Capital pecking order in small
and medium sized enterprises
What do size and crises have to do with it
Yorben Blok
0312614
October 2014
Abstract
Trade off theory, Agency theory and the Pecking order theory explain the capital structure for businesses. This thesis analyses theoretically the explanatory power of these theories for small and medium sized enterprises (SME). I find that most of the theory is applicable to larger SMEs and that for smaller SMEs other aspects, such as the entrepreneur, play a large role in financial decision making. The theoretical analysis leads to a “best fit” pecking order which is investigated empirically for differently sized SMEs. Using a sample of 5 countries with over 700 thousand firms for a 10 years period from 2003 to 2012 I find evidence that the pecking order does not hold for the EU SMEs tested. With the SMEs divided in three specified size classes most of the evidence points to a confirmation of the pecking order for Medium SMEs but rejects it for Micro SMEs. I also find a strong similarity for SMEs pecking order within countries and confirm that smaller SMEs use less external finance than larger SMEs. Hereby this thesis provides evidence that there is no such thing as a universal pecking order for SMEs. Alongside it is investigated to what extent the crisis years played a role in the capital structure of SMEs. As expected evidence points out that in the period 2007-2012 smaller SMEs are affected worse by the crises than larger SMEs. This is due to both larger SMEs ability to retain internal finance and acquire short term external finance. Lastly I find that all SMEs have considerable difficulty with acquiring long term external finance which arguable hampers investment and growth.
Master Thesis Finance
Supervisor: Dr. Jeroen Ligterink
MSc Business Economics: Finance
Amsterdam Business School
University of Amsterdam
Index
1.
Introduction ... 1
2.
SMEs and entrepreneurs ... 3
2.1 SMEs ... 3
2.2 Entrepreneurs... 4
3.
Capital structure theory ... 6
3.1 Trade-off theory (TOT) ... 10
3.1.1 Empirical and qualitative results (TOT) ... 11
3.2 Agency theory, SME risk and information asymmetry ... 13
3.2.1 Agency theory and moral hazard ... 13
3.2.2 SME specific risk & adverse selection (ex ante) ... 15
3.2.3 Information asymmetry... 17
3.3 Pecking order theory (POT) ... 19
3.3.1 Empirical and qualitative results (POT) ... 21
3.4 The financial growth cycle ... 23
3.4.1 Empirical and qualitative resuts ... 25
3.4.2 Country and industry effects ... 27
3.5 Trade off theory, Pecking order theory and Agency theory ... 27
4.
Hypotheses, data and methodology ... 29
4.1 Hypotheses ... 29
4.1.1 Internal and external finance ... 31
4.2 Methodology ... 33
4.3 Data handling and overview ... 34
5.
Empirical results ... 40
5.1 Pecking order theory ... 40
5.2 Internal vs. External finance in crises ... 48
6.
Discussion and conclusion ... 54
6.1 Discussion ... 54
6.2 Conclusion ... 56
1
1. Introduction
The idea for the subject of this thesis on small and medium sized enterprises (SMEs) capital structure originated due to the problems with regard to credit availability to said enterprises. Since the
financial crisis in 2007 it is suggested that SMEs find it harder to acquire external financing for the funding of their business (ECB, 2014). The initial line of research, coupled with an internship at PerspeXo1, focused on the entrepreneur and the demand side of acquiring external financing. This showed two distinctive results, first that there was considerable difference between different sized enterprises and the overall influence of the entrepreneur on financial decision making. Secondly the subsequent academic research suggested that there was considerable difficulty for SMEs to acquire external financing which leads to a greater reliance on internal finance. Limited use and possible supply of external financing for SMEs emphasises the role of retained earnings as a means of
financing the business. However economic conditions do not only negatively affect credit availability but also profits and subsequently retained earnings for SMEs: a proverbial double whammy. This insight led to a concentration on capital structure and the difference in funding between internal and external finance during crisis years.
Eventually the research comprehended so many discernments that including them all in a single thesis would make it to “wide”. Therefore this thesis focusses solely on the capital structure and role of internal and external funding for different sized European SMEs during the past ten years of worsening macroeconomic conditions. It uses a theoretical and empirical approach to hypothesise and analyse changes in SMEs’ capital structure. The main research question this thesis will answer is: is the capital structure of differently sized and located European SMEs similar as suggested by the pecking order theory between 2003 and 2012. The thesis expands on the existing literature by using a 10 year dataset of cross country data (Belgium, Germany, Italy, Spain and the UK). Furthermore by defining and testing three different size classes, the diversity of the SME sector with regard to capital structure is reflected and underlines the difference between small and large SMEs. Finally the inclusion of interaction dummies will isolate the effect of the crisis years 2007-2012 for differently sized and located SMEs, to establish the effects of internal and external finance on SMEs total asset funding.
The entrepreneur is not completely excluded from this thesis but most of the in-depth demand side research is referred to in the appendix. By including the entrepreneur in the appendix to this thesis two things will become clear. Firstly the limited usage of external financing is most definitely not just a supply side problem and secondly the role of the entrepreneur, especially in
2 Micro and Small SMEs, is likely far more important with regard to the capital structure of a SME than is suggested by existing empirical literature (Appendix I.I, p. 2).
The importance of understanding this sector and shed light on capital decisions and structure of these firms is important for the economy since SMEs represent a vast portion of the European economy. Roughly 99,8% of the companies are SMEs and two out of three jobs in the private sector are provided by SMEs (EC, 2002). However, as Zingales (2000, p. 1629) points out “empirically, the emphasis on large companies has led us to ignore (or study less than necessary) the rest of the universe: the young and small firms, who do not have access to public markets”. Most of the capital structure research, initially established by the seminal work of Modigliani and Miller (1958), has focused on financial decisions of large, public firms. Large public firms optimise their capital structure with regard to minimising cost and maximising (shareholder) value. SMEs, which are usually not listed and usually have a single shareholder, differ substantially from large (public) companies due to specific factors. The first is information opacity and specific internal and external agency problems SMEs suffer related to asymmetric information and size related information requirements as well as moral hazard and adverse selection. The second is the perceived increased risk of SMEs due to potential volatility, growth, profitability and continuity of the business, which also imply a greater sensitivity to economic cycles. The third is due to the entrepreneur, the owner-manager role they play as shareholder and managing director in the decision making process and the effect they have on both aforementioned issues. All of the arguments above are closely linked to SMEs demand and supply for capital since they are all intrinsically related to the process of acquiring capital. Adding the macroeconomic conditions of 2007-2012 makes the EU SME sector an interesting subject for study, both empirically and theoretically.
The outset of this thesis is as follows: the first chapter will introduce SMEs and entrepreneurs and give an initial sketch of SME specifics. The next chapter will thoroughly review the literature on capital structure theory. Well known issues such as moral hazard, adverse selection and information asymmetry will be reviewed with respect to the specific environment SMEs and entrepreneurs operate in. Furthermore it will include full-fledged capital structure theories such as trade-off theory, agency theory and the pecking order theory and will review the well-established financial growth cycle to hypothesise a SME pecking order. This will be the foundation for chapter 4 which will formulate the hypotheses, give a data overview and show the methodology. Chapter 5 shows the empirical results which will be discussed in turn. Herein SME’s capital structure and the effects of the crisis will empirically be verified with regard to the hypothesis postulated in chapter 4. Chapter 6 will discuss and conclude this thesis.
