SME Performance: Separating Myth from Reality

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

02

Printed and bound by MPG Books Group, UK

Contents PART I 1

BACKGROUND

Introduction

PART II 2 3

9 10

11

GROWTH FINANCING FOR SMES 69 87

NETWORKING AND SME PERFORMANCE

The association between networking and performance Networking: comparing female- and male-controlled SMEs

PART VI

31

47 53 59

A qualitative analysis A quantitative analysis

PART V

11

COMPARING THE PERFORMANCE OF FEMALE- AND MALE-CONTROLLED SMES

Failure rates Relating outputs to inputs Adjusting for risk

PART IV 7 8

SME PERFORMANCE

Defining and measuring SME failure/success The effects of age, size, industry and the economy on SME failure rates

PART III

4 5 6

3

101 116

CONCLUSIONS

Conclusions, implications and areas for future research

References Index

133 140 153

v

PART I

Background

1. 1.0

Introduction 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 following 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 eventually closed their businesses only when they felt it was time to retire. Were these owners failures because they eventually retired and closed their businesses? 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 onethird 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 demonstrate that failure rates could be reduced, and performance improved, if SME owners sought (and acted on) appropriate advice from professional 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 performance 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 maleowned SMEs (when appropriate adjustments and controls are incorporated 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 disadvantaged, 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 subsequent 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 riskadjusted 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 performance (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 providers 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 excessively high failure rates. Chapter 2 looks at how we might consider defining 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 limitations with the definition being used. Otherwise, as noted in Chapter 1, inappropriate policies or actions might be adopted to the potential detriment of SME owners and the future health of the economy. For example, if we define failure as bankruptcy proceedings being initiated 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 opportunity 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 depending 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. 2.0

Defining and measuring SME failure/success INTRODUCTION

How to adequately assess SME failure and success has long been a controversial 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 necessarily 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 performance 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 measuring and analysing business failure and success. In particular, the following 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 subjective 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 businesses are more likely to discontinue. Therefore, it is important (if possible) 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, therefore, 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 information 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 discontinuance (sale or closure) of a business and the initiation of formal bankruptcy 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 opportunities’. 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 ownership 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 discontinuances (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 business 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 MañezCastillejo (2008). These studies reported average annual failure rates varying from 3% to 17%. Interestingly, even though this is a much narrower 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 performing 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

Defining and measuring SME failure/success

15

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 business 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 necessarily 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 partnerships are treated differently from companies. Further, it is difficult to see how recording as failures all businesses that are closed or sold, irrespective 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 homogeneous 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 certainly 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 determine 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 (thirdparty) 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

Defining and measuring SME failure/success

17

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 (particularly 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 definitions (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 termination 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 definition 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 interpreting 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 community, 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 successful 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 sufficient 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 business 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 attempting 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 discontinuance 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

Defining and measuring SME failure/success

Table 2.1

19

Classification of business as failed/non-failed under the various definitions of failure

Definition of Failure

Watch repair and jewellery shop

Bookshop

Fruit and vegetable shop

Overall failure rate (%)

Discontinuance (of ownership and/or business) Business closurea Bankruptcy Disposed of to prevent further losses Failing to ‘make of go of it’

Failed

Failed

Not Failedb

66

Failed Not Failed Not Failed

Not Failed Failedc Not Failed

Not Failedb Not Failed Not Failed

33 33 0

Not Failed

Not Failed

Failedd

33

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, researchers 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 businesses 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

Defining and measuring SME failure/success

Table 2.2

21

Reasons for sale or closure of businesses located within Australian managed shopping centres 1961–90

Reason for Sale/Closure

Number

Percent

179 415 267 126 916 277 34

3.4 8.0 5.1 2.4 17.6 5.3 0.7

2214 329

42.6 6.3

Total sale or closures Continuing businesses

2543 2653

48.9 51.1

Total start-ups

5196

1. 2. 3. 4. 5. 6a. 6b.

Bankruptcy To avoid further losses Did not ‘make a go of it’ Retirement or ill health To realize a profit Other – not faileda Other – failed

7.

Unknown

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 discontinuances (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 definition 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

22

Bankruptcy To avoid further losses Did not make ‘a go of it’ Retirement or ill health To realize a profit Other – not failed Other – failed

Totals

7. Unknown

1. 2. 3. 4. 5. 6a. 6b.

Reason for Sale/Closure

594

594

179 179

179 415

To Prevent Further Losses

179

Bankruptcy

1061

895 166b

34

179 415 267

Failed to ‘Make a Go of It’

2543

2214 329

179 415 267 126 916 277 34

Discont. of Ownership (sale or closure)

Definition of Failurea

1002

836 166

114 270 162 37 152 78 23

38 50

64 65 61 29 17 28 68

Business Closure (%)

Table 2.3 Analysis of reasons for SME sale or closure grouped by failure definition, 1961–90, in Australian managed shopping centres

23

3.4 0.7 72

% of all businesses (n=5196) Average annual failure rate (%) Businesses < 5years old (%)

11.4 2.3 76

23 20.4 4.1 75

42 49 9.4 75

100

19 3.9 66

39

Source:

Adapted from Watson and Everitt (1996, Tables 3 and 7).

Notes: 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.

7

% of Discontinuances

24

Table 2.4

SME performance

SME failure rates within first five years of start-up, 1961–90a (%)

Definition of Failure

Bankruptcy To prevent further losses Failed to ‘make a go of it’ Discontinuance Closure of business

Years Since Start-up

Cum. Average 5 Year Annual Failure Failure Rate Rate

0–1

1–2

2–3

3–4

4–5

0.7 1.6

1.1 2.5

1.0 3.3

0.7 2.4

0.4 2.7

3.8 11.9

0.8 2.5

3.7

6.2

7.7

5.0

6.5

26.0

5.8

5.9 2.5

10.5 3.9

14.4 6.0

12.2 4.5

15.1 5.8

46.3 20.7

11.2 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 discontinued 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 misinterpretation 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 operation 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 approximately 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

Defining and measuring SME failure/success

25

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 shopping centre failed within five years, or, alternatively, almost 80% of businesses 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 inefficient 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 comparative 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 information 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 shopping 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 discontinued businesses; and for Australian SMEs located within managed shopping 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 bankruptcy. 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 comprehensive studies that focused on business closure. The first study is a large-scale longitudinal survey of Australian employing 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, forestry and fishing; electricity, gas and water supply; communication services; government administration and defense; education; and health and community services. Data collection was through self-administered questionnaires 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 confidentiality 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 businesses (representing approximately 1.25% of eligible Australian SMEs)

Defining and measuring SME failure/success

Table 2.5

27

Australian (ABS) SME closure rates 1995–98

Details of Business Closures

No.

Businesses active in 1994–95 Businesses closed in 1995–96 % of businesses closed

5030 487 10

Businesses active in 1995–96 Businesses closed in 1996–97 % of businesses closed

4543 337 7

Businesses active in 1996–97 Businesses closed in 1997–98 % of businesses closed

4206 339 8

Businesses active in 1997–98

3867

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 comprehensive 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 longitudinal study undertaken by Box (2008) that included 2154 Swedish jointstock 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 expansion) 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 businesses. Forsyth (2005) reported a 60% four-year survival rate on average for all rural firms. This survival rate is reasonably similar to the 50% fouryear 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 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’. Table 2.6 provides a summary of the bankruptcy and business closure

Defining and measuring SME failure/success

Table 2.6

29

Summary bankruptcy and closure rates from selected studies

Author(s)

Watson and Everett (1996)

ABS Data (unpublished) Hudson (1997)

Headd (2003)

Country/Sample

Australia: 5196 SMEs located within managed shopping centres Australia: 5014 randomly selected SMEs UK: All firms from 1950–90 US: All firms from 1950–90 US: 12 185 new firms

Forsyth (2005)

US: 4103 new small rural firms

Box (2008)

Sweden: 2154 joint-stock companies Belgium: Registered small firms 1986–2002

Dewaelheyns and Van Hulle (2008)

Reported Failure Rates Bankruptcy

Closure

All firms: 0.7% p.a.

All firms: 3.9% p.a.

All firms: 8% p.a.

Approx. 1% p.a. Approx. 0.7% p.a. Employing firms: <4yrs: 16% p.a. (estimated) All firms <4yrs: 10% p.a. Employers: 5.75% p.a. Non-employers: 11% p.a. Firms <4yrs: 6.25% p.a. Approx 1.8% p.a.

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 consistency 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 prevention 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 necessarily 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 necessarily 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, questioned 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 definition of failure used.

3. 3.0

The effects of age, size, industry and the economy on SME failure rates 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

Probability of failure (%)

14 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 gradually 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 afterwards’. This mortality process has been labelled the ‘liability of adolescence’, 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

Table 3.1

33

Average annual failure rates by age classification (%) Firm Age (years) 0–5

Bankruptcy Prevent further losses Failed to ‘make a go of it’ Discontinuance of ownership Discontinuance (closure) of business

0.8* 2.9** 5.0** 10.4 4.3*

5–10 0.6 1.8** 3.2** 9.8 4.0

10+

Av.

