles, and SME Owner Characteristics. International Journal of
Entrepreneurial Behaviour and Research, 11(2), 129–143.
Watson, J., Newby, R. and Mahuka, A. (2006). Comparing the Growth and
External Funding of Male- and Female-Controlled SMEs in Australia,
in Growth-Oriented Women Entrepreneurs and their Businesses. Ed. C.
G. Brush, N. M. Carter, E. J. Gatewood, P. G. Greene and M. M. Hart,
Cheltenham, UK and Northampton, MA, USA: Edward Elgar. 205–231.
Watson, J., Newby, R. and Mahuka, A. (2009). Gender and the SME
‘Finance Gap’. International Journal of Gender and Entrepreneurship,
1(1), 42–56.
Watson, J. and Robinson, S. (2003). Adjusting for Risk in Comparing
the Performances of Male- and Female-Controlled SMEs. Journal of
Business Venturing, 18(6), 773–788.
Williams, M. L. (1993). Measuring Business Starts, Success and Survival: Some
Database Considerations. Journal of Business Venturing, 8(4), 295–299.
Wilson, H., (1979), Interim Report on The Financing of Small Firms,
London: Her Majesty’s Stationery Office.
Winborg, J. and Landstrom, H. (2001). Financial Bootstrapping in Small
Businesses: Examining Small Business Managers’ Resource Acquisition
Behaviors. Journal of Business Venturing, 16(3), 235–254.
Zhao, L. and Aram, J. D. (1995). Networking and Growth of Young
Technology-Intensive Ventures in China. Journal of Business Venturing,
10(5), 349–370.
Index
accountants Australian Bureau of Statistics (ABS)
frequency of contact 118–19, 120 data
Institute of Chartered Accountants age of business 33–4, 35
3 failure rates 26–7, 29
and performance 123–6, 127, 128, female-controlled SMEs 48–52
130, 137, 139 growth, relationship with funding
age of business 90–95
and failure rates 31–4, 35, 42, 135 networks 104–5, 117–20
female-controlled SMEs 44, 49–50, output/input relationship, as
51–2, 57–8, 63–4 performance measure 54–8
and growth 91–5 risk, attitudes to 62–4
and networks 108–15, 121–2, 124,
125 Ballantine, J.W. 59
Aldrich, H. 99, 100, 116, 128, 137 bank funding
Allen, K.R. 67 attitudes to 71–8, 79–86, 136–7
Anna, A.L. 44, 51 demand for funding 69–70
Ansic, D. 61, 89 female-controlled SMEs 6–7, 67,
Aram, J.D. 103 79–86, 88–90, 136
assets 54–8 focus group results 70–78
Australia, SMEs in and growth 67–8, 90–97, 136
age of business, and failure rates owners’ considerations 70–86
31–4, 35 Pecking Order Theory 87, 88,
bankruptcies 26, 29 89–90, 94–7
case studies 17–20 questionnaire results 78–85
closure rates 26–7, 29 see also ‘finance gap’
economic conditions, effect of bankruptcy
39–40 age of business 31–3
failure statistics 20–25, 27, 28 comparative studies 25–6, 29, 31
female-controlled SMEs 48–52 economic conditions, effect of 39,
focus group survey of demand for 40, 135
finance 70–78 as failure 15–16, 18–20, 133
growth, relationship with funding failure statistics 20–25
90–95 industry sector 37–8, 42
networks 104–5, 117–20 size of business 36–7, 42, 135
output/input relationship, as banks 118–19
performance measure 54–8 Barber, B.M. 61
questionnaire regarding demand for Bates, T. 14, 15, 25, 31, 129, 138
finance 78–85 Becchetti, L. 69, 85, 91
risk, attitudes to 62–4 Belgium 25, 29
size of business, and failure rates Bennett, R.J. 105, 118, 128, 137
36–7 Berger, A.N. 69, 87
153
154 SME performance
Berggren, B. 70 creditors, losses to 15–16