3
2. SMEs and entrepreneurs
2.1 SMEs
When watching daily (business) news or studying business economics one could easily be mistaken that our economy is dominated by large, publicly traded companies. The media is filled with stock exchange trading, mergers and acquisitions and, unfortunately in this day and age, huge lay-offs by Forbes 500 companies. Academic literature and research is, quite possibly due to this public
attention, also focused on large companies. These companies are, or usually will be in the short term, publicly traded and employ thousands of people world-wide. What can get lost in the media and academic research is that there are many more small businesses defined as small and medium sized enterprises (SMEs). In the EU roughly 99% of all businesses are in fact SMEs. They provide two out of three jobs in the private sector and more than half of total value added is created by them (EC, 2002, 2012). A SME is officially defined in the EU by three variables and a single criterion. These variables are: staff headcount, annual turnover or annual balance sheet. The criterion is that the SME has to be independent. Depending on the
criterion and variables SMEs are classified as “Micro”, “Small” or “Medium”-sized (Figure 1, Source: EC, 2005, p. 15).
The “size-definition” of a company in the EU is given along the aforementioned related variables for clarification. However size is a definition not purely based on the above variables or used for defining groups of SMEs. Size can and often is, used as a proxy for information opacity due to
differences in accounting standards,
or plays a role in the study of organization theory due to the role of the entrepreneur. It is therefore of importance to clarify that in this thesis size, if not explicitly mentioned otherwise, will refer to the first variable of the EU definition: staff headcount2. The main argument for this is twofold: firstly decision-making in smaller companies is usually done by a single person. The larger (staff) a company
2 In the empirical part of this thesis these exclusive search criteria will be used to match the EU SME definition exactly.
4 gets, the more decisions are decentralised. Secondly, since many SMEs are service orientated (EC, 2013, p. 12), business is usually done by employees, making staff headcount a good proxy for size. Percentagewise SMEs are divided along the EU variables in the following order: 92,2% of SMEs fall in the Micro group with less than 10 employees, 6,5% is Small (<50 employees) and only 1,1% is Medium (<250 employees) (EC, 2012). A remarkable observation is that there are very few Medium but many Micro SMEs. One explanation could be that many SMEs today are service related. Service related businesses are easier to start due to no or low initial investments (start-up costs) and no sufficient negative economies of scale effects (Hutchinson, 1999). This explanation by Hutchinson would account for start-ups, but not why they remain small. A related explanation from Berger and Udell (1998) and Cressy and Olofsson (1997) is due to the contentment of the entrepreneur with the current size of the SME and personal income, which suggests that they have no motivation to grow and are not looking for capital. This would make the entrepreneur responsible for Micro SMEs to stay Micro (Appendix I.II, pp. 2-3).
2.2 Entrepreneurs
The foremost aspect that makes SMEs different from large businesses is the way they are organized. Small businesses are usually dominated by one, or a small team of entrepreneurs (partners/co-owners) whom have considerable, if not absolute decision power and are also the major investors (Pettit, 1985). Where large, public (European) businesses usually have layers of management, stockholders, two tier boards and a large group of stakeholders. In SMEs the entrepreneurs, are the driving force of the business and it is their vision that make the SME successful in what it does. Since especially smaller SMEs have a larger role for the entrepreneur, their influence on SMEs financial behaviour and capital structure is greater as is argued in the Appendix I.I (p. 2) and some academic research.
As Pettit and Singer (1985, p. 58) posit: “the level of debt and equity in a smaller firm is more than likely a function of the characteristics of the firm and its managers.” Or along a similar vine Levin and Travis (1987, p. 30): “The owners’ attitude towards public risk – not the capital structuring policies public companies use – determine what amounts of debt and equity are acceptable.” SMEs are therefore more dependent on personal preferences, capabilities, behaviour and motivation of the entrepreneur and are much more demand driven when it comes to the capital structure. Furthermore as Ang (1992) and other authors point out, small businesses are prone to make mistakes due to overconfidence or ignorance on the entrepreneur’s side. It is thus of importance to research the entrepreneur and the specific role they play in the financial decision making process with regard to capital structure. It is not always a rational trade off as is assumed in most of the
5 capital structure theory. Most of the theoretical research in Appendix I.II and I.III3 would namely suggest that there is a large group of (irrational) entrepreneurs either unwilling to, not looking for or incapable of acquiring external finance. This would suggest that for SMEs with considerable decision power for the entrepreneurs, arguably smaller SMEs, there is little to no external finance available and investment has to come solely from retained earnings.
This insight from the academic (behavioural) research, combined with an internship and a questionnaire conducted among almost 600 entrepreneurs by PerspeXo, leads to the following assumption: for smaller SMEs demand side factors play a larger role with respect to the capital structure puzzle than supply side factors, while for larger SMEs supply side factors are the dominant constraint (Appendix I). That is not to say smaller SMEs have less supply side problems than larger SME, on the contrary, as is argued in the remainder of this thesis, those issues are also larger for smaller SMEs. However demand side problems dominate the capital structure choice for small SMEs. The reason to include this research in the appendix instead of the main body of this thesis is twofold. First for the sake of clarity, to avoid a very bulky thesis that is too broad and loses focus. Secondly to emphasise instead of leaving out the importance of entrepreneurial demand, demand constraints and opportunities in small SMEs, which will give a more complete capital structure picture than previous empirical academic work. Furthermore a short version of this thesis and appendix will be part of the PerspeXo research literature and hopefully contribute to an awareness among
entrepreneurs with regard to the financial choices they have.
The financial explanation of an abundance of (small) SMEs is mostly sought in the inability and limited availability (demand and supply) to attain external finance, arguing the existence of a finance gap. SMEs have limited access to capital which serves as a growth constraint (Chittenden et al., 1996, Beck and Demirguc-Kunt, 2006 and Beck et al., 2006). Therefore the remainder of this thesis will focus on the financial behaviour of businesses that forms the capital structure. Theoretical explanations with respect to capital structure such as trade-off theory (TOT), agency theory, pecking order theory (POT) and the financial growth cycle paradigm, as well as the underlying issues such as financial distress, taxes, moral hazard, adverse selection and information asymmetry, are no less important for SMEs and their capital structure (Ang, 1992).
3
In Appendix I a succinct review of the literature on entrepreneurs is included. It contains an overview of qualitative traits of entrepreneurs and how they fit into small business capital structure theory.