0.3** 1.0** 2.0** 6.9** 2.6**

0.7 2.4 4.2 10.0 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 consecutive 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 government 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

Businesses active in 1994–95 Businesses closed in 1995–96 % of businesses closed

1563 204 13**

1224 124 10

2243 159 7**

5030 487 10

Businesses active in 1995–96 Businesses closed in 1996–97 % of businesses closed

1359 213 16**

1100 68 6

2084 56 3**

4543 337 7

Businesses active in 1996–97 Businesses closed in 1997–98 % of businesses closed

895 195 22**

1164 58 5**

2147 86 4**

4206 339 8

16**

7**

5**

8

a

Average annual closure rate (%)

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

Effects of age, size, industry and the economy

Table 3.3

35

Australian SME closure rates 1995–96 to 1997–98 by size (%)

Firm Age

Firm Size (employees) 1–20

< 5 years 5–10 years > 10 years Average annual closure rate

16 7 5 9

>20 16 7 4* 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 relationship 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 difference in the average annual closure rates of the small versus the mediumsized 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 appropriate) 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 business 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 typically 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 investigation 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 discontinuance 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 justified 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 consider 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.

Effects of age, size, industry and the economy

Table 3.4

37

Comparing failure rates for SMEs located in larger and smaller shopping centres (%) Shopping Centre Size

Bankruptcy To prevent further losses Failed to ‘make a go of it’ Discontinuance Closure of business

Larger

Smaller

0.8 2.4 4.1 9.0 3.7

0.6* 2.2 4.1 10.2* 4.3*

Notes: * Significantly different from average failure rate in larger shopping centres at 5% using a two-tailed test. Source:

3.3

Adapted from Watson and Everett (1996b, p.279).

THE EFFECT OF INDUSTRY

Many studies examining the incidence of SME failure have reported significant 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 association 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 businesses with high start-up costs and against businesses with low start-up

SME performance High

38

Failure rate

Bankruptcy

Failed to ‘make a go of it’

Low

Business closure

Low Source:

Start-up costs

High

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 appropriate 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

Effects of age, size, industry and the economy

Table 3.5

Significant relationships between macro-economic indicators and SME failure Interest rate (current)

Bankruptcy Positive To prevent further losses Failed to ‘make a go of it’ Discontinuance Business closure Source:

39

Employment rate (lagged 6 months)

Retail sales (current)

Retail sales (lagged 6 months)

Positive

Negative

Positive Positive

Negative Negative

Unemployment rate (lagged 6 months)

Positive

Positive

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 summarized in Table 3.5. The table reports only those macro-economic variables 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 businesses with substantial borrowings do fail, they have a high probability of being placed into bankruptcy. The positive association between interest rates and bankruptcy supports the earlier findings by Hall (1986) and Wadhwani (1986). The six-month lagged employment rate is the only significant macroeconomic variable associated with businesses that are sold (or closed) to prevent further losses. However, the direction of the association (positive) is not as one might expect. It appears that the probability of failure increases with increases in the lagged employment rate; that is, as employment 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 significantly 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 opportunity 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 prediction 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.

Effects of age, size, industry and the economy

41

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 unemployment 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) indicate 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 unemployment (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’; discontinuance of ownership; or discontinuance of business). These findings are consistent 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, voluntarily, 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 associated 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 understanding 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 particular: 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 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 smaller, compared to larger 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 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 folklore 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, industries 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 speculation, on differences between male and female entrepreneurs and their businesses.’ Studies examining and comparing the performance of maleand female-controlled businesses have generally reported that femalecontrolled 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 discriminated against (by lenders, for example) and/or because of other systematic 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 different 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 differences 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 explanation 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, construction, and manufacturing’. Similarly, Rosa, Carter and Hamilton (1996) argued that at least some of the gender difference in business performance 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 malecontrolled 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 business. 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 definitions 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 flexibility (particularly if they have family responsibilities) than male owners (Brush 1992; Rosa et al. 1994). Unfortunately, even after controlling for certain demographic differences, 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 femalecontrolled SMEs. I trust that by the end of these chapters, the reader will be convinced that female-controlled SMEs do not under-perform malecontrolled 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 underperformance 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 opportunity 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 fouryear period 1995–1998, as discussed in Chapter 2.

4.1

COMPARING CLOSURE RATES FOR MALEAND 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 Businesses closed in 1995–96 % of businesses closed

2868 252 9

178 18 10

3046 270 9

Businesses active in 1995–96 Businesses closed in 1996–97 % of businesses closed

2616 185 7

160 20 13

2776 205 7*

Businesses active in 1996–97 Businesses closed in 1997–98 % of businesses closed

2431 166 7

140 9 6

2571 175 7

Businesses active in 1997–98

2265

131

2396

Average annual closure rate

8

10

8*

Note: * Female closure rate significantly higher than male closure rate at 5% using a onetailed 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 malecontrolled 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

Firm age (%) Less than 2 years 2 years to less than 5 years 5 years to less than 10 years 10 years to less than 20 years 20 years and older Total

Female Controlled n = 178

13 15 23 27 21

22** 22** 21 24 11**

100

Firm size in 1994–95 Number of employees Note: ** Source:

Male Controlled n = 2868

27

100 13**

Significantly different to male-controlled SMEs at 1% using a two-tailed test.

ABS.

50

SME performance

Table 4.3

Female-controlled SMEs by industry with industry closure rates (%)

Industry Construction Wholesale trade Mining Manufacturing Finance and insurance Property and business services Retail trade Transport and storage Cultural and recreational services Accommodation, cafes and restaurants Personal and other services Average Source:

% Female Controlled

Closure Rate

1 3 4 5 7 8 9 10 13 17 33

22 17 38 21 27 25 28 21 28 31 29

7

23

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 significant age and size differences it is not possible to draw informed conclusions 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 underrepresented 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% compared 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 definition 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 industry, 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 incorporating 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 significant. 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

Table 4.4

SME performance

Results of logistic regression examining SME closure against, size, industry and sex of major decision maker

Sex of major decision maker Firm age Less than 2 years 2 years to less than 5 years 5 years to less than 10 years 10 years to less than 20 years Firm Size Industry Mining Manufacturing Construction Wholesale trade Retail trade Accommodation, cafes & restaurants Transport & storage Finance & insurance Property & business services Cultural & recreational services Constant

B

S.E.

Wald

df

Sig.

Exp(B)

−0.02

0.20 0.16 0.17 0.17 0.16 0.00

0.49 0.01 −0.18 −0.36 0.10 0.36

0.64 0.35 0.39 0.37 0.37 0.44

0.94 0.00 0.00 0.00 0.09 0.39 0.84 0.01 0.44 0.97 0.65 0.33 0.79 0.41

0.99

2.54 0.70 0.28 0.14 0.00

0.01 1 392.91 4 237.96 1 16.45 1 2.81 1 0.73 1 0.04 1 23.63 10 0.60 1 0.00 1 0.21 1 0.97 1 0.07 1 0.68 1

−0.67 0.57 −0.09 −0.51 −1.98

0.47 0.40 0.36 0.53 0.38

2.09 2.08 0.06 0.94 26.35

1 1 1 1 1

0.15 0.15 0.80 0.33 0.00

12.61 2.02 1.32 1.15 1.00 1.64 1.01 0.84 0.70 1.10 1.43 0.51 1.78 0.91 0.60 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. 5.0

Relating outputs to inputs 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 businesses 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 measures. For example, Rosa, Carter and Hamilton (1996) reported that businesses 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 measures 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

5.1

SME performance

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: ROE 5

profit 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 leverage, as shown below: ROE 5

profit assets 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 separate 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 femalecontrolled SMEs, will be explored further.

5.2

COMPARING MALE- AND FEMALECONTROLLED 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

Table 5.1

55

Comparing the mean annual outputs and inputs for male- and female-controlled SMEs

Output / Input Measures ($)

Mean

Sig. Level

Male

Female

Output measures Total income Profit

6 158 168 354 181

1 661 729 22 314

0.000 0.000

Input measures Total assets Owner’s equity

4 432 502 1 509 569

734 070 158 866

0.000 0.000

Source:

Adapted from Watson (2002, Table 2).

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 femalecontrolled SMEs. While these results indicate that, on average, malecontrolled SMEs are larger than female-controlled SMEs, they do not indicate that the former outperform the latter. Before we can appropriately 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 maleand female-controlled SMEs. However, an examination of the histograms for each of these performance measures (with a normal curve fitted) suggests 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 measured 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

All firms Total income to total assets (TITTA) Profit to total assets (ROA) Profit to total equity (ROE) Log of total income to total assets (Log TITTA) Profitable firms Profit to total assets (ROA) Log of profit to total assets (Log ROA) Profit to total equity (ROE) Log of profit to total equity (Log ROE) Loss making firms Absolute loss to total assets (ABROA) Log of absolute loss to total assets (Log ABROA) Absolute loss to total equity (ABROE) Log of absolute loss to total equity (Log ABROE) Source:

Mean

Sig. Level

Male

Female

10.5392 0.3744 0.7824 1.1583

10.1048 0.5478 1.8938 1.2406

0.960 0.197 0.334 0.142

0.5398 −1.6369 1.0139 −0.6930

0.8339 −1.3604 2.5461 −0.5015

0.048 0.000 0.321 0.005

0.5902 −2.0283

0.6373 −1.9877

0.920 0.759

1.7095 −0.9210

1.3829 −1.0729

0.674 0.248

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 femalecontrolled 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

Table 5.3

ANOVA of log ROA and log ROE and various business demographics

Source of Variation Log of ROA Industry Firm age Days business operated Sex of owner Log of ROE Industry Firm age Days business operated Sex of owner Source:

57

Sum of Squares

DF

Mean Square

F

Sig. of F

348.40 390.96 28.94

10 4 1

34.84 97.74 28.94

14.96 41.97 12.43

0.00 0.00 0.00

10.97

1

10.97

4.71

0.03

143.24 494.23 0.59

10 4 1

14.32 123.56 0.59

7.06 60.90 0.29

0.00 0.00 0.59

1.67

1

1.67

0.82

0.36

Adapted from Watson (2002, Tables 5 and 6).

following analysis of firm performance, the SMEs in the sample were classified 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.