Bernasek, A. 61 see also bankruptcy; disposal, to
Bhattacharjee, A. 41 limit losses
Bhide, A. 134 Cressy, R. 32, 72–3, 89, 91
Birley, S. 14, 69, 99, 105, 116, 117, 118, Cromie, S. 99, 116, 117, 118, 119
119, 120
Boden, R.J. 47, 49 data 3–7
Bowman-Upton, N. 61, 89 see also Australian Bureau of
Box, M. 14, 28, 29, 32 Statistics (ABS) data
Brüderl, J. 32, 37, 126, 128 Davidsson, P. 90
Bruno, A.V. 67 debt 54, 58, 88–90
Brush, C. 69, 102 debt to asset ratio 90–95
Building Owners and Managers DeCarolis, D. 70
Association in Australia Deeds, D. 70
(BOAMA) 20 Delmar, F. 90
Burt, R.S. 102, 126, 128 Dewaelheyns, N. 25, 29
business consultants 118–19 discontinuance (sale or closure) of
business plans 74, 75, 78 business
Buttner, E.H. 70, 134 age of business 31–4
economic conditions, effect of 39,
Canada 61 40
capital 44, 45, 47, 54–8 as failure 13–15, 18–20, 133–4
Carpenter, R.E. 67, 94 failure statistics 20–25
Carter, N.M. 47, 49, 67, 69, 102 industry sector 42
Carter, S. 44, 53, 119 size of business 35–7, 42, 135
Chaganti, R. 70 see also closure of business; sale of
Cleveland, F.W. 59 business
Cliff, J.E. 61, 70, 95 disposal, to limit losses
closure of business age of business 31–3
age of business 31–4 economic conditions, effect of 39,
comparative studies 26–8, 29, 31 40
economic conditions, effect of 39, 40 as failure 16, 18–20, 133
as failure 13–15, 18–20, 133–4 failure statistics 20–25
failure statistics 20–25 size of business 36–7
industry sector 37–8, 42 Donckels, R. 102
size of business 35–7, 42 Duchesneau, D.A. 101
see also discontinuance (sale or Dunkelberg, W.C. 14, 105, 118
closure) of business; sale of Dyke, L.S. 43
business
Cochran, A.B. 11, 13, 16, 17 economic conditions, effect of 38–41,
Coleman, J.S. 101 42, 135
construction businesses 50, 52, 92, 93, education, effect of 91–4, 121–2
121–2, 125 employment rates 39, 40, 41
control equity funding 54–8, 96, 106–15
female-controlled SMEs 83–5, 89 Everett, J.E. 20–25, 29, 39, 41
finance for growth 83–5 experience of owner, effect of 91–4,
male-controlled SMEs 83–5, 89 121–2
retention by owner 72–3, 75, 76, 78 external funding
Cooper, A.C. 14, 47, 100, 105, 118, attitudes to 71–8, 79–86, 136–7
128, 137 demand for funding 69–70
Index 155
female-controlled SMEs 6–7, 67, intensity (access frequency) of