6
3. Capital structure theory
The traditional concept regarding financial behaviour, defended until the 1950’s, revolved around investment decisions and the optimal financial structure to support said investment. This concept was based on (López-Gracia and Aybar-Arias, 2000, p. 56):
- Forecasted profitability of possible projects, implicitly assuming all projects found the necessary funds
- Optimal capital accumulation occurred alongside optimal financing structure
It was the work of Modigliani and Miller in 1958 that put this concept into doubt. Their theory of capital structure irrelevance in perfect markets showed that under perfect market conditions, investment and financing decisions occurred separately. The average cost of capital remained the same regardless of financial structure and as a result capital structure did not affect firm value. Modigliani and Miller’s (1958) (M&M) work sparked more research that relaxed some of the M&M assumptions. In particular the irrelevance of capital structure on firm value and imperfections in financial markets. Issues such as transactions costs, agency costs, information asymmetries and taxes were incorporated into different models. Each separate strand of research providing a different piece of the overall puzzle on capital structure. However no theory is universally accepted, as Myers (2001, p 81) mentions: “There is no universal theory of the debt-equity choice, and no reason to expect one. There are several useful conditional theories however.” These conditions rest with the specific firm characteristics one focusses on which, in this case is: being a SME
Ang (1992) argues that financial theory was not developed with SMEs in mind. Most financial (capital structure) theory is based on large, public companies which are fewer in number and
arguably more homogenous. Even though the European SME numbers give a general idea of how important this sector is today, Michaelas et al. (1999) note that the SME sector has mostly been ignored in financial theoretical and empirical research. Denis (2004) counts the amount of articles with respect to entrepreneurial finance over the past decades in the following journals: Journal of Finance, Journal of Financial Economics, Review of Financial Studies, Journal of Business, Journal of Financial and Quantitative Analysis and the Journal of Corporate Finance. He finds that research picked up in 1990 but finds only 6 articles are published a year. He argues that: “finance scholars have tended to view entrepreneurship as entirely separate from the field of corporate finance.” (2004, p. 303). Van der Wijst and Thurik (1993, p. 55) find similar results arguing that the (financial) theory for small businesses is seldom discussed and even less frequently tested quantitatively. Hereby showing a large discrepancy between the importance of the sector and research focus.
The first indication of small businesses inability to fund all their (profitable) projects is in the MacMillan Report (Stamp, 1931). The term that was used in the MacMillan Report was the
7 MacMillan Gap, later dubbed the finance gap, which argued that banks were not servicing the needs of their (small) industrial debtors. The MacMillan report describes the gap as a situation in which a firm had grown to a size where it made maximum use of short-term finance but was not big enough for capital markets (Chittenden et al., 1996). The specific lack of interest into small businesses in the following two decades was due to the publishing of Gibrat’s law of proportionate effect in the same period (Gibrat, 1931). Gibrat’s law states that size and growth are not related, which led to a lack of concern about firm size and stage of development. It was not until the works of Bates (1971) and subsequently Bolton (1971) and Wilson (1980) that raised the possibility - and also showed
quantitatively - that there were significant differences in financial structure between large and small firms. It was this insight, coupled with developing theories such as agency theory and life cycle approaches in the eighties that provided an explanation which incorporated the concept of a finance gap with capital structure theory. Eventually it were the spectacular IPO successes in the 90s and growing concerns that many small businesses were not able to obtain sufficient external finance that created public focus and subsequent theoretical and empirical research into SMEs capital structure (Berger and Udell, 1998, p. 659 and Vos et al., 2007).
The resulting flow of empirical research into SME financing focusses specifically on which capital structure theory applies best to SMEs and tries to quantitatively prove one of these theories for capital structure choices. The theories that are reviewed are: Trade off theory (TOT), Agency Theory and Pecking Order Theory (POT). Table 1 and Table 2 on the next page summarise the empirical results from the authors whom test specific determinants on debt levels in SMEs. Table 1 summarises the articles that research determinants on total equity ratio’s or total debt-to-asset ratio’s while Table 2 summarises articles that research the effects on short term debt (STD) and long term debt (LTD) ratio’s separately. The last column in both tables sums up the total tests done and if the relationship is negatively or positively significant. This gives a clear view on which
hypotheses are uniformly confirmed or rejected. In the next subchapters on capital structure theory the results in both tables will be discussed when applicable.
The first theory that will be explained in full is the trade-off theory which focuses mainly on tax benefits of lending. TOT will be followed by agency theory which will emphasise the problems of entrepreneurs’ intertwinement and heterogeneous SMEs. Subsequently information asymmetry will be introduced for SMEs followed by the POT. Large business empirical research will only be covered abridged, mostly to show general findings and relevance of the theories over the years. Since SMEs rarely participate in public markets, theory that emphasises the role of public equity and share price are only mentioned abridged.
8
TABLE 1SUMMARY EMPIRICAL RESULTS (DEPENDENT VARIABLE:TOTAL DEBT)
AYBAR-ARIAS ET AL. (2003) VAN CANEGHEM, VAN CAMPENHOUT (2012) HEYMAN ET AL. (2008) LÓPEZ-GRACIA, SOGORB-MIRA (2008) TOTAL (NEGATIVE/POSITIVE)
CASH FLOW Negative Negative Negative Negative 4 (4/0)
SIZE Positive Negative Negative Positive 4 (2/2)
AGE Negative Negative - Negative 3 (3/0)
ASSET
STRUCTURE - Positive Positive - 2 (0/2)
GROWTH - Positive Negative - 2 (1/1)
TAXES - - - Not sign. 1 (0/0)
NDTS - - - Negative 1 (1/0)
SAMPLE
N (FIRMS) ≈ 6000 79,097 1,132 3,569
COUNTRY Spain Belgium Belgium Spain
YEAR
1995-2001 2007 1996-2000 1995-2004
Notes: The table reviews the sign of the coefficient (Negative/Positive) and whether it is significant with regard to total debt. The last column shows the total amount of tests and sums the number of significant negative or positive coefficients. NDTS = Non Debt Tax Shields
9
TABLE 2SUMMARY EMPIRICAL RESULTS (DEPENDENT VARIABLE:SHORT TERM DEBT OR LONG TERM DEBT)
CHITTENDEN ET AL. (1996) DEGRYSE ET AL. (2012) HALL ET AL. (2004) HUTCHINSON (2004) MICHAELAS ET AL. (1999) VAN DER WIJST, THURIK (1993) ZOPPA, MCMAHON (2002) TOTAL (NEGATIVE/POSITIVE)
CASH FLOW STD Negative Negative Negative Negative Negative Negative Negative 7 (7/0)
LTD Not sign. Not sign. Not sign. 7 (4/0)
SIZE STD Negative Negative Negative Negative Negative Not sign. Positive 7 (5/1)
LTD Positive Positive Positive Positive Positive Not sign. 7 (0/5)
AGE STD Negative - Negative Negative Negative - Negative 5 (5/0)
LTD Not sign. Not sign. Not sign. 5 (2/0)
ASSET STRUCTURE
STD
Positive Positive Negative Negative Positive Negative Negative 7 (4/3)
LTD Positive Positive Positive Positive 7 (0/7)
GROWTH STD Not sign. Not sign. Positive Not sign. Positive - Positive 6 (0/3)
LTD Positive Not sign. Not sign. 6 (0/2)
TAXES STD - Not sign. - - Not sign. - - 2 (0/0)
LTD 2 (0/0)
NDTS STD - Positive - - Not sign. Not sign. - 3 (0/1)
LTD Negative 3 (1/0)
SAMPLE
N (FIRMS) 3,480 ≈ 25,000 3,951 1,000+ 3,500 15-17
Industries 871
COUNTRY UK Netherlands EU UK UK
West-Germany Australia
YEAR 1988-1993 2002-2005 1995 1991-1995 1986-1995 1950-1980 1995-1997
Notes: The table reviews the sign of the coefficient (Negative/Positive) and whether it is significant with regard to short or long term debt. The last column shows the total amount of tests and sums the number of significant negative or positive coefficients.