5.3

COMPARING MALE- AND FEMALECONTROLLED 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 performance 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 controlling for various demographic variables, there is no difference in the performance of male- and female-controlled SMEs. Interestingly, if we use a 5% cut-off, the results suggest that the female-controlled SMEs outperformed 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 femalecontrolled 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 femalecontrolled SMEs do not underperform male-controlled SMEs. Indeed, there is some evidence to suggest that, with respect to ROA (asset management), female-controlled SMEs might outperform male-controlled SMEs.

6. 6.0

Adjusting for risk 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 substantial 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 propensities 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 performance 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 investors 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 experimental study, asked a sample of entrepreneurs leading America’s fastest growing firms to make choices among a series of hypothetical new ventures. 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 performance 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 reasonable 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

6.2

61

COMPARING MALE AND FEMALE ATTITUDES TO RISK

Barber and Odean (2001) found that for different risk measures (portfolio 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 financial 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 strategies 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 concerned 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

6.3

SME performance

COMPARING MALE- AND FEMALECONTROLLED 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 femalecontrolled 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 histograms 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 Average annual profit (’000) Log of average annual profit Standard deviation of annual profit (’000) Log of standard deviation of annual profit Sharpe ratio Log of Sharpe ratio

Male N=2236 448 4.4 356 4.2 1.7 0.2

Female N=131 63** 3.3** 83** 3.0** 1.6 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).

Adjusting for risk

63

are related to the standard deviation in those profits, using the Sharpe ratio (Sharpe 1975), no difference exists between the male- and femalecontrolled 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.

6.4

COMPARING MALE- AND FEMALECONTROLLED 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 explaining 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 Squares

DF

Mean Square

F

Sig of F

Log of average annual profit Firm age Industry No. of employees Sex of owner

25.68 144.88 1963.80 24.22

4 10 3 1

6.42 14.49 654.60 24.22

2.96 6.69 302.18 11.18

0.02 0.00 0.00 0.00

Log of st. dev. of annual profits Firm age Industry No. of employees Sex of owner

8.94 197.46 2327.81 21.70

4 10 3 1

2.24 19.75 775.94 21.70

1.39 12.26 481.64 13.47

0.24 0.00 0.00 0.00

4.91 23.04 6.69 0.36

4 10 3 1

1.23 2.30 2.23 0.36

0.99 1.87 1.81 0.29

0.41 0.05 0.14 0.59

Log of Sharpe ratio Firm age Industry No. of employees Sex of owner 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 findings reported in Table 6.1 and again support the proposition that there is no significant difference in the performances of male- and femalecontrolled SMEs, provided performance is measured appropriately. In other words, although females may adopt different strategies in founding, 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 appropriately, and adequate controls are included in the analysis, male- and female-controlled SMEs perform equally well.

Adjusting for risk

65

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 previous 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 appropriate 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 institutions (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 employment, 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 significantly 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 external 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. 7.0

A qualitative analysis 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 consciously 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 structure 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 thresholds are smaller than those set by their male counterparts. Female entrepreneurs 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 performance (Cooper 1993), this chapter reviews two studies designed to help better understand the various demand-side issues relating to the financing 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 creditconstrained 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 N=18

Discouraged N=8

Total N=26

No.

(%)

No.

(%)

No.

(%)

Operational issues – lack of appropriate staff, facilities, time, supplies and expertise/knowledge in assessing strategy, market size, competition and how to go about growing the business

60

(59)

18

(39)

78

(53)**

Funding issues – lack of internal cash flow or external funds at reasonable rates and terms

24

(24)

11

(24)

35

(24)

Government regulation intrusion, lack of support, fees and taxes

7

(7)

15

(33)

22

(15)**

Fear of failure/unsure of benefits

9

(9)

1

(2)

10

(7)

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 introduction, both groups agreed that the major issue centred on the risks involved. These included the obvious concern about the potential for interest 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

A qualitative analysis

Table 7.2

73

Reasons SME owners might choose not to seek external funding

Reason

Successful N=18

Discouraged N=8

No.

(%)

No.

(%)

No.

(%)

Adequate internal funds

18

(28)

4

(15)

22

(24)

Risk/uncertainty about future, potential to lose control, fear of failure

11

(17)

5

(19)

16

(17)

9

(14)

5

(19)

14

(15)

Terms unacceptable – interest rates, security, etc.

10

(15)

3

(11)

13

(14)

Expand (don’t want to) – can’t cope, too many hassles, lack confidence

1

(2)

7

(26)

8

(9)**

Time and aggravation of trying to get a bank loan

6

(9)

2

(7)

8

(9)

Unavailable because of poor/no credit rating

4

(6)

1

(4)

5

(5)

Other

6

(9)

0

0

6

(7)

Total

65

(100)

27

92

(100)

Burden (don’t want) – can’t service more debt

(100)

Total N=26

Notes: Figures do not always add up to 100% because of rounding. ** ‘Discouraged’ group significantly different from ‘successful’ group at 1% using a twotailed test. Source:

Watson, Newby and Mahuka (2006, Table 9.8).

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 (business plan and track record); and the perceived risks (including lack of

74

SME performance

Table 7.3

Main disadvantages of debt funding

Reason

Successful N=18

Discouraged N=8

Total N=26

No.

(%)

No.

(%)

No.

(%)

Risk – interest rate changes, can’t repay debt, spend too much

26

(47)

12

(43)

38

(46)

Burden – worry, work, stress, paperwork

12

(22)

4

(14)

16

(19)

Dislike of financial institutions – terms, fees, disclosures

4

(7)

4

(14)

8

(10)

Control may be lost – dependence on others is increased

2

(4)

4

(14)

6

(7)**

Costs

5

(9)

0

(0)

5

(6)

Other

6

(11)

4

(14)

10

(12)

Total

55

(100)

28

(100)

83

(100)

Notes: Figures do not always add up to 100% because of rounding. ** ‘Discouraged’ group significantly different from ‘successful’ group at 1% using a twotailed test. Source:

Watson, Newby and Mahuka (2006, Table 9.9).

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 ‘discouraged’ 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 existence of a ‘finance gap’.

A qualitative analysis

Table 7.4

75

Reasons for bank refusing a loan application

Reason

Successful N=18

Discouraged N=8

Total N=26

No.

(%)

No.

(%)

No.

(%)

Business plan – don’t have one or not convincing or bank doesn’t understand

23

(29)

12

(31)

35

(29)

Track record – poor credit rating or don’t have one

19

(24)

9

(23)

28

(24)

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. Source:

Watson, Newby and Mahuka (2006, Table 9.10).

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 participant 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 participants 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 N=18 No.

1 2 3 4 5 6 7 Mean Std Dev

14 2 1 – – – 1 1.63 1.54

Discouraged N=8

(%)

No.

(78) (11) (6) – – – (6)

8 – – – – – – 1.00 0.00

Total N=26

(%)

No.

(%)

(100) – – – – – –

22 2 1 – – – 1 1.38 1.24

(85) (8) (4) – – – (4)

Notes: Figures do not always add up to 100% because of rounding. (1 = very important, 7 = unimportant) Source:

Watson, Newby and Mahuka (2006, Table 9.11).

Table 7.6

Comparing male and female SME owners’ attitudes to risk

Attitude to Risk Women more conservative Men more conservative No difference Depends on personality Total

Males

Females

Total

4 0 0 2 6

3 1 2 1 7

7 1 2 3 13

Note: Where a participant spoke on more than one occasion, his/her view was only recorded once. Source:

Watson, Newby and Mahuka (2006, Table 9.12).

the firm were unlikely to want to raise additional funding for fear of overcapitalizing 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 ‘discouraged’ group. It is difficult to know on what basis they formed their views (perhaps it was from their previous experiences or, alternatively, from

A qualitative analysis

Table 7.7

77

Factors likely to influence an SME owner’s decision to access external funding

Key Thoughts (Ideas) Growth: Without profit it isn’t worth it Can’t just decide to grow – there are competitors Business conditions: Less likely to borrow if business is volatile and control could be lost If firm is failing you might borrow to try and save the business Exiting the firm: Additional funding isn’t needed if you are planning to exit the firm Unless you want to build the business up ready for sale Or have someone (whom you trust) willing to buy in Although finding that trustworthy person isn’t easy I would prefer to sell the business to my staff I would be prepared to offer key personnel a stake in the business I am planning to hand the business over to my kids Accessing debt funding: As you get older it is harder to get a loan (repayments) As you get older it is easier to get a loan (security) As you get older you may not want to borrow Age makes no difference Being married helps when you borrow But for women the husband must often co-sign Banks want security Banks don’t lend to businesses they don’t understand Banks will lend you money when you don’t need/want it Bigger firms find it easier to borrow Bigger loans attract lower interest rates More likely to borrow to modernize equipment But over-capitalizing isn’t helpful Total Source: Watson, Newby and Mahuka (2006, Table 9.13).

No. of times mentioned 4 2 2 2 4 1 1 2 1 1 1 1 1 1 2 1 1 2 2 1 1 1 1 1 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 virtually 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 demandside 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

A qualitative analysis

Table 7.8

79

Application rate and success of application by sex of owner Sex of Owner Female

Applied for funding in last 3 years?