79–86, 88–90, 136 networks 118–20
focus group results 70–78 vs. male-controlled SMEs 43–5
and growth 67–8, 90–97, 136 motivation 45, 134
owners’ considerations 70–86 networks 99–100, 116–17, 120–130,
Pecking Order Theory 87, 88, 137–9
89–90, 94–7 output/input relationship, as
questionnaire results 78–85 performance measure 53–8
see also ‘finance gap’ performance 99–100, 129–30
range (number) of networks 117–18
failures risk, attitudes to 45, 59, 61–4, 75,
and age of business 31–4, 35, 42, 76, 78, 83–6, 89
135 size of business 44–5, 49–50, 51–2,
case studies 17–20 63–4
definitions 7, 13–17, 18–20, 133–4 finance
economic conditions 38–41, 42, 135 attitudes to 71–8, 79–86, 136–7
female-controlled SMEs 47–52, demand for funding 69–70
135–6 female-controlled SMEs 6–7, 67,
industry sector 37–8, 42, 135 79–86, 88–90, 136
male-controlled SMEs 47–52, 136 focus group results 70–78
selection criteria 11–13 and growth 67–8, 90–97, 136
size of business 34–7, 42, 135 male-controlled SMEs 79–86
statistics 3–7, 20–28, 29–30, 31 owners’ considerations 70–86
family networks Pecking Order Theory 87, 88,
female-controlled SMEs 117, 126, 89–90, 94–7
129 questionnaire results 78–85
frequency of contact 119, 137–8 ‘finance gap’
survival of the business 124, 130, and growth 94–7, 136
139 meaning of 69, 87–8
Fasci, M.A. 56 survey results 71, 74, 78, 79–80
female-controlled SMEs financial businesses 50, 52, 92, 93,
and accountants 118–19, 120, 121–2, 125
123–6, 127, 128, 130 financial institutions 71–8, 79–86,
age of business 44, 49–50, 51–2, 88–90
57–8, 63–4 see also bank funding
control, retention of 83–5, 89 financial return, as motivation 134
data 6–7 Fischer, E. 43, 59, 101, 115
debt levels 88–90 Florin, J. 99
failure rates 47–52, 135–6 Forlani, D. 59, 60
finance for 6–7, 67, 79–86, 88–90, formal networks
136 female-controlled SMEs 118–20,
formal networks 118–20, 123–8, 129 123–8, 129
growth, relationship with funding intensity (access frequency) 104–5
90–97 male-controlled SMEs 118–20,
and industry associations 118–19, 123–8
124, 125, 126, 127, 130 nature of 103
industry sector 44, 50–52, 57–8, and performance 111–15, 137–9
63–4 range (number of) 104–5
informal networks 118–20, 123–8, Forsyth, G.D. 14, 28, 29
129 Foster, G. 39
156 SME performance
Fox, M.A. 69, 85, 136 industry sector

Fraser, S. 79, 80 and failure rates 37–8, 42, 135
Fredland, E.J. 13, 37, 39 female-controlled SMEs 44, 50–52,
Freel, M.S. 119 57–8, 63–4
Fried, V.H. 68 and growth 91–4
friend networks and networks 108–15, 121–2, 125
female-controlled SMEs 117, 126, informal networks
129 female-controlled SMEs 118–20,
frequency of contact 119, 137–8 123–8, 129
survival of the business 124, 130, 139 intensity (access frequency) 104–5
funding see external funding male-controlled SMEs 118–20,
123–8
Gallagher, C. 14, 25, 26, 31, 37 nature of 103
Ganguly, P. 14, 32 and performance 111–15, 130,
Garrod, P. 13–14, 26 137–9
Gartner, W.B. 90, 101 range (number of) 104–5
Gimeno-Gascon, J.F. 100, 128 input/output relationship, as
goals, of SME owners 134, 139 performance measure 53–8
government regulation 71, 72 Institute of Chartered Accountants
Granovetter, M.S. 101, 102, 103, 126, 3
128 see also accountants
growth insurance businesses 50, 52, 92, 93,
and finance for SMEs 67–8, 90–97, 121–2, 125
136 interest rates 39, 40, 41
and ‘finance gap’ 94–7, 136
and networks 106–15, 120–128, 129, Jack, S.L. 119
138 Japan 26
obstacles to 70–78 Jianakoplos, N.A. 61
and Pecking Order Theory 94–7 Jovanovic, B. 31, 34, 35, 36, 51
see also performance Julien, P.A. 99
Hall, K.S. 40 Kalleberg, A.L. 47