10
3.1 Trade-off theory (TOT)
Trade-off theory suggest that the optimal capital structure is a trade-off of the benefits of debt financing, lower taxes due to interest deductibility, and the costs of debt financing namely higher cost of financial distress. To be more specific the trade-off theory states that a business will borrow up to the point where the marginal value of tax shields on additional debt is just offset by the increase in the present value of possible costs of financial distress (Myers, 2001, p. 88). As Myers (1984) showed, trade-off theory implies that businesses strive to some sort of optimum level or target of debt to equity ratio were the benefits and costs of debt-financing exactly offset one another. Subsequently the hypotheses are that businesses with higher profitability and higher tax rates will have higher debt ratios to increase tax shields.
Shyam-Sunder and Myers (1999) give an overview of evidence over the past 50 years of trade-off theory. Some articles find direct evidence that firms adjust toward a target debt ratio, while others show specific determinants such as intangible assets which cannot be used as collateral in the event of bankruptcy, negatively related to debt ratios in congruence with the trade-off theory. In the article by Bradley et al. (1984) most of the earlier theoretical and empirical literature is reviewed on large business capital structure and they conclude that their findings support the modern trade-off theory of capital structure for large businesses. Summarised the literature confirms two hypotheses:
- Businesses do reverse actual debt ratios towards an optimum (target adjustment)
- There is a cross-sectional relation between average debt ratios and asset risks, profitability, tax status and asset type.
However, as Shyam-Sunder and Myers (1999) argue, most of the empirical literature is guided almost exclusively by the assumption of an optimal debt ratio. They argue that some of the evidence is inconsistent with optimal debt ratios or can be interpreted differently. Firstly, the effects of valuation of equity in the presence of information asymmetry is not included in the trade-off theory. Since “new” information changes the valuation of equity, it also changes the value of the firm and thereby the actual and (new) optimum debt ratio. Second, of more importance to SME and trade-off
research, other authors find strong negative relationships between past profitability and debt ratios, also inconsistent with TOT.
Since none of the articles has systematically compared the explanatory power of their fitted equations with alternatives, nor strived to find whether their equations seemed to work even when actual financing is driven by other forces, Shyam-Sunder and Myers (1999) test the statistical power of the trade-off theory. They use a database of mature public firms with investment-grade debt ratings, which are specifically chosen due to their ability to issue nearly risk free debt and are therefore less constrained when trying to reach an optimum debt target. Shyam-Sunder and Myers
11 (1999) find that even though the trade-off theory performs well statistically, the target adjustment model for mature companies does not hold. In Shyam-Sunder and Myers’ (1999, p. 242) words: “If our sample of companies did have well-defined optimal debt ratios, it seems that their managers were not much interested in getting there.” More importantly however, in a simulation experiment conducted by Shyam-Sunder and Myers, the target adjustment model of the TOT is not rejected, even when financing decisions are made purely on pecking order behaviour.
A more recent article by Myers (2001) mentions the considerable lack of evidence of the trade-off theory among all (large and small) businesses. Many established companies that are able to increase tax shields with a low probability of financial distress do not do so. Furthermore, the most profitable companies with more taxable income to shield, tend to borrow the least. There is however evidence that tax driven tactics exist in large companies with public equity. MacKie-Mason (1990) finds evidence for businesses with low marginal tax rates, with less income to shield, to be more likely to issue equity compared to more profitable companies and Graham (1996) finds evidence that changes in long term debt (LTD) are positively and significantly related to the marginal tax rate of the business.
3.1.1 Empirical and qualitative results (TOT)
There has been strong evidence against trade-off theory theoretically and statistically. Statistically by the aforementioned article by Shyam-Sunder and Myers (1999) on statistical power. Theoretically by Briozzo (2007) on the ease of use of the TOT in congruence with the limited financial capabilities of entrepreneurs to find such an optimum level of debt (Appendix I.III pp. 5-6). Also research conducted empirically on SME capital structure as summarised in Table 1 and 2 find results inconsistent with TOT. All the articles in Table 1 and 2, with the exception of López-Gracia and Sogorb-Mira (2008), suggest that entrepreneurs operate without targeting an optimal level of debt. For the researched relation between profitability and debt four out of four articles find a negative relation with total debt (Table 1) and seven out of seven articles a negative relation with short term debt (STD) (Table 2). Four out of seven articles find a negative relation with long term debt (LTD) and three are non-significant (Table 2). Furthermore out of the three articles that research a relationship between taxes and debt levels, none finds a significant relation with total, short or long term debt (Table 1 and 2).
López-Gracia and Sogorb-Mira (2008) give an explanation in response to the non-significant relation between effective tax rate and debt. They suggest that SMEs use other fiscal constructions to shield their taxable income. These are non-debt tax shields (NDTS), such as depreciation on assets or tax deductible investments that diminish the need for interest on debt as a tax shield. López-Gracia and Sogorb-Mira (2008) find empirical evidence that NDTSs influence the capital structure of SMEs but the results do not show that SMEs are using NDTSs to shield all of their profitable income.
12 Since NDTSs does little to financial distress costs, given that it is purely an accounting principle without actual cash flow, it would not devalue the possible advantage of a debt tax shield for remaining profitable income. But even for the remaining income, López-Gracia and Sogorb-Mira do not find that businesses attract debt, similar to the results found by Shyam-Sunder and Myers (1999) for large companies. Another argument put forward by López-Gracia and Sogorb-Mira (2008) is that SMEs have less (continuous) profitable income to shield. Tax advantages are especially of interest to (continuously) profitable and less volatile SMEs which diminishes the need for debt tax shields. However the two aspects mentioned, profitability and less volatile income, are at odds with the very specifics of an SME, begging the question how relevant trade-off theory for SMEs actually is.
Table 2 shows that Michaelas et al. (1999) find that neither the NDTS nor the effective tax rate is significantly related to debt levels. Van der Wijst and Thurik (1993) find similar results for NDTSs, suggesting that entrepreneurs do not include tax effects in their capital structure decisions. Degryse et al. (2012) do find some limited evidence that NDTS have a negative relation with regard to long term debt levels, which could indicate that tax considerations become more important in the longer term capital structure decisions. It is however hard to argue, let alone empirically prove, that a business’s tax status has predictable effects on its debt policy. As also mentioned by Myers (1984) classifying businesses by tax status would implicitly classify them on other dimensions as well. As Myers (1984) explains, firms with large tax loss carry forwards may also be firms in financial distress with high debt ratios. Alternately, firms with high profitability may also have valuable intangible assets and growth opportunities. The issue whether either of these companies ends up with a higher or lower than average debt ratio remains, to quote Myers (1984, p. 588), “hard to say”.
Another remarkable finding is that all related articles in support of trade-off theory use the same model to find a target adjustment coefficient to optimum debt ratio called Lambda. Lambda is the speed of adjustment with which businesses adjust their debt levels in face of transaction costs. In three studies Lambda is considerably different on datasets that should produce similar results. Namely Spanish SMEs in a similar time frame that produce a value of Lambda that varies from 0.36 to 0.86. This at the very least suggests that either the datasets are extremely different, even though it is contained to a single country and similar time-frame, or the variable is highly volatile with respect to testing. The validity and, considering the article of Shyam-Sunder and Myers (1999), statistical power of such a variable should be taken into consideration. (De Miguel and Pindado (2001), López-Gracia and Sogorb-Mira (2008), Sogorb-Mira and López-Gracia, 2003)
The main principal on which trade-off theory is founded, higher debt levels if there is more income to shield or taxes are high, is attractive theoretically for large businesses. Practically one could wonder if entrepreneurs are even considering or just plainly able to calculate the more intrinsic
13 parts of the costs and benefits of tax shields, financial distress cost and are actually able to make the trade-off (Briozzo, 2007 and Appendix I.III, p. 5-6). Empirically the evidence on the matter of tax effects and tactics is mixed. Some authors find NDTS as a significant determinant for capital structure, which could show the presence of tax tactics and explain to some extent the absence of debt. However none of the aforementioned articles find any significant evidence for a positive relation between effective tax rates and debt or a positive relation between profitability and debt to confirm the main hypotheses of TOT.