Male

13 54% 55 11 46% 43 24 98 Pearson c2 = 0.030, p value = 0.863

56% 44%

Application for funding successful? Yes 11 85% 46 No 2 15% 6 Total 13 52 Pearson c2 = 0.143, p value = 0.706

88% 12%

Total

Yes No

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) concludes ‘that most SMEs are getting the finance they want’, this suggests that it is also unlikely (given the similar rejection rates) that any substantial 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: 1 = Strongly disagree to 7 = Strongly agree The firm had sufficient funds under its existing arrangements The firm did not need additional funds Procedures to obtain funding from a bank were too complicated Interest rates were too high It was unlikely the bank would provide the full amount A previous loan application was rejected Source:

Mean scores for

2-tailed p value

Females

Males

6.75

5.77

0.08

6.10

5.73

0.48

2.38

3.82

0.11

1.50 1.00

2.87 2.53

0.01 0.00

1.00

1.42

0.41

Watson, Newby and Mahuka (2009, Table 3).

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 arrangements; 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, compared 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,

A qualitative analysis

Table 7.10

81

Funding terms Mean scores for Females

How long did it take the bank to approve the loan (days) Term that applied to funding (months) Interest rate at inception (per annum) Interest rate currently (per annum) Source:

19 119 7.41% 8.88%

21

0.85

56 8.35% 8.50%

0.22 0.50 0.76

Watson, Newby and Mahuka (2009, Table 4).

Table 7.11

Level of satisfaction with lending institution

7 point scale: 1 =Totally dissatisfied to 7 = Totally satisfied The amount granted by the bank relative to the amount requested Customer service The time taken to process the application The interest rate charged The guarantees required The security required The service fees charged Source:

Males

2-tailed p value

Mean scores for

2-tailed p value

Females

Males

6.64

6.48

0.97

4.73 5.00

5.22 4.96

0.41 0.95

4.55 5.00 4.82 3.82

4.60 4.16 4.15 4.04

0.94 0.19 0.31 0.72

Watson, Newby and Mahuka (2009, Table 5).

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 provides 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

7 point scale: 1 = Strongly Disagree to 7 = Strongly Agree Male applicants are more likely to have their application approved Female applicants are more likely to have to provide security Female applicants are more likely to need loan guarantors Female applicants are more likely to incur higher interests rates Source:

Mean scores for

2-tailed p value

Females

Males

3.83

3.84

0.99

3.46

3.72

0.58

3.33

3.66

0.48

2.50

2.91

0.25

Watson, Newby and Mahuka (2009, Table 6).

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 experienced by both the females and males with respect to their lending institution 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 majority 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 applying for funding from a financial institution. These responses suggest that neither the female nor the male respondents believed women suffer any

A qualitative analysis

Table 7.13

83

Comparing the personal characteristics of SME owners

7 point scale: 1 = Low to 7 = High

Risk-taking propensity (av. of 10 items)a Internal locus of control (av. of 8 items)b

Mean scores for Females (Non-Applicants)

Males (Applicants)

2.85 (3.00) 5.01 (5.08)

3.37 (3.47) 5.15 (5.16)

2-tailed p value

0.01 0.00 0.47 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). Source:

Watson, Newby and Mahuka (2009, Table 7).

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 consistent with the findings reported in Chapter 6, indicating that femalecontrolled 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 maintain control), the respondents were asked to assess the relevance of various factors in potentially discouraging them from applying for external

84

Table 7.14

SME performance

Factors discouraging SME owners from applying for funding in the future

7 point scale: 1 = Not relevant to 7 = Very relevant

Mean scores for Females Males (Non-Applicants) (Applicants)

2-tailed p value

Likelihood of unreasonable terms and conditions

5.13 (5.42)

5.40 (5.28)

0.54 0.70

Risk of not being able to repay the loan

5.08 (4.87)

4.39 (4.08)

0.12 0.04

Potential to lose control of the business

4.65 (4.15)

3.75 (3.60)

0.07 0.17

Amount of documentation required

3.96 (3.97)

4.18 (4.34)

0.62 0.28

Probability of not getting the loan

3.13 (2.93)

2.56 (2.30)

0.19 0.07

Apprehension over having the loan application rejected

2.91 (2.46)

2.05 (1.89)

0.02 0.05

Source: Watson, Newby and Mahuka (2009, Table 8).

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

A qualitative analysis

Table 7.15

Most likely use of surplus funds

7 point scale: 1 = Highly Unlikely to 7 = Highly Likely Repay debt Grow the current business Improve lifestyle Invest in other ventures

Source:

85

Mean scores for Females Males (Non-Applicants) (Applicants) 5.88 (5.71) 4.71 (5.09) 4.67 (4.16) 4.17 (4.45)

5.20 (4.89) 4.60 (4.06) 4.13 (4.34) 4.36 (4.15)

2-tailed p value

0.05 0.03 0.80 0.00 0.23 0.63 0.70 0.44

Watson, Newby and Mahuka (2009, Table 9).

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 indicates that SME owners have an aversion to borrowing funds from financial 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 femalecontrolled 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 discrimination 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. 8.0

A quantitative analysis 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 financing options where insufficient internal funds are available to take advantage 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 external 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 Source:

Firm age

Older

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 ‘informationally 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 femaleowned 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 internal (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 comparing 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 significantly 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 external) 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 controlled 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

A quantitative analysis

91

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 facilitate 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 malecontrolled 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

Average profits ’000 (median)

Growth

High

Low

High

Low

36

62**

69

30**

51 49

49 51

Debt-to-asset ratio (%) High (above the median) Low (below the median) Age of firm (%) Less than 10 years old (younger) 10 or more years old (older)

47 52

53 48

51 49

49 51

Industry (%) Mining Manufacturing Construction Wholesale trade Retail trade Accomodation, cafes & restaurants Transport & storage Finance & insurance Property & business services Cultural & recreational services Personal & other services

9 49 55 49 61 62 62 41 44 44 54

** 91 51 45 51 39 38 38 59 56 56 46

9 48 54 50 51 40 54 54 50 55 60

91 52 46 50 49 60 46 46 50 45 40

Type of legal organization (%) Incorporated (limited liability) Unincorporated (unlimited liability)

47 58

53 42

51 48

49 52

Sex of owner (%) Female Male

40 51

60 49

56 50

44 50

Education (%) School Trade Tertiary (non-business) Tertiary (business)

55 54 43 44

45 46 57 56

50 49 52 49

50 51 48 51

Years of experience (median)

14

13

14

14

Note: * Source:

*

**

*

**

Significantly different at 5%; ** Significantly different at 1%. Adapted from Watson (2006, Table 1).

A quantitative analysis

93

Table 8.2

Logistic regression models for SMEs with high/low growth

Variables

Model 1 Wald

Sig

Debt-to-asset ratio (dtar) Average profits (logged) 31.30 Firm age 2.99 Sex of owner 2.73 Industry 17.89 Mining 8.08 Manufacturing 2.35 Construction 0.47 Wholesale trade 2.60 Retail trade 0.85 Accom., cafes & restaurants 4.02 Transport & storage 0.01 Finance & insurance 0.13 Property & business services 1.22 Cultural & recreation services 1.17 Type of legal organization 3.17 Owner education 3.18 School 1.32 Trade 0.94 Tertiary (non-business) 3.14 Years of experience 0.03

0.00 0.08 0.10 0.06 0.00 0.13 0.49 0.11 0.36 0.05 0.92 0.72 0.27 0.28 0.08 0.37 0.25 0.33 0.08 0.87

Nagelkerke R Square

0.039

Model 2 Exp(B) Wald

Sig

Exp(B)

2.54 32.54 3.31 3.10 17.48 7.86 2.34 0.47 2.59 0.95 4.16 0.03 0.09 1.13 0.89 3.44 3.43 1.25 0.90 3.42 0.04

0.11 0.00 0.07 0.08 0.06 0.01 0.13 0.49 0.11 0.33 0.04 0.86 0.76 0.29 0.35 0.06 0.33 0.27 0.34 0.06 0.84

0.86 1.17 1.20 1.46

1.16 1.19 1.42 0.04 0.60 0.77 0.57 0.72 0.38 0.96 0.86 0.68 0.60 1.20 1.17 1.15 1.30 1.00

0.04 0.60 0.77 0.57 0.71 0.38 0.93 0.88 0.69 0.63 1.21 1.16 1.15 1.31 1.00

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)’. Source:

Watson (2006, Table 4).

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 association 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 manufacturing 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 highgrowth 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 significantly 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

76 54

1027 1204

16 12

25 26

10 5

13 16**

All firms

130

2231

15

26

8

15**

Note: **

Significantly different at 1% using a two-tailed test.

Younger (< 10 years) Older (≥ 10 years)

Source:

Watson (2006, Table 3).

A quantitative analysis

Table 8.4

Four-year growth rates for female- and male-controlled SMEs by age

Firm age

Number

Younger (< 10 years) Older (≥ 10 Years) All firms Note: * Source:

95

Mean (%)

Median (%)

Female

Male

Female

Male

Female

Male

77 54

1028 1208

25 41

59 26

10 18

11 8*

131

2236

32

41

17

9

Significantly different at 5% using a two-tailed test. Watson (2006, Table 5).

