SME Performance
SME Performance Separating Myth from Reality
John Watson Professor of Accounting and Finance, The Un...
29 downloads
881 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
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: 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