3.2 Agency theory, SME risk and information asymmetry
The remainder of this chapter will outline the general concepts of agency theory such as agency costs, moral hazard, SME specific risk and adverse selection for SMEs. In general all these issues contribute to the assumption that SMEs, especially smaller SMEs, rely more heavily on internal financing (retained earnings) and have more difficulty, either due to demand or supply constraints, with acquiring external financing. Therefore in the last subchapter information asymmetry issues and the POT will be introduced which theorises an order for internal and external finance with retained earnings as the primary source of financing for SMEs. Lastly the financial growth cycle, which combines agency costs and information asymmetry, will be discussed. The goal of reviewing these theories is to hypothesise a SME pecking order to be empirically tested.
3.2.1 Agency theory and moral hazard
Agency theory qualitatively describes the factors that impact the firms’ capital structure. External financing is amenable to transaction, contracting and agency costs which disadvantages external finance compared to internal finance. Michaelas et al. (1999) argue that agency theory is more complex in small firms. The factor that is responsible for this complexity is first and foremost the entrepreneur. Due to the fact entrepreneurs are usually manager but also major shareholder. The agency relations are therefore difficult to define by existing theory since shareholder, owner, management and employee positions can, but not necessarily are, intertwined within a single individual. Due to the role of the entrepreneur agency theory can theorise contradicting effects that can either decrease or increase agency costs dependent on how intertwined the entrepreneur is4. Furthermore a major problem with agency theory is that it is hard to test empirically and results acquired with proxies are far less straightforward than hypothesised and can generate results
4 In the thesis it is assumed that entrepreneur and SME are legally not the same entity (limited liability) unless specifically
mentioned otherwise. In the Appendix I.IV it is argued that when entrepreneur and SME are highly intertwined that the entrepreneur either is the SME (full liability), or more or as important as the SME when it comes to agency theory and moral hazard issues (Appendix I.IV, pp. 7-8).
14 consistent with several theories. This begs the question how universally true agency theory can be for SMEs. (Chittenden et al., 1996 and Myers, 2001)
One of the main issues with agency theory is moral hazard. Capital suppliers fear that when borrowers, in this case the SME, receive financing that they will start displaying undesirable
behaviour. The most severe case of moral hazard and primary credit risk is that the loan is not repaid. Since SME credits are not traded in financial markets and their value does not change until maturity, it is either full repayment or default (Dietsch and Petey, 2002). The only way suppliers of capital can get their money back (in The Netherlands) is by liquidating remaining assets if they have a prior claim, or after a civil suit if serious failure on the part of the entrepreneur can be ascertained. Less severe forms of moral hazard that can influence capital structure are: the asset substitution effect (risk shifting), the underinvestment problem and the free cash flow problem which will be covered in the next paragraph (Myers, 2001).
The asset substitution mechanism depends on entrepreneurs having a different kind of risk assessment and attitude when it comes to projects not financed by their own money. For instance, investing in riskier assets increases the upside for shareholders (the entrepreneur), the downside (default) is absorbed by creditors. This would entice entrepreneurs to increase external financing, especially debt, if they want to pursue risky projects but also dissuade banks to make a loan. The underinvestment problem, also called the debt overhang problem, suggests that entrepreneurs will reject positive NPV projects since the gain accrues to creditors rather than shareholders
(entrepreneur), which would argue lower debt levels for SMEs to avoid this problem. Finally free cash flow theory argues in favour of debt from the supply side, to diminish agency costs such as empire building and acquiring perks. A “firm pay out” model like interest on debt as substitute for “loose pay out” models, such as dividends at the discretion of the entrepreneur, forces money to creditors. The supposed result is that entrepreneurs wish to diminish debt, as to avoid any form of firm pay out model and favour equity that has a loose pay out model in the form of dividends. However that supposes that it outweighs the effects of entrepreneurs’ aversion to equity due to loss of autonomy or other personal preferences (Appendix I.II and I.IV). Lastly debt can also be used to pay out cash dividends to shareholders (entrepreneur) when “playing for time” to conceal financial problems and prevent creditors from filing for bankruptcy. The question however is how either of the
aforementioned influences affects the capital structure. Since both parties are aware of these influences it remains a question whether it is the supply side that dictates availability or the demand side that steers towards the most attractive form of finance, which leaves the total effect on capital structure uncertain. (Jenssen, 1986, Myers, 2001, Berger and Udell, 1998 and Chittenden et al. 1996, Appendix I.IV, pp. 7-8)
15 Due to the crisis years a greater likelihood of financial distress is also expected which
exacerbates moral hazard issues. In this regard size and age might play a considerable role since small (start-up) firms with no reputation or tangible assets to lose, are less discouraged to pursue very risky strategies. The smaller firms’ absence of employees also means that there is on a small group of employees to take into consideration or justify actions too. In both cases there is little to lose and possibly much to gain. For larger businesses this is arguably less the case since they have more employees, collateral and a credit reputation, thereby suggesting that a larger and older firm is less inclined to behave poorly resulting in lower agency costs. This suggest that larger businesses are internally forced to behave properly and are therefore better positioned to acquire external finance in the crises years then smaller businesses.
In the appendix it is argued that for very small SMEs and/or highly intertwined entrepreneurs agency costs are lower due to the dependence of income or a feeling of responsibility to ones
employees (Appendix I.IV pp. 7-8). Similarly to the last paragraph, but through other motivations, autonomous entrepreneurs would be internally forced to behave properly, lowering moral hazard issues. The caveat therefore with size and age is that they are not perfect proxies for entrepreneurial intertwinement and consequently the seriousness of moral hazard issues and height of agency costs. All of the arguments, in this thesis and appendix, borne from agency theory therefore have merit, but affect moral hazard and agency costs and subsequently the capital structure in ambiguous ways. This makes it very hard to construct an unifying capital structure theory out of agency theory for all Micro, Small and Medium sized SMEs. Furthermore any results found by hypotheses testing could be explained along different theoretical lines that support contradicting effects and implicitly also confirm other capital structure theories. This makes agency theory a less then suitable candidate to uniformly explain the capital structure of SMEs.
3.2.2 SME specific risk & adverse selection (ex ante)
In academic theory it is argued that beyond conventional risks, SMEs have specific risks that large companies do not have. These specific risks are an added difficulty for both external financiers and entrepreneurs since they are not easily identified, available in financial disclosure or improved upon and hinder the acquirement of external finance. Furthermore with regard to debt, the margin on small SME loans is very low and should cover the risk and incurred research costs. The more difficult it is to find out what kind of SME is applying for a loan, the lesser the chance the risk premium on a loan covers the costs.