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 difference 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

8.6

SME performance

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 personal 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

A quantitative analysis

97

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 provision 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 networking 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 embedded 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, potentially 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 compelling, 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 association between the networking activities of SME owners and the performance of their firms. Chapter 9 examines the relationship between networking and SME performance generally, and then Chapter 10 specifically 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 networks of the SME owner, rather than on the organizational networks of the business (Brüderl and Preisendörfer 1998).

9. 9.0

The association between networking and performance 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 decision making but is costly to obtain and that networks provide a means by which important information can potentially be acquired in a costeffective 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 relationship 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 performance 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 advisers, 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 discussion of the three conceptualizations of social capital found in the literature. 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 networks 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 connected) 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 embedded 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 networks 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, business 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 question, namely, should SME owners attempt to have a very broad support network that they access on a limited basis, or should they have fewer supports 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 narrower 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 Business consultant External accountant Industry associations SBDC Solicitor/lawyer Tax office

38 72 20 60 84 43 59

35 18 35 21 13 34 31

27 10 45 19 3 23 10

Average for formal networks

54

27

19

Informal networks Family & friends Local businesses Others in the industry

65 74 45

19 16 29

16 10 25

Average for informal networks

61

22

17

Average for all networks

56

25

19

Source:

Watson (2007, Table 1).

Networking and performance

105

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 associations and family and friends being the most frequently accessed sources of advice. For example, 45% of SME owners accessed an external accountant 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 networking 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, consistent 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

Probability of Survival

1.0

0.9

0.8

0.7 Linear 0.6 –10

Source:

Quadratic 0

10 Networking Score

20

30

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 relationship between networking and both income growth and return on equity (ROE), respectively. To assess the relationship between networking and both firm growth and ROE, the analysis focuses on firms in the upper and lower quartiles for these two performance measures. Firms in

Networking and performance

107

0.58

Probability of High Growth

0.56 0.54 0.52 0.50 0.48 0.46 0.44 Linear

0.42 0.40 –10

Source:

Quadratic 0

10 Networking score

20

30

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 survival 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 variables. Table 9.2 provides the results of examining the relationship between

108

SME performance 0.54

Probability of High ROE

0.52

0.50

0.48

0.46

0.44 Linear 0.42 –10

Source:

Quadratic 0

10 Networking score

20

30

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 demographic 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 negatively 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,

109

B

Industry Mining Manufacturing Construction Wholesale trade Retail trade Accom., cafes, rest’s

−0.24 0.02 0.28 0.45 0.11 −0.10

−0.20 −0.04 0.14 0.38 0.06 −0.11

0.10

1.05

0.79 1.02 1.33 1.56 1.12 0.90

−0.28

−0.44

0.65*

0.70*

−3.34

B

0.82 0.96 1.15 1.46 1.06 0.90

1.10

0.76

0.64*

** 0.04**

Exp(B)

Model 2

** 0.04**

Exp(B)

Model 1

Survival

−0.47 −0.25 0.45 −0.18 −0.22 −0.66

0.22

0.28

0.41

0.96

B

* 0.63 0.78 1.57 0.83 0.81 0.52

1.24

1.33*

1.50*

** 2.62**

Exp(B)

Model 1 B

−0.43 −0.28 0.44 −0.21 −0.23 −0.66

0.23

0.28

0.40

* 0.65 0.76 1.55 0.81 0.80 0.52

1.26

1.32*

1.50*

** 2.63**

Exp(B)

Model 2

0.97

Growth

Logistic regression models of survival, growth and ROE against networking

Firm age Less than 2 years −3.21 old 2 years to less −0.43 than 5 5 years to less than −0.35 10 10 years to less 0.05 than 20

Variables

Table 9.2

−1.02 −0.70 −0.15 −0.56 −0.63 −0.69

0.37

0.34

0.40

0.47

B

** 0.36 0.50* 0.86 0.57 0.53* 0.50

1.45**

1.40*

1.49*

* 1.60*

Exp(B)

Model 1

0.36

0.33

0.39

0.47

B

** 0.35* 0.50* 0.87 0.58 0.53* 0.49*

1.43*

1.39*

1.47*

* 1.59*

Exp(B)

Model 2

−1.06 −0.70 −0.14 −0.55 −0.64 −0.71

ROE

110

1.00

0.327

0.00 1.45** 0.98**

1.00

1.20

1.17

0.80

1.22

0.392

0.37 −0.02

0.00

0.19

0.16

−0.22

0.20

1.00

0.82

1.03

0.90

0.78

Exp(B)

0.032

0.00

−0.19

0.03

−0.10

−0.24

B

Model 1 B

1.10** 1.00*

1.00

0.84

1.03

0.90

0.039

0.10 −0.01

0.00

−0.18

0.03

−0.11

0.78

Exp(B)

Model 2

−0.24

Growth

1.00

0.30**

0.73

0.60

0.39**

Exp(B)

0.024

0.00

−1.20

−0.32

−0.51

−0.95

B

Model 1

B

0.98 1.00

1.00

0.30**

0.72

0.60

0.027

−0.02 0.00

0.00

−1.22

−0.32

−0.51

0.38**

Exp(B)

Model 2

−0.96

ROE

Source:

Watson (2007, Table 3).

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’.

Nagelkerke R Square

Networking Networking2

Firm size

1.25

0.23

1.17

0.85

−0.17

0.16

1.27

0.24

Exp(B)

Model 2 B

Survival

Exp(B)

Model 1

B

(continued)

Transport & storage Finance & insurance Property & bus. serv. Cultural & rec. serv.

Variables

Table 9.2

111

** 0.04**

0.65*

0.76

1.09

0.78 0.92 1.12 1.39

−0.43

−0.27

0.09

Industry Mining −0.25 Manufacturing −0.08 Construction 0.11 Wholesale 0.33 trade

Exp(B)

−3.33

B

Model 3

−0.22 −0.03 0.06 0.40

0.07

−0.26

−0.46

−3.33

0.81 0.97 1.07 1.50

1.08

0.77

0.63*

** 0.04**

Exp(B)

Model 4 B

Survival

−0.46 −0.29 0.45 −0.23

0.23

0.30

0.43

0.98

B

** 0.63 0.75 1.56 0.80

1.25

1.35*

1.54**

** 2.67**

Exp(B)

Model 3 B

−0.42 −0.29 0.43 −0.23

0.22

0.27

0.40

0.66 0.75 1.54 0.80

1.25 **

1.31*

1.49**

** 2.64**

Exp(B)

Model 4

0.97

Growth

−1.06 −0.70 −0.15 −0.55

0.36

0.33

0.38

0.46

B

** 0.35* 0.50* 0.86 0.58

1.43*

1.39*

1.47*

* 1.59*

Exp(B)

Model 3

0.36

0.33

0.40

0.47

B

** 0.35* 0.50* 0.88 0.57

1.43*

1.39*

1.48*

* 1.60*

Exp(B)

Model 4

−1.06 −0.69 −0.13 −0.56

ROE

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

Firm age Less than 2 years old 2 years to less than 5 5 years to less than 10 10 years to less than 20

Variables

Table 9.3

112

1.11

1.21

0.11

0.19

1.59** 0.97**

1.00

0.79

−0.24

0.00

1.20

0.18

Formal networks 0.46 Formal networks2 −0.03

Firm size

1.04 0.88

Exp(B)

0.00

0.23

0.10

−0.25

0.20

0.06 −0.18

1.00

1.26

1.10

0.78

1.22

1.06 0.83

Exp(B)

Model 4 B

Survival

0.04 −0.13

B

Model 3

(continued)

Retail trade Accom., cafes, rest’s Transport & storage Finance & insurance Property & bus. serv. Cultural & rec. serv.

Variables

Table 9.3

0.13 −0.01

0.00

−0.15

0.02

−0.11

−0.25

−0.22 −0.65

B

1.14** 0.99*

1.00

0.86

1.02

0.90

0.78

0.81 0.52

Exp(B)

Model 3 B

0.00

−0.17

0.02

−0.11

−0.25

1.00

0.84

1.02

0.90

0.78

0.79 0.51

Exp(B)

Model 4

−0.24 −0.68

Growth

−0.03 0.00

0.00

−1.22

−0.32

−0.51

−0.96

−0.64 −0.71

B

0.98 1.00

1.00

0.29**

0.73

0.60

0.38*

0.53* 0.49

Exp(B)

Model 3

B

0.00

−1.23

−0.33

−0.52

−0.97

1.00

0.29**

0.72

0.60

0.38**

0.53* 0.49

Exp(B)

Model 4

−0.63 −0.71

ROE

113

0.392

0.406

8.92** 0.52**

1.04 1.00

1.01

0.94

0.041

0.01

−0.07

0.78 1.09

1.26** 0.98*

0.041

−0.25 0.09

0.23 −0.02

1.00

1.01

0.027

−0.01

0.01

0.93 1.06

0.95 1.00

0.027

−0.07 0.06

−0.06 0.00

Source:

Watson (2007, Table 4).

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’.

Nagelkerke R Square

2.19 −0.65

Network intensity Network intensity2

0.98

−0.02 0.04 0.00

1.20*

0.19

Network range Network range2

Informal networks Informal networks2

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 supports 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

Networking and performance

115

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 providing support for the assertion by Fischer and Reuber (2003) that owners of rapid-growth firms should be interested in (and should support) government policy aimed at developing a network-based approach to facilitating firm growth. Having explored the relationship between networking and SME performance for all firms, the following chapter will specifically look at possible networking differences in male- and female-controlled SMEs.