Hamilton, D. 44, 53 Kent, P. 102
Hamilton, R.T. 69, 85, 136 Kirchoff, B.A. 14, 37
Harada, N. 15, 26 Knott, A.M. 7
Headd, B. 14, 15, 21, 27–8, 29 Koeller, T.C. 59
Hill, R.C. 134 Kon, Y. 74
Hisrich, R.D. 68, 102 Kraimer, M.L. 102
hours worked, effect of 45, 56–8
Hudson, J. 15, 25, 29 Lam, W. 119
human capital 45, 47, 91–4 Lambrecht, J. 102
Hustedde, R.J. 102 Landstrom, H. 69
Hutchinson, A.R. 14, 35–6, 38, 44, 51 Larsson, E. 102
Hutchinson, R.G. 14, 35–6, 38, 44, 51 legal organisation (status) of the
Hutchinson, R.W. 70 business 91–4
Leicht, K.T. 47
Ibarra, H. 103 Lerner, M. 102
income 54–8, 90, 106–15, 120–127 Levenson, A.R. 71, 79
industry associations 118–19, 124, leverage 54, 58
125, 126, 127, 130, 139 Lewis, J. 4, 11
Index 157
Liden, R.C. 102 size of business, and failure rates

Littunen, H. 103 36–7
losses 15–16 manufacturing businesses 50, 52, 92,
see also bankruptcy; disposal, to 93, 121–2, 125
limit losses McGrath, R.G. 7
Lowe, J. 15, 37 McKenna, J. 15, 37
Lubatkin, M. 99 McPherson, J.M. 103, 116
Miklius, W. 13–14, 26
‘make a go of it’, failure to mining businesses 50–51, 52, 92, 93,
age of business 31–3 121–2, 125
economic conditions, effect of 39, Moore, D.P. 70, 134
40 Moore, G. 103, 117
as failure 16–17, 18–20, 133 Morris, C.E. 13, 37, 39
failure statistics 20–25 motivation 45, 134, 139
industry sector 38 Mukhtar, S.-M. 89
size of business 36–7 Mullins, J.W. 59, 60
male-controlled SMEs Munch, A. 103, 116
and accountants 118–19, 120, Murphy, G.B. 134
123–6, 127, 128 Myers, S.C. 87
control, retention of 83–5, 89
failure rates 47–52, 136 Nelson, G.W. 138
vs. female-controlled SMEs 43–5 networks
finance, demand considerations definition 99
79–86 female-controlled SMEs 99–100,
formal networks 118–20, 123–8 116–17, 120–130, 137–9
growth, relationship with funding and growth 106–15, 120–128, 129,
90–97 138
and industry associations 118–19, intensity (access frequency) 103–5,
124, 125, 126, 127, 130, 139 114, 115, 117–20
informal networks 118–20, male-controlled SMEs 99–100,
123–8 116–17, 120–130, 137–9
intensity (access frequency) of nature of 102–3
networks 118–20 and performance 99–100, 101–2,
networks 99–100, 116–17, 120–130, 105–14, 120–130, 137–9
137–9 range (number of) 103–5, 114,
output/input relationship, as 115
performance measure 54–8 see also formal networks; informal
performance 99–100, 120–130, networks
137–9 New Zealand 69
range (number) of networks Newby, R. 61
117–18 Newcomer, M. 14, 36, 38, 44, 51
risk, attitudes to 59, 61–4, 75, 76, Nielsen, A. 13, 16, 17
78, 83–6, 89 Nucci, A.R. 14, 25, 31, 47, 49
managed shopping centres
age of business, and failure rates Odean, T. 61
31–3 Olofsson, C. 70
as data sources 6 Orhan, M. 117
economic conditions, effect of output/input relationship, as
39–40 performance measure 53–8
failure statistics 20–25, 27, 28 owner withdrawals 96
158 SME performance
partnerships 14–15, 36 Reynolds, P.D. 5, 14, 47