The SME specific risks that affect the capital structure are collected from Cressy and Olofsson (1997), Ang (1992) Petit and Singer (1985) and Williamson (1996) and include: income risk, lower fixed assets ratio, a lack of diversification, financial scale disadvantages market and bargaining power
16 and the entrepreneur5. Since younger and smaller SMEs are involved in a learnings experiment their sales will be more volatile than those of larger SMEs. This increase in income risk affects SMEs ability to acquire external finance (equity) from parties with no prior claim on assets. In congruence with another SME specific risk, not having fixed assets as collateral, might also dissuade other external financiers to invest in the SME. The second SME specific risk comes from no to low diversification. Diversification is very broad and extents to product, service, supplier and customer. Obviously the risk for a non-diversified SME is captured by the dependence on a single entity: product, supplier or customer. Potential risks could be the threat of substitutes for your product or new entrants and the overreliance on a single customer/supplier. On the client side, a SMEs customer could force the price down or, on the supplier side, force the price up. This would lower profit margins and thereby the ability to pay the cost of capital and repay the principal. Subsequently SMEs simply do not have the market or bargaining power to place their product, to reach a broader group of customers, or close contract with multiple suppliers. Similarly the financial scale disadvantages limits the diversification in supply of external finance as well, which reflects the limited prospects for growth and product diversification. Any increase in SME specific risk would argue higher costs for external finance and favour internally generated resources. Additionally is likely that all of these issues become more important in a time of crises.
Adverse selection is also more likely in the SME sector mostly due to the limited (public) information available and highly differentiated growth and profit prospects of SMEs. External financiers do not know beforehand which SME-“risk/growth-type” applies for a loan or investment. Even if the SME has been profitable for several years and there is considerable supportive financial information available, it still says little about future profitability and volatility in income. This future profitability is closely linked to the diversification of the business in products, services, clients and suppliers, the professionalism with which it is run, including the entrepreneur and future (sectorial) prospects. To diminish adverse selection capital suppliers often rely on a commitment such as collateral, made by the SME or in some cases by the entrepreneur, to signal whether they are a good or bad borrower (Appendix I.VI, pp. 8-9). However one of the arguments with regard to SMEs specific risks concerns the absence of fixed assets for collateral. Small businesses, especially service
orientated ones and young businesses, with little to no reputation, have little to no tangible assets to offer up as collateral for a loan or equity (Cressy and Olofson, 1997). This means that the smaller SMEs that need collateral the most as a signal for commitment, are the least likely to have it.
5
The entrepreneur can be a SME specific risk as well as is explained in Appendix I.II (pp. 2-5 ) on capabilities and Appendix I.V (p. 8) on Entrepreneurial specific risk
17 However, as with the last chapter on agency theory and moral hazard, size and age are not perfect proxies for all the specific SME risks or adverse selection issues. For instance fixed assets for collateral might still not be available in larger/older research intensive businesses while
smaller/younger SMEs with wealthy entrepreneurs might have considerable (personal) collateral. Subsequently a large firm might not be diversified and sell only one highly sophisticated product to a few customers, while a smaller firm might be very diversified and sell many simple products to a large pool of customers. Furthermore in the Appendix (I.V, p. 8) it is argued that the entrepreneurial risk can either decrease or increase overall risk in a SME, dependent on which effects are greatest. This variety in SMEs makes the effect of size and age on specific risk and adverse selection not uniformly true. Nonetheless, in general it is assumed that most of the SME specific risks described, have a negative effect on acquiring external finance for SMEs. Especially so for the absence of collateral, since the empirical results in Table 1 and 2 show that a less tangible asset structure is negatively linked to total and long term debt levels. Furthermore it is also assumed that for most SMEs, these specific risks decrease with growth. SME specific risk likely also increase due to worsening macroeconomic conditions, resulting in an additional increase in both the difficulty of acquiring and price of external finance arguably more so for smaller SMEs. This would exacerbate the reliance on internal finance for smaller SMEs in crises periods. However, since it is likely that profits and thereby internal finance are also under pressure due to macroeconomic conditions, it leaves the question if SMEs are able to finance at all.
3.2.3 Information asymmetry
One of the main issues when it comes to SMEs acquiring finance is the availability and quality of external information by which SME credibly convey internal knowledge to outside (financial) parties. Conveying internal knowledge is meant in the broadest sense which consists of, but is not limited to: (personnel) contracts, guarantees, collateral, (personal) credit ratings and financial statements of the company. For large businesses lack of information and opaqueness are partially solved by public pricing, mandatory (financial) statements and large accountancy firms checking validity. SMEs are usually not listed on stock exchanges, have no public contracts, whether it is labour, supplier or customer contracts, have limited mandatory disclosure with respect to financial data and use “simple” bookkeepers (Berger and Udell, 1998). This could also mean that smaller SMEs are less opaque as larger SMES but would arguably shift the attention to entrepreneurs which is covered in the Appendix I.VI (p. 8). Furthermore entrepreneurs might decide on not divulging any information to outside parties simply because it takes too much time to gather, is concerned about his/her
competition or does not want the world to know how much profit the SME generates, since this number is likely closely linked to the entrepreneurs income (van der Wijst and Thurik, 1993). This all
18 argues against less opaque SMEs, especially smaller ones whom have no mandatory financial
information disclosure. Many countries have regulations in which smaller SMEs are exempt from publishing certain financial statements. This is done to lighten the administrative load and gives protection from competitors, but it does not benefit the smaller SME when it comes to transparency.
Research suggests that the low transparency of SMEs compared to large (public) firms makes it difficult for any investor or bank to validate a loan or investment. The above argument, echoed by many authors, suggest that this higher degree of information asymmetry is present within the SME sector, especially with respect to financial information (Ang, 1992, Michaelas et al., 1999, Cressy and Olofsson, 1998, Pettit and Singer, 1985). Empirical research by van Caneghem and van Campenhout (2012) shows a positive significant relationship between the quantity and quality of external financial information available and SMEs leverage ratios suggesting that availability and quality of external information is important for acquiring loans. Limited availability and quality of financial information to outside parties could lead to a misrepresentation of the SME, which would increase adverse selection issues and subsequently increase both the difficulty of acquiring and the price of external finance. Since information asymmetry decreases with size due to increased mandatory accounting disclosure for larger SMEs, it could be argued that smaller SMEs rely more heavily on internal capital than larger SMEs.
In that regard entrepreneurs personal information plays a role as well, specifically if there is more information about the entrepreneur’ credit history than about the SME6. As mentioned by Berger and Udell (1998): due to the intertwinement of the SME and entrepreneur, it could be hard to tell where the owners’ credit rating ends and the companies’ credit rating starts. The degree of informational intertwinement is likely also related to size and age of the firm. Younger, service oriented SMEs are more likely to be started up by initial funds (private capital) from the
entrepreneur or familiar investors due to both lack of credit reputation and information on credit reliability from the SME. Older and larger SMEs in general have more extensive credit reputations and financial statements, which gives the larger/older SME an information advantage when acquiring external finance.