10. 10.0

Networking: comparing femaleand male-controlled SMEs 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 different 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 professional 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 (particularly family and friends). Although the literature, and the vast majority of past research, indicates 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 business, 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 incorporating 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 10 9 8 7 6 5 4 3 2 1 0 Note: Source:

Number of networks accessed by male and female SME owners Male n = 2919

Female n = 181

%

Cum %

%

Cum %

3 5 8 12 11 13 12 11 7 6 12

3 8 16 28 39 52 64 75 82 88 100

2 4 5 8 14 12 12 9 11 7 16

2 6 11 19 34 46 58 67 77 84 100

Chi-Square test comparing males and females not significant at 5%. 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 associations 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

Networking: female- and male-controlled SMEs

Table 10.2

119

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 Bank Solicitor Industry association Business consultant Tax office SBDC

19 36 41 57 71 58 84

20 44 48 75 82 65 87

34 36 35 23 19 32 13

36 39 40 15 13 30 12

47 28 24 20 10 10 3

44 18** 12** 10** 5** 6 1

Av. formal

52

60

27

26

20

13*

Informal Others in the industry Family & friends Local businesses

44 63 73

48 52 75

30 20 17

26 23 15

27 17 10

27 25** 9

Av. informal

60

58

22

21

18

20

Av. all networks

55

60

26

25

20

16

Note: *, ** respectively. Source:

Chi-Square test significantly different for males and females at 5% and 1%,

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 significant 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 accountant on more than three occasions during the year. This finding is consistent 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 achieving high growth, respectively. Note from Chapter 9 that a firm’s networking 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 indicate 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

Networking: female- and male-controlled SMEs

Table 10.3

121

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

−0.03 0.09 −0.02

0.17 0.20 0.19

0.49 0.03 0.19 0.02

3 1 1 1

0.92 0.87 0.67 0.90

0.97 1.09 0.98

0.00

0.01

0.10

1

0.76

1.00

Industry Mining Manufacturing Construction Wholesale trade Retail trade Accom., cafes & restaurants Transport & storage Finance & insurance Property & bus. services Cultural & rec. services

0.58 −0.27 0.05 0.23 0.04 −0.35 0.09 −0.67 0.01 −0.23

0.80 0.44 0.49 0.46 0.46 0.53 0.54 0.50 0.45 0.58

14.67 0.53 0.38 0.01 0.26 0.01 0.44 0.03 1.81 0.00 0.16

10 1 1 1 1 1 1 1 1 1 1

0.14 0.47 0.54 0.92 0.61 0.94 0.51 0.87 0.18 0.98 0.69

1.79 0.76 1.05 1.26 1.04 0.70 1.09 0.51 1.01 0.79

Firm age Less than 2 years old 2 years to less than 5 5 years to less than 10 10 years to less than 20

−3.03 −0.40 −0.12 0.21

0.21 0.24 0.22 0.22

483.34 219.28 2.80 0.27 0.91

4 1 1 1 1

0.00 0.00 0.09 0.60 0.34

0.05 0.67 0.89 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

Education School Trade Non-business degree Experience

Percentage predicted correctly Survived/Discontinued/ Overall Nagelkerke R Square

42.7

96

88.4 0.36

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 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

Table 10.4

SME performance

Modelling firm growth and networking

Variables

B

S.E.

Wald

df

Sig.

Exp(B)

Sex of owner

0.08

0.25

0.11

1

0.74

1.09

Education School Trade Non-business degree

0.09 0.12 0.00

0.16 0.18 0.18

0.80 0.29 0.45 0.00

3 1 1 1

0.85 0.59 0.50 0.98

1.09 1.13 1.00

Experience

0.00

0.01

0.15

1

0.70

1.00

−0.62 −0.70 0.11 −0.44 −0.66 −1.13

0.79 0.47 0.50 0.48 0.49 0.62

19.42 0.62 2.23 0.05 0.84 1.80 3.37

10 1 1 1 1 1 1

0.04 0.43 0.14 0.82 0.36 0.18 0.07

0.54 0.50 1.12 0.64 0.52 0.32

−0.90

0.55

2.72

1

0.10

0.41

−0.44

0.52

0.72

1

0.40

0.65

−0.43

0.48

0.80

1

0.37

0.65

−0.40

0.62

0.41

1

0.52

0.67

1.03

0.26

16.05 15.81

4 1

0.00 0.00

2.79

0.36

0.20

3.29

1

0.07

1.43

0.36

0.18

4.02

1

0.05

1.43

0.26

0.17

2.25

1

0.13

1.29

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

−0.17

0.54

0.09

1

0.76

0.85

Industry Mining Manufacturing Construction Wholesale trade Retail trade Accom., cafes & restaurants Transport & storage Finance & insurance Property & bus. services Cultural & rec. services Firm age Less than 2 years old 2 years to less than 5 5 years to less than 10 10 years to less than 20

Constant

Networking: female- and male-controlled SMEs

Table 10.4

123

(continued)

Variables Percentage predicted correctly Low/High growth/Overall Nagelkerke R Square

B

63.2

S.E.

48.3

Wald

df

Sig.

Exp(B)

55.7 0.04

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 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 modelling 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 stepwise (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 compared 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 Models

Male-Controlled

Female-Controlled

Wald

Sig.

Exp(B)

Wald

Sig.

Exp(B)

Firm age 434.74 Less than 2 years 212.94 old 2 years to less than 3.10 5 5 years to less than 0.07 10 10 years to less than 0.69 20

0.00 0.00

0.05

37.84 14.55

0.00 0.00

0.01

0.08

0.66

0.15

0.70

0.59

0.80

0.95

0.48

0.49

0.44

0.41

1.21

0.33

0.56

2.37

0.24 0.76

10.18 9.40 0.78

0.01 0.00 0.38

0.09 0.53

5.52 0.47 2.91

0.06 0.49 0.09

0.60 4.87

32.25

11.24

0.00

77.25

88.5

94.5

Formal networks External accountant Never 1–3 times Industry association Never 1–3 times Informal networks Others in the industry Never 1–3 times Family and friends Never 1–3 times Constant

77.07 68.23 2.77 13.27 10.86 1.49

0.00 0.00 0.10 0.00 0.00 0.22

9.48 0.06 5.42

0.01 0.81 0.02

201.84

Percentage predicted correctly Survived/Discont./ 96.6 Overall Chi-square significance 22 Log likelihood Nagelkerke R Square

0.00 39.4

0.50 0.74

0.96 1.58

66.7

89.0

0.00

0.00

1716 0.36

91 0.62

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 each network the last category is ‘More than 3 times’. Source:

ABS.

Networking: female- and male-controlled SMEs

Table 10.6

125

Modeling firm growth and individual network contact for male- and female-controlled SMEs

Variables in the Final Models

Male-Controlled

Female-Controlled

Wald

Sig.

Firm age Less than 2 years old 2 years to less than 5 5 years to less than 10 10 years to less than 20

17.06 15.71 2.92 5.03 1.49

0.00 0.00 0.09 0.03 0.22

2.78 1.39 1.49 1.23

Industry Mining Manufacturing Construction Wholesale trade Retail trade Accom., cafes & restaurants Transport & storage Finance & insurance Property & bus. services Cultural & rec. services

21.65 0.93 2.47 0.00 1.55 2.62 4.03

0.02 0.33 0.12 0.97 0.21 0.11 0.05

0.44 0.42 0.98 0.49 0.39 0.22

3.26 1.16 1.22 0.60

0.07 0.28 0.27 0.44

0.32 0.52 0.53 0.57

External accountant Never 1–3 times

6.71 6.36 0.08

0.04 0.01 0.78

0.64 0.96

Industry association Never 1–3 times

6.46 0.09 3.07

0.04 0.76 0.08

0.95 1.37

Constant

0.69

0.41

1.62

Percentage predicted correctly High/Low/Overall 53.2 Chi-square significance 22 Log likelihood Nagelkerke R Square

63.0

Exp(B)

58.1 0.00 1683 0.06

Wald Sig. Exp(B)

8.10 3.02 7.25

0.02 0.08 0.01

0.25 0.24

3.52

0.06

2.00

62.9

71.1

67.1 0.01 92 0.15

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’; 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 provided 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 counterproductive. 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 malecontrolled 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 valuable than networks that are focused on only weak or only strong ties. It

Networking: female- and male-controlled SMEs

127

should also be noted that the model for predicting the survival of femalecontrolled 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 accountants 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 highgrowth 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 external accountants (possibly because of problems associated with accessing 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 networks. 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 structural 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 particular, banks, solicitors, industry associations and business consultants). Further, with the exception of the relationship between industry associations 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 maleand 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 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.

Networking: female- and male-controlled SMEs

129

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 femalecontrolled 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 invaluable 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 performances 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 networking (more than three times during a year) might be counterproductive. This was particularly true of the association between firm survival and the use of certain informal networks (others in the industry for malecontrolled 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 femalecontrolled SMEs); industry associations (for the survival and growth of male-controlled SMEs); others in the industry (for the survival of malecontrolled SMEs); and family and friends (for the survival of femalecontrolled 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 substantial 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. 11.0

Conclusions, implications and areas for future research 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 misperceptions. 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 businesses 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 businesses 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 particularly 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 performance 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 considered 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 comparing 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 limitations 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 businesses) 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, industries 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 positively 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 differences (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 appropriately and/or have not controlled for key demographic differences. As a result, most of the previous research in this area is incomplete and probably 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 preferences, 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 relatively 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 structure 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 differences 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 business 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 significantly 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 unsuccessful companies’, it contrasts with the results of Robson and Bennett (2000) and Cooper et al. (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 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 performances 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 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 networking (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 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).