Pecking Order Theory 87, 88, 89–90, Riding, A. 69
94–7 risk
performance controlling for 59–60
and accountants 123–6, 127, 128, female-controlled SMEs 45, 59,
130, 137, 139 61–4, 75, 76, 78, 83–6, 89
female-controlled SMEs 99–100, finance for growth 72, 73–4, 75, 76,
129–30 78, 83–5
and formal networks 111–15, 137–9 male-controlled SMEs 59, 61–4, 75,
and informal networks 111–15, 130, 76, 78, 83–6, 89
137–9 Robson, P. 105, 118, 119, 120, 128,
male-controlled SMEs 99–100, 137
120–130, 137–9 Rosa, P. 44, 51, 53
and networks 99–100, 101–2,
105–14, 120–130, 137–9 sale of business
and professional advice 100, 101–2, age of business 31–3
104–5, 118–19 economic conditions, effect of 40
see also growth as failure 13–15, 18–20, 133–4
performance measures failure statistics 20–25
case studies of failure 17–20 size of business 35–7
definitions of failure 13–17, 18–20, see also closure of business;
133–4 discontinuance (sale or closure)
failure statistics 20–28, 29–30, 31 of business
female-controlled SMEs 47–52, sales 39, 40, 41, 90, 125, 127
135–6 Schulze, W. 99
selection criteria 11–13 Scott, M. 4, 11
see also failures Seibert, S.E. 102
Petersen, B.C. 67, 94 service businesses 50, 52, 92, 93,
Phillips, B.D. 14, 37 121–2, 125
Posen, H.E. 7 Sexton, D.L. 61, 89
Potts, A.J. 5, 24, 101, 128, 137 Shailer, G. 39
Powell, M. 61, 89 Sharpe ratio 60, 62–4
Preisendörfer, P. 32, 37, 126, 128 Sharpe, W.F. 39, 59–60
professional advice 100, 101–2, 104–5, Shaw, E. 119
118–19 shopping centres
see also accountants age of business, and failure rates
profits 20–21, 54–8, 60–64, 90–94, 96 31–3
property businesses 50, 52, 92, 93, as data sources 6
121–2, 125 economic conditions, effect of
Pulver, G.C. 102 39–40
failure statistics 20–25, 27, 28
Reese, P.R. 100 size of business, and failure rates
regulation 71, 72 36–7
relationship lending 78 Silver, L. 70
retail businesses 50, 52, 92, 93, 121–2, size of business
125 and failure rates 34–7, 42, 135
retail sales 39, 40, 41, 90, 125, 127 female-controlled SMEs 44–5,
return on assets (ROA) 54–8 49–50, 51–2, 63–4
return on equity (ROE) 54–8, 106–15 growth, and funding 90–94
Reuber, R.A. 43, 101, 115 and networks 108–15, 121–2
Index 159
SMEs (small- and medium-sized Udell, G.F. 69, 87

enterprises) Ulmer, M.J. 13, 16, 17
case studies 17–20, 28–30 unemployment rates 39, 40–41
definitions of failure 7, 13–17, United Kingdom (UK) 25, 29, 41,
18–20, 133 79
economic conditions, effect of United States (US)
38–41, 42, 135 bankruptcies 26
future research 139 closure rates 27–8, 29
misconceptions about 3–4, 6–7 economic conditions, effect of
statistics 3–7, 20–28, 29–30, 31 39
see also age of business; failures; failure rate statistics 4
female-controlled SMEs; female-controlled SMEs 47, 89
finance; industry sector; male- finance for SMEs 67, 69, 79
controlled SMEs; performance; growth, relationship with funding
size of business 94
Smith-Lovin, L. 103, 116 risk, attitudes to 61
social resource theory 102 Uzzi, B. 126
sole traders 14–15, 36
solicitors 118–19 Valdez, J. 56
Stanworth, J. 4, 14 Van Hulle, C. 25, 29
Stewart, H. 14, 25, 26, 31, 37 variability, and risk 60
Storey, D.J. 74
structural holes 102, 126, 128 Wadhwani, S.B. 40
survival of the business 106–15, Watson, J. 20–25, 29, 39, 41, 61, 105,
120–128, 129, 130 106
Sweden 28, 29, 70 weak tie theory 102, 126, 128
Swift, C.S. 69 wholesale trade 50, 52, 92, 93, 121–2,
125
Taggart, R.A. 60 Willard, K.L. 71, 79
tax office 119 Williams, M. 47
Tibbits, G. 15, 37 Wilson Committee 4
Trailer, J.W. 134 Winborg, J. 69
transaction-based lending 78 women see female-controlled SMEs
transport businesses 50, 52, 92, 93, Woo, C. 14, 100, 105, 118, 128
121–2, 125
Trovato, G. 69, 85, 91 Zhao, L. 103
Tyebjee, T.T. 67 Ziegler, R. 32, 37