A higher information asymmetry contributes to the difficulty ascertaining the credit reliability of the SME and obtaining external finance. Due to mandatory disclosure standards and the increased activity of older and larger SMEs in the economy, it is therefore also better linked to the proxy of size in general and age to a lesser extent. This makes information asymmetry theory applicable to SMEs
6
The extreme case in which there is no SME information (start-up) leaves only information on the entrepreneur which is referred to in Appendix I.VI (p. 8)
19 capital structure theory. However even with size and/or age as proxy for information asymmetry, it is important to note that there are other explanations not necessarily correlated to size and
information asymmetry, that might alleviate the information problem such as collateral. For instance larger and older research intensive SMEs might have high quality information but still might be unable to pledge a significant amount of tangible collateral, where a small start-up with no
reputation and limited financial statements might be funded by wealthy entrepreneurs, whom could pledge extensive collateral. A higher quality of information therefore contributes to access to external finance, but by itself it is not the only determinant that plays a role.
For external capital suppliers resolving the information asymmetry issues is hard since information gathering is difficult. Suppliers of capital incur considerable research costs to sort out the SMEs applying for capital, subsequently driving up the cost of external finance. This makes external finance less attractive and internal finance more attractive for SMEs with no reputation, higher opaqueness and limited financial statements. The increased difficulty and importance of information asymmetry is due to the fact that it is a concept that not only applies to every stage of acquiring finance, ex ante or ex post a deal, it is also applicable to both the entrepreneur and SME.
Furthermore, any other SME related risk or agency issue, adverse selection and/or moral hazard is more complex due to limited availability of external information. (Berger and Udell, 2005, 2006)
3.3 Pecking order theory (POT)
The basics of POT is founded with asymmetric information between borrower/investee and lender/investor driving the costs of capital. Since the insider (creditor/investee) has all the
information, any kind of external finance required from the outsider (debtor/investor) will be more expensive than internal finance such as retained earnings. Furthermore capital choices also reflect information to the outside which will further affects costs. If external funds are required due to positive NPV investments opportunities and lack of cash flow, businesses will issue the least
information sensitive securities first, which means debt before external equity. Vice versa if cash flow exceeds investment opportunities the surplus is used to pay down debt rather than buyback equity (Myers, 1984, p. 92). Therefore as Myers (1984, 2001) argues, businesses debt ratios reflect
cumulative requirements for external financing, contrary to the optimal target ratio the trade-off theory suggests. Financing policies from businesses therefore follow a hierarchy for internal over external funding and for debt over equity.
Businesses issue debt before equity because of information asymmetry between managers and shareholders. If managers feel stock is undervalued, they will not issue stock since the price is too low. Vice versa when managers feel stock is overvalued, the price is high, and announce an issue, shareholders assume that the price is (too) high and an immediate price drop in shares is observed.
20 This price drop of shares should not be interpreted as transaction cost for issuing stock like
underwriting spreads and other expenses, but costs bared from asymmetric information. Studies by Dierkens (1991) and D’Mello and Ferris (2000) in Myers (2001) confirm this with several proxies. The greater the price drop of shares, the larger the information asymmetry between managers and shareholders. Besides, since debt has the prior claim on assets and earnings, investors are less exposed to errors in valuing firms. Issuing debt minimizes the information advantage of managers and reduces costs and possible share price drops. Therefore external equity issues will be less attractive if debt is available on fair terms and in equilibrium debt is preferred. (Myers, 1984, 2001 and Myers and Majluf, 1984)
Initially, the POT was a way of explaining the financing practices of large (public) businesses. Scherr et al. (1993) were one of the first to consider the POT an appropriate description of SMEs’ financing practices. Subsequently articles by Holmes and Kent (1991), Norton (1991), Fama and Fench (2000), Cosh and Hughes (1994), Hall et al. (2000) and Myers (2001) suggested adjustments to the POT. One of the adjustments proposed by Holmes and Kent’s (1991) revolves around equity issues for SMEs. They suggested that due to considerable differences between large businesses and SMEs, equity is treated differently and SMEs would end up with a truncated POT. There are three motivations as to why external equity is treated differently by SMEs. Firstly an aversion to issuing equity since it would dilute entrepreneurs’ ownership. Secondly, as posited by Baumol (1965) in Myers (2001), an aversion of equity due to the discipline of equity markets. Listed companies need to abide to the rules equity markets dictate. Mandatory (financial) statements, strategic plans and shareholder meetings are all costs that have to be dealt with. Costs that are arguably greater the smaller a firm is since their financial statements are not (yet) up to par, not frequently enough composed, or require a licensed accountant. Furthermore most resources in a SME, such as time and human capital is scarce compared to large businesses. Lastly SMEs access to public equity markets is limited compared to large (public) businesses. Even if the entrepreneur might be willing to forego on a certain percentage of ownership and abide by the rules, the barrier to access a public equity market is much greater than looking for internal means or debt. Holmes and Kent (1991) call this the constrained POT in the light of aversion to dilution, abiding by public market rules and having limited access to issue additional equity7. This would not only root external equity as the least preferred type
7 Some consideration with regard to aversion of dilution and access to public equity markets is of importance. Alternative
mean of issuing equity have increased such as: credit unions, clubs of informal investors, SME specific indexes and crowd-funding. These alternatives offer easy access to equity and limited to no loss of ownership. Hamilton and Fox (1998) found under 190 respondents that 41% would look for new equity if that did not affect their control of the business, compared to 22% if it would.
21 of capital in the POT, but argue the presence of an unbridgeable (finance) gap to acquire external (public) equity for smaller firms, hence the constrained POT.
Fama and French (2000) reveal another possible flaw in the application of POT for SMEs with regard to debt. They suggest that some SMEs keep leverage low to have (low-risk) debt capacity available to finance future investments. By doing so they seem to violate the POT by making large net new issues of stock while they have debt capacity available. It could be argued, as Fama and French (2000) mention themselves, that this is mostly a concern for small businesses with potential high growth. The kind of business that needs a significant capital injection to invest in tangible assets and/or cross an entry barrier of industry. It is very likely that banks are unwilling to take that kind of risk, where a venture capitalist might see the potential gain. This would disqualify debt as a viable option for high growth and thereby change the hierarchy of the POT for high growth SMEs.
So far the POT assumed debt without a separation between short term debt (STD) and long term debt (LTD). However many articles make a separation between STD and LTD - not on the basis of POT - to isolate the effects of the chosen determinants on debt. Van der Wijst and Thurik (1993) namely found that the effects of determinants on long and short term debt cancel out when looking at total debt only. Hutchinson et al. (1998) similarly find the influence of some variables on total debt to be the net effect of opposite influences on long and short term debt.
3.3.1 Empirical and qualitative results (POT)
The most important result echoed by all research on SMEs POT is that entrepreneurs have a strong preference for retained earnings. Internal revenue comes first when they start looking for capital. Most empirical research test this relation by using cash flow, ROI or ROA as determinant. The first row in Table 1 shows that a positive cash flow is negatively related to total debt in all four articles. In Table 2 it is shown that this negative relation is especially strong for STD were 7 out of 7 articles find a negative relation. For LTD only 4 out of 7 find a negative relation, the remaining three results are non-significant. These results thereby support the POT but not solely so. All other theories that assume retained earnings as the preferred method of financing, such as agency theory and entrepreneurial preferences seem to be acknowledged by this significant relation as well.