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 appropriately 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 external 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 WomenOwned 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 WomenOwned 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 of Business Venturing, 13(6), 523–542. 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. Journal of Business Venturing, 8(3), 241–253. 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. Journal of Business Venturing, 9(5), 371–395. 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? Journal of Small Business Management, 13(3), 1–5. 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. Journal of Small Business Management, 41(2), 205–212. 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. Journal of Small Business Management, 16(3), 44–49. 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 of Business Venturing, 5(5), 327–340. 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 FemaleControlled 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 SexRoles, 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 frequency of contact 118–19, 120 Institute of Chartered Accountants 3 and performance 123–6, 127, 128, 130, 137, 139 age of business and failure rates 31–4, 35, 42, 135 female-controlled SMEs 44, 49–50, 51–2, 57–8, 63–4 and growth 91–5 and networks 108–15, 121–2, 124, 125 Aldrich, H. 99, 100, 116, 128, 137 Allen, K.R. 67 Anna, A.L. 44, 51 Ansic, D. 61, 89 Aram, J.D. 103 assets 54–8 Australia, SMEs in age of business, and failure rates 31–4, 35 bankruptcies 26, 29 case studies 17–20 closure rates 26–7, 29 economic conditions, effect of 39–40 failure statistics 20–25, 27, 28 female-controlled SMEs 48–52 focus group survey of demand for finance 70–78 growth, relationship with funding 90–95 networks 104–5, 117–20 output/input relationship, as performance measure 54–8 questionnaire regarding demand for finance 78–85 risk, attitudes to 62–4 size of business, and failure rates 36–7

Australian Bureau of Statistics (ABS) data age of business 33–4, 35 failure rates 26–7, 29 female-controlled SMEs 48–52 growth, relationship with funding 90–95 networks 104–5, 117–20 output/input relationship, as performance measure 54–8 risk, attitudes to 62–4 Ballantine, J.W. 59 bank funding attitudes to 71–8, 79–86, 136–7 demand for funding 69–70 female-controlled SMEs 6–7, 67, 79–86, 88–90, 136 focus group results 70–78 and growth 67–8, 90–97, 136 owners’ considerations 70–86 Pecking Order Theory 87, 88, 89–90, 94–7 questionnaire results 78–85 see also ‘finance gap’ bankruptcy age of business 31–3 comparative studies 25–6, 29, 31 economic conditions, effect of 39, 40, 135 as failure 15–16, 18–20, 133 failure statistics 20–25 industry sector 37–8, 42 size of business 36–7, 42, 135 banks 118–19 Barber, B.M. 61 Bates, T. 14, 15, 25, 31, 129, 138 Becchetti, L. 69, 85, 91 Belgium 25, 29 Bennett, R.J. 105, 118, 128, 137 Berger, A.N. 69, 87

153

154

SME performance

Berggren, B. 70 Bernasek, A. 61 Bhattacharjee, A. 41 Bhide, A. 134 Birley, S. 14, 69, 99, 105, 116, 117, 118, 119, 120 Boden, R.J. 47, 49 Bowman-Upton, N. 61, 89 Box, M. 14, 28, 29, 32 Brüderl, J. 32, 37, 126, 128 Bruno, A.V. 67 Brush, C. 69, 102 Building Owners and Managers Association in Australia (BOAMA) 20 Burt, R.S. 102, 126, 128 business consultants 118–19 business plans 74, 75, 78 Buttner, E.H. 70, 134 Canada 61 capital 44, 45, 47, 54–8 Carpenter, R.E. 67, 94 Carter, N.M. 47, 49, 67, 69, 102 Carter, S. 44, 53, 119 Chaganti, R. 70 Cleveland, F.W. 59 Cliff, J.E. 61, 70, 95 closure of business age of business 31–4 comparative studies 26–8, 29, 31 economic conditions, effect of 39, 40 as failure 13–15, 18–20, 133–4 failure statistics 20–25 industry sector 37–8, 42 size of business 35–7, 42 see also discontinuance (sale or closure) of business; sale of business Cochran, A.B. 11, 13, 16, 17 Coleman, J.S. 101 construction businesses 50, 52, 92, 93, 121–2, 125 control female-controlled SMEs 83–5, 89 finance for growth 83–5 male-controlled SMEs 83–5, 89 retention by owner 72–3, 75, 76, 78 Cooper, A.C. 14, 47, 100, 105, 118, 128, 137

creditors, losses to 15–16 see also bankruptcy; disposal, to limit losses Cressy, R. 32, 72–3, 89, 91 Cromie, S. 99, 116, 117, 118, 119 data 3–7 see also Australian Bureau of Statistics (ABS) data Davidsson, P. 90 debt 54, 58, 88–90 debt to asset ratio 90–95 DeCarolis, D. 70 Deeds, D. 70 Delmar, F. 90 Dewaelheyns, N. 25, 29 discontinuance (sale or closure) of business age of business 31–4 economic conditions, effect of 39, 40 as failure 13–15, 18–20, 133–4 failure statistics 20–25 industry sector 42 size of business 35–7, 42, 135 see also closure of business; sale of business disposal, to limit losses age of business 31–3 economic conditions, effect of 39, 40 as failure 16, 18–20, 133 failure statistics 20–25 size of business 36–7 Donckels, R. 102 Duchesneau, D.A. 101 Dunkelberg, W.C. 14, 105, 118 Dyke, L.S. 43 economic conditions, effect of 38–41, 42, 135 education, effect of 91–4, 121–2 employment rates 39, 40, 41 equity funding 54–8, 96, 106–15 Everett, J.E. 20–25, 29, 39, 41 experience of owner, effect of 91–4, 121–2 external funding attitudes to 71–8, 79–86, 136–7 demand for funding 69–70

Index female-controlled SMEs 6–7, 67, 79–86, 88–90, 136 focus group results 70–78 and growth 67–8, 90–97, 136 owners’ considerations 70–86 Pecking Order Theory 87, 88, 89–90, 94–7 questionnaire results 78–85 see also ‘finance gap’ failures and age of business 31–4, 35, 42, 135 case studies 17–20 definitions 7, 13–17, 18–20, 133–4 economic conditions 38–41, 42, 135 female-controlled SMEs 47–52, 135–6 industry sector 37–8, 42, 135 male-controlled SMEs 47–52, 136 selection criteria 11–13 size of business 34–7, 42, 135 statistics 3–7, 20–28, 29–30, 31 family networks female-controlled SMEs 117, 126, 129 frequency of contact 119, 137–8 survival of the business 124, 130, 139 Fasci, M.A. 56 female-controlled SMEs and accountants 118–19, 120, 123–6, 127, 128, 130 age of business 44, 49–50, 51–2, 57–8, 63–4 control, retention of 83–5, 89 data 6–7 debt levels 88–90 failure rates 47–52, 135–6 finance for 6–7, 67, 79–86, 88–90, 136 formal networks 118–20, 123–8, 129 growth, relationship with funding 90–97 and industry associations 118–19, 124, 125, 126, 127, 130 industry sector 44, 50–52, 57–8, 63–4 informal networks 118–20, 123–8, 129

155

intensity (access frequency) of networks 118–20 vs. male-controlled SMEs 43–5 motivation 45, 134 networks 99–100, 116–17, 120–130, 137–9 output/input relationship, as performance measure 53–8 performance 99–100, 129–30 range (number) of networks 117–18 risk, attitudes to 45, 59, 61–4, 75, 76, 78, 83–6, 89 size of business 44–5, 49–50, 51–2, 63–4 finance attitudes to 71–8, 79–86, 136–7 demand for funding 69–70 female-controlled SMEs 6–7, 67, 79–86, 88–90, 136 focus group results 70–78 and growth 67–8, 90–97, 136 male-controlled SMEs 79–86 owners’ considerations 70–86 Pecking Order Theory 87, 88, 89–90, 94–7 questionnaire results 78–85 ‘finance gap’ and growth 94–7, 136 meaning of 69, 87–8 survey results 71, 74, 78, 79–80 financial businesses 50, 52, 92, 93, 121–2, 125 financial institutions 71–8, 79–86, 88–90 see also bank funding financial return, as motivation 134 Fischer, E. 43, 59, 101, 115 Florin, J. 99 Forlani, D. 59, 60 formal networks female-controlled SMEs 118–20, 123–8, 129 intensity (access frequency) 104–5 male-controlled SMEs 118–20, 123–8 nature of 103 and performance 111–15, 137–9 range (number of) 104–5 Forsyth, G.D. 14, 28, 29 Foster, G. 39