SME Performance Separating Myth From Reality (PDFDrive)

Uploaded by

Copyright:

Available Formats

You might also like

SME Performance Separating Myth From Reality (PDFDrive)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SME Performance Separating Myth From Reality (PDFDrive)

Uploaded by

Copyright:

Available Formats

SME Performance

All rights reserved. No part of this publication may be reproduced, stored in a

Edward Elgar Publishing, Inc.

Library of Congress Control Number: 2009940729

ISBN 978 1 84542 977 5

Printed and bound by MPG Books Group, UK

PART II SME PERFORMANCE

2 Defining and measuring SME failure/success 11

PART III COMPARING THE PERFORMANCE OF

PART IV GROWTH FINANCING FOR SMES

PART V NETWORKING AND SME PERFORMANCE

9 The association between networking and performance 101

11 Conclusions, implications and areas for future research 133

I first became interested in the small and medium enterprise (SME)

information on the subject, these perceptions are permitted to continue

with an idea! I would try to access information on SMEs from managed

1.1 OUTLINE OF THE REMAINING CHAPTERS

It should be noted that the focus in this section is on dichotomous

2.1 CRITERIA FOR SELECTING A MEASURE OF

Prior to reviewing a number of alternative definitions of failure that have

The use of an objective measure makes replication of results by researchers

The selected measure should faithfully represent that which it purports to

Reliability/Freedom from Bias

It is important that the measure selected should, within reason, be free

Ultimately, the choice of a failure/success measure is likely to involve a

2.2 ALTERNATIVE DEFINITIONS OF SME

As noted by Bruno and Leidecker (1988, p.51):

No two experts agree on a definition of business failure. Some conclude that

Because there are no formal reporting requirements for the majority of

Discontinuance (Sale or Closure)

This definition of failure is the least homogeneous. Fredland and Morris

studies, it is sometimes not possible to distinguish between change of own-

as a failure would clearly be inappropriate. For example, as noted earlier,

into question the use of ‘business closure’ as a meaningful measure of busi-

objective (verifiable) and simple measure of business failure and is cer-

Disposed of to Prevent Further Losses

Failing to ‘Make a Go of It’

Cochran (1981, p.52) suggested that ‘failure should mean inability to

cost of independence, making it difficult (and possibly even inappropriate)

2.3 AN EXAMINATION OF THREE CASE STUDIES

The following examples, based on real businesses from my local com-

Example 1: Watch Repair and Jewellery Shop

A Swiss couple founded this business soon after they immigrated to

Example 3: Fruit and Vegetable Shop

These three examples illustrate some of the difficulties inherent in attempt-

Table 2.1 Classification of business as failed/non-failed under the various

Definition of Failure Watch repair Bookshop Fruit and Overall

2.4 ESTIMATED FAILURE RATES UNDER

As noted earlier, lack of data is a major problem in trying to conduct

Table 2.2 Reasons for sale or closure of businesses located within

Reason for Sale/Closure Number Percent

Source: Adapted from Watson and Everitt (1996, Table 2).

classified as ‘failures’; a result that was subsequently supported by Headd’s

Reason for Definition of Failurea

Definition of Failure Years Since Start-up Cum. Average

more detailed information, business closure might provide a useful proxy

2.5 FAILURE RATE COMPARISONS

Most of the studies on business bankruptcy do not provide any infor-