Related research on the POT, without relying on determinants, is undertaken by Watson and Wilson (2002). They look at the actual SME capital structure instead of using specific determinants such as cash flow, age and size. They do so by using and empirical investigation based on the accounting balance of assets, liabilities and equity. Watson and Wilson (2002) construct a model in which total asset growth is a dependent variable and liabilities and equity the explanatory variable. Their research therefore not only shows how total assets growth is funded by different types of financial claims, it also shows the relative importance of each financial claim with regard to total
22 asset growth. For a sample of roughly 700 British SMEs, Watson and Wilson (2002) find that the pecking order, as suggested by the POT holds. In Figure 2 (Watson and Wilson, 2002, p. 573, Table 2) is shown that total assets is mostly funded by retained earnings followed by debt and lastly new equity (share capital). They also split their sample into “closely-held” (owner-managed) and “other firms”, in which they find that for closely held firms the pecking order remains the same but with a greater emphasis on retained earnings and remarkably, also new equity. Furthermore the interaction dummy retained profit and closely held is highly significant. These results are in line with the theory that closely held firms in which entrepreneurs are more intertwined place greater emphasis on retained earnings, due to personal preferences (Appendix I.II) and some strands of agency theory, but also supports the theory that investment through equity are more common too. In another specification, where debt is divided into certain subgroups, Watson and Wilson (2002) finds that new equity for closely held firms is actually just as important for funding as other types of debt. It is however remains unclear whether these equity contributions come from private wealth of the entrepreneur/personal relations or comes from external investors/funds.
An alternative definition which could serve as an explanation to the results found by Watson and Wilson was proposed by Ang (1992). He suggested a modified POT in which internal equity comes
FIGURE 2TABLE FROM WATSON AND WILSON (2002) IN THE JOURNAL OF BUSINESS FINANCE &ACCOUNTING
23 after retained earnings but before all other forms of external finance. Private equity or debt invested in the self-owned enterprise are more common in SMEs, since they have smaller funding needs. It is more likely that an entrepreneur of a SME has sufficient private wealth to match small funding needs. In comparison, large business funding needs are likely much greater and near impossible to fill with personal equity. In a survey by Berger and Udell (1998) on U.S. small businesses show that smaller businesses (<20 employees and <$1 million sales) are funded by internal (own) equity and/or debt for 50% and larger businesses for 30%. This would support Ang’s (1992) ranking of capital contributions from owners behind retained earnings but in front of external debt finance. The caveat he argues, similar to Berger and Udell (1998) and Zoppa and McMahon (2002), is that many private equity contributions beyond (documented) cash investments or loans are hard to measure. There are also implicit contributions such as reduced or below market pay, overtime and no dividends, simply translated into not expecting any kind of compensation for the private equity invested. Howorth (2001) finds evidence among the 13 cases interviewed that director loans including salary, bonuses and dividends was chosen as the 2nd preferred form of finance after retained profits. Howorth’ (2001) results with regard to entrepreneurs’ actual preferences, similar to work like Hamilton and Fox (1998), seem to follow the modified POT as well: entrepreneurs prefer internal finance to external finance and debt over equity.
The empirical and qualitative research above suggests that SMEs follow to great extent the POT, but that there are limitations and possible adaptations to be made such as the modified or truncated POT. However, taking into account the differences among SMEs, agency theory and the entrepreneurs’ preferences, it could very well be that an SME ends up with a temporarily or
permanently truncated pecking order or a pecking order that is dynamic and changes over time, but is not necessarily driven exclusively by pecking order theory.
3.4 The financial growth cycle
Both information asymmetry through the POT and agency theory suggest that external finance is harder to acquire for smaller than for larger SMEs. This suggests two, non-mutually exclusive arguments:
- Some demand for capital is not met, in favour of a finance gap, covered abridged in chapter 3 (pp. 6-7)
- Some types of capital are preferred over others dependent on agency cost and information asymmetry
Both would result in different capital preferences for different SMEs dependent on how strongly one of either theory fits with an SME. In general both would argue that acquiring external finance
24 large firms was firstly argued by Weston and Brigham (1972) using a life cycle approach. Small (high growth) firms started with the owners’ recourses and, if successful, could move on to trade credit and short term loans. Long term funds would be hard to access which would result to an over-reliance on short term funds and end with illiquidity. In this instance the firm would face the finance gap in which it either acquires a (costly) stock quotation or venture capital. Berger and Udell (1998, pp. 622-624) argue that the idea that firms evolve trough a financial growth cycle is well established in the literature as a descriptive concept, but that data constraints have made it difficult to
empirically analyse. To give a general idea on how different types of capital become available at different points in the Financial growth cycle (FGC), Berger and Udell adapt and update a figure from Carey et al. (1993, Fig. 10) shown in Figure 3 (1998, p. 623, Figure 1). As firms grow and mature the “pecking order” adapts, dependent on their stage in the financial growth cycle. Berger and Udell (1998) emphasize that it is neither a conclusive nor uniform growth cycle paradigm to fit all small businesses. Furthermore the sketched beginning and end of funding is not intended to be definitive.
FIGURE 3FINANCIAL GROWTH CYCLE IN THE JOURNAL OF BANKING &FINANCE (BERGER AND UDELL,1998, P. 623,FIGURE 1)
25 On the horizontal axis of Figure 3 it is shown that size, age and information availability are the three main drivers of the financial growth cycle. As the company grows/matures external agency costs would decrease, which makes access to external finance easier. Subsequently as the company grows/matures the information advantage to external parties will also decrease. Size and age are in this regard proxies for the agency costs and information asymmetry but as argued, they are not perfectly correlated.
Age is a difficult variable to describe and test due to possible high correlation with size, growth, asset structure and profitability. It is plausible that all firms that are older are also larger, have more steady income streams and have more tangible assets. However age could also mean none of the sort, for instance if the SME of 5 years old remains the same SME 20 years later.
Numbers from the SME sector reflect that many SMEs remain in a specific size class and barely grow. It could be argued that by far most SMEs do not change in size when they get older and the effects on profitability and asset structure are ambiguous. The only effect age might have on the acquiring of capital could be captured by the relationship with a bank and the possible reputation the SME has built up over the years. These age effects do however not diminish information asymmetry and agency costs beyond that relationship.
Size is a more straightforward, objective determinant than age. If a business grows it is more likely that certain aspects are highly correlated such as tangible assets and profitability. Furthermore larger firms have arguably different asset structures to support collateral and more intricate report, management and control systems which would reduce information asymmetry and agency costs. The effect of size is not straightforward for all SMEs, there might be Medium SMEs more akin to Micro SMEs than large businesses and vice versa. The general trend however suggest that larger (older) SMEs have easier access to external finance.
3.4.1 Empirical and qualitative resuts
Articles by Beck et al. (2006, 2008) and Beck and Demirguc-Kunt (2006) show that (limited) access to finance is apparent and an important growth constraint for SMEs. While researching SMEs actual capital structure internationally, they find that size, age and ownership play an important role with respect to finance obstacles. The bigger and/or older a firms is, the less they report financing
obstacles (Beck et al., 2006). Specifically, the probability that a firm lists financing as a major obstacle is 39% compared to 36% and 32% from small to medium to large. Beck and Demirguc-Kunt (2006, p.2936) subsequently show that small firms finance a smaller share of their investment with formal sources of external finance. While capital from banks is the largest share of potential investment for both small, medium and large firms, there is a considerable difference between the percentages of bank capital used by large or medium firms compared to small. From small to large the share of bank