156

SME performance

Fox, M.A. 69, 85, 136 Fraser, S. 79, 80 Fredland, E.J. 13, 37, 39 Freel, M.S. 119 Fried, V.H. 68 friend networks female-controlled SMEs 117, 126, 129 frequency of contact 119, 137–8 survival of the business 124, 130, 139 funding see external funding Gallagher, C. 14, 25, 26, 31, 37 Ganguly, P. 14, 32 Garrod, P. 13–14, 26 Gartner, W.B. 90, 101 Gimeno-Gascon, J.F. 100, 128 goals, of SME owners 134, 139 government regulation 71, 72 Granovetter, M.S. 101, 102, 103, 126, 128 growth and finance for SMEs 67–8, 90–97, 136 and ‘finance gap’ 94–7, 136 and networks 106–15, 120–128, 129, 138 obstacles to 70–78 and Pecking Order Theory 94–7 see also performance Hall, K.S. 40 Hamilton, D. 44, 53 Hamilton, R.T. 69, 85, 136 Harada, N. 15, 26 Headd, B. 14, 15, 21, 27–8, 29 Hill, R.C. 134 Hisrich, R.D. 68, 102 hours worked, effect of 45, 56–8 Hudson, J. 15, 25, 29 human capital 45, 47, 91–4 Hustedde, R.J. 102 Hutchinson, A.R. 14, 35–6, 38, 44, 51 Hutchinson, R.G. 14, 35–6, 38, 44, 51 Hutchinson, R.W. 70 Ibarra, H. 103 income 54–8, 90, 106–15, 120–127 industry associations 118–19, 124, 125, 126, 127, 130, 139

industry sector and failure rates 37–8, 42, 135 female-controlled SMEs 44, 50–52, 57–8, 63–4 and growth 91–4 and networks 108–15, 121–2, 125 informal networks female-controlled SMEs 118–20, 123–8, 129 intensity (access frequency) 104–5 male-controlled SMEs 118–20, 123–8 nature of 103 and performance 111–15, 130, 137–9 range (number of) 104–5 input/output relationship, as performance measure 53–8 Institute of Chartered Accountants 3 see also accountants insurance businesses 50, 52, 92, 93, 121–2, 125 interest rates 39, 40, 41 Jack, S.L. 119 Japan 26 Jianakoplos, N.A. 61 Jovanovic, B. 31, 34, 35, 36, 51 Julien, P.A. 99 Kalleberg, A.L. 47 Kent, P. 102 Kirchoff, B.A. 14, 37 Knott, A.M. 7 Koeller, T.C. 59 Kon, Y. 74 Kraimer, M.L. 102 Lam, W. 119 Lambrecht, J. 102 Landstrom, H. 69 Larsson, E. 102 legal organisation (status) of the business 91–4 Leicht, K.T. 47 Lerner, M. 102 Levenson, A.R. 71, 79 leverage 54, 58 Lewis, J. 4, 11

Index Liden, R.C. 102 Littunen, H. 103 losses 15–16 see also bankruptcy; disposal, to limit losses Lowe, J. 15, 37 Lubatkin, M. 99 ‘make a go of it’, failure to age of business 31–3 economic conditions, effect of 39, 40 as failure 16–17, 18–20, 133 failure statistics 20–25 industry sector 38 size of business 36–7 male-controlled SMEs and accountants 118–19, 120, 123–6, 127, 128 control, retention of 83–5, 89 failure rates 47–52, 136 vs. female-controlled SMEs 43–5 finance, demand considerations 79–86 formal networks 118–20, 123–8 growth, relationship with funding 90–97 and industry associations 118–19, 124, 125, 126, 127, 130, 139 informal networks 118–20, 123–8 intensity (access frequency) of networks 118–20 networks 99–100, 116–17, 120–130, 137–9 output/input relationship, as performance measure 54–8 performance 99–100, 120–130, 137–9 range (number) of networks 117–18 risk, attitudes to 59, 61–4, 75, 76, 78, 83–6, 89 managed shopping centres age of business, and failure rates 31–3 as data sources 6 economic conditions, effect of 39–40 failure statistics 20–25, 27, 28

157

size of business, and failure rates 36–7 manufacturing businesses 50, 52, 92, 93, 121–2, 125 McGrath, R.G. 7 McKenna, J. 15, 37 McPherson, J.M. 103, 116 Miklius, W. 13–14, 26 mining businesses 50–51, 52, 92, 93, 121–2, 125 Moore, D.P. 70, 134 Moore, G. 103, 117 Morris, C.E. 13, 37, 39 motivation 45, 134, 139 Mukhtar, S.-M. 89 Mullins, J.W. 59, 60 Munch, A. 103, 116 Murphy, G.B. 134 Myers, S.C. 87 Nelson, G.W. 138 networks definition 99 female-controlled SMEs 99–100, 116–17, 120–130, 137–9 and growth 106–15, 120–128, 129, 138 intensity (access frequency) 103–5, 114, 115, 117–20 male-controlled SMEs 99–100, 116–17, 120–130, 137–9 nature of 102–3 and performance 99–100, 101–2, 105–14, 120–130, 137–9 range (number of) 103–5, 114, 115 see also formal networks; informal networks New Zealand 69 Newby, R. 61 Newcomer, M. 14, 36, 38, 44, 51 Nielsen, A. 13, 16, 17 Nucci, A.R. 14, 25, 31, 47, 49 Odean, T. 61 Olofsson, C. 70 Orhan, M. 117 output/input relationship, as performance measure 53–8 owner withdrawals 96

158

SME performance

partnerships 14–15, 36 Pecking Order Theory 87, 88, 89–90, 94–7 performance and accountants 123–6, 127, 128, 130, 137, 139 female-controlled SMEs 99–100, 129–30 and formal networks 111–15, 137–9 and informal networks 111–15, 130, 137–9 male-controlled SMEs 99–100, 120–130, 137–9 and networks 99–100, 101–2, 105–14, 120–130, 137–9 and professional advice 100, 101–2, 104–5, 118–19 see also growth performance measures case studies of failure 17–20 definitions of failure 13–17, 18–20, 133–4 failure statistics 20–28, 29–30, 31 female-controlled SMEs 47–52, 135–6 selection criteria 11–13 see also failures Petersen, B.C. 67, 94 Phillips, B.D. 14, 37 Posen, H.E. 7 Potts, A.J. 5, 24, 101, 128, 137 Powell, M. 61, 89 Preisendörfer, P. 32, 37, 126, 128 professional advice 100, 101–2, 104–5, 118–19 see also accountants profits 20–21, 54–8, 60–64, 90–94, 96 property businesses 50, 52, 92, 93, 121–2, 125 Pulver, G.C. 102 Reese, P.R. 100 regulation 71, 72 relationship lending 78 retail businesses 50, 52, 92, 93, 121–2, 125 retail sales 39, 40, 41, 90, 125, 127 return on assets (ROA) 54–8 return on equity (ROE) 54–8, 106–15 Reuber, R.A. 43, 101, 115

Reynolds, P.D. 5, 14, 47 Riding, A. 69 risk controlling for 59–60 female-controlled SMEs 45, 59, 61–4, 75, 76, 78, 83–6, 89 finance for growth 72, 73–4, 75, 76, 78, 83–5 male-controlled SMEs 59, 61–4, 75, 76, 78, 83–6, 89 Robson, P. 105, 118, 119, 120, 128, 137 Rosa, P. 44, 51, 53 sale of business age of business 31–3 economic conditions, effect of 40 as failure 13–15, 18–20, 133–4 failure statistics 20–25 size of business 35–7 see also closure of business; discontinuance (sale or closure) of business sales 39, 40, 41, 90, 125, 127 Schulze, W. 99 Scott, M. 4, 11 Seibert, S.E. 102 service businesses 50, 52, 92, 93, 121–2, 125 Sexton, D.L. 61, 89 Shailer, G. 39 Sharpe ratio 60, 62–4 Sharpe, W.F. 39, 59–60 Shaw, E. 119 shopping centres age of business, and failure rates 31–3 as data sources 6 economic conditions, effect of 39–40 failure statistics 20–25, 27, 28 size of business, and failure rates 36–7 Silver, L. 70 size of business and failure rates 34–7, 42, 135 female-controlled SMEs 44–5, 49–50, 51–2, 63–4 growth, and funding 90–94 and networks 108–15, 121–2

Index SMEs (small- and medium-sized enterprises) case studies 17–20, 28–30 definitions of failure 7, 13–17, 18–20, 133 economic conditions, effect of 38–41, 42, 135 future research 139 misconceptions about 3–4, 6–7 statistics 3–7, 20–28, 29–30, 31 see also age of business; failures; female-controlled SMEs; finance; industry sector; malecontrolled SMEs; performance; size of business Smith-Lovin, L. 103, 116 social resource theory 102 sole traders 14–15, 36 solicitors 118–19 Stanworth, J. 4, 14 Stewart, H. 14, 25, 26, 31, 37 Storey, D.J. 74 structural holes 102, 126, 128 survival of the business 106–15, 120–128, 129, 130 Sweden 28, 29, 70 Swift, C.S. 69 Taggart, R.A. 60 tax office 119 Tibbits, G. 15, 37 Trailer, J.W. 134 transaction-based lending 78 transport businesses 50, 52, 92, 93, 121–2, 125 Trovato, G. 69, 85, 91 Tyebjee, T.T. 67

159

Udell, G.F. 69, 87 Ulmer, M.J. 13, 16, 17 unemployment rates 39, 40–41 United Kingdom (UK) 25, 29, 41, 79 United States (US) bankruptcies 26 closure rates 27–8, 29 economic conditions, effect of 39 failure rate statistics 4 female-controlled SMEs 47, 89 finance for SMEs 67, 69, 79 growth, relationship with funding 94 risk, attitudes to 61 Uzzi, B. 126 Valdez, J. 56 Van Hulle, C. 25, 29 variability, and risk 60 Wadhwani, S.B. 40 Watson, J. 20–25, 29, 39, 41, 61, 105, 106 weak tie theory 102, 126, 128 wholesale trade 50, 52, 92, 93, 121–2, 125 Willard, K.L. 71, 79 Williams, M. 47 Wilson Committee 4 Winborg, J. 69 women see female-controlled SMEs Woo, C. 14, 100, 105, 118, 128 Zhao, L. 103 Ziegler, R. 32, 37