MASTER THESIS. Muhammad Suffian Tariq * MSc. Finance - CFA Track ANR Tilburg University. Supervisor: Professor Marco Da Rin

Transcription

1 MASTER THESIS DETERMINANTS OF LEVERAGE IN EUROPE S PRIVATE EQUITY FIRMS And Their comparison with Factors Effecting Financing Decisions of Public Limited Liability Companies Muhammad Suffian Tariq * MSc. Finance - CFA Track ANR Tilburg University Supervisor: Professor Marco Da Rin * I would like to thank my Supervisor Professor Marco Da Rin, for providing me with valuable feedback and suggestions which enhance the rigor of this academic study. I would also like to thank him and my fellow Research Assistants for helping me in collecting and organizing this unique dataset.

2 TABLE OF CONTENTS ABSTRACT... 4 I. INTRODUCTION... 4 II. LITERATURE REVIEW... 6 a. AGENCY COST THEORY... 6 i. Conflict between Shareholder & Manager... 7 ii. Conflict between Shareholder & Debt holders... 8 b. PECKING ORDER THEORY... 8 c. STATIC - TRADEOFF THEORY III. HYPOTHESIS BUILDING a. TANGIBILITY b. PROFITABILITY c. SIZE d. BUSINESS RISK e. GROWTH POTENTIAL IV. DATA DESCRIPTION V. METHODOLOGY a. DEPENDENT VARIABLES VI. RESULTS a. TANGIBILITY b. PROFITABILITY c. SIZE d. BUSINESS RISK e. GROWTH POTENTIAL f. INDUSTRY EFFECTS g. DIFFERENCES ACROSS COUNTRIES h. DIFFERENCES BETWEEN PRIVATE AND PUBLIC COMPANIES VII. CONCLUSION: VIII. REFERENCES Page 2 of 37

3 IX. TABLES AND APPENDICES TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE Page 3 of 37

4 ABSTRACT This study uses a unique data set on privately held companies to access how well the capital structure theories based on public limited liability companies apply to European Private Equity Firms. This paper provides a very good basis to compare the effect of different factors on leverage across countries and with the public limited liability companies. I find that Leverage in privately held companies is strongly positively correlated with Tangibility, whereas it is negatively correlated with Profitability, Business Risk and Growth Potential, and the effect of Size varies in different countries. Moreover, I observe that there is remarkable similarity in effects of these factors on public and private equity firms but the sensitivity of these factors vary from that of public firms; e.g. private firms are much more sensitive to Tangibility, Profitability and size as compared to Growth and Business Risk. I also find that these factors have somewhat similar correlation across countries, however, their relative importance and explanatory power differs substantially in different economies, which can be attributed to Institutional differences. I. INTRODUCTION Even after decades of work in the field, one of the basic questions in Finance still remains under discussion: What influences the financing decisions of firms? Many theories have been put forward over the period to explain the drivers that affect the capital structure of the firm but there hasn t been a consensus as to which factors are most important and how exactly do those factors affect the choice of capital structure. Capital Structure is the mix of debt, equity and hybrid securities that a firm uses to finance its operations. Firms try to finance through a combination of different kinds of securities as all these securities have a different affect on firm s value. This combination of debt and equity is chosen such that the value of the firm is maximized. The foundation of theory of capital structure was laid with the pioneering paper by Modigiliani and Miller (1958), more than five decades ago. Since then countless studies have been done by researchers to solve the Capital Structure puzzle. Several other theories such as Static Tradeoff theory, Agency Cost theory and Pecking Order theory have been put forward to further explain the financing decisions of the firm and many authors such as Myers (1984), Haris and Raviv (1991), Rajan and Zingales (1995), Lemon et al (2008) etc. have tried to empirically test these theories and come up with their own explanation of capital structure. Page 4 of 37

5 Most of the empirical studies started with the public equity firms in US, but since then studies have been conducted on other developed countries in Europe, G7 countries (Rajan and Zingales, 1995), developing countries (Booth et al., 2001), and in the recent years papers have been written on capital structure choices of almost all the major developed and developing countries. All these studies have shown remarkable consistency in the effect of some of the factors but many of the determinants had contradictory effects in different countries. This shows that the capital structure not only depends in the firm s fundamentals but it is also dependent on Industries and institutional differences between countries such as creditor protection, tax laws and bankruptcy laws. In spite of the vast research that has been done in this field, most of the research has been focused on Public Limited liability firms due to easily available data on public companies. Not much is known about the behavior of privately held companies or the effect that these well established theories have on financing choices of private firms. Private firms differ drastically from public equity firms in their access to capital. Moreover, the ownership structure of privately held firms and absence of limited liability causes them to behave quite differently from the public equity firms. In this paper I focus on private equity firms of Europe s top four economies and try to answer the following questions; - What are the factors that affect the leverage of private equity firms? - How do financing decisions of Private Equity Firms differ from those of Public Limited Liability firms? - How do the determinants of Capital Structure differ among the Big Four economies of Europe? I start by finding the factors that determine capital structure of private companies. I choose the factors whose effect is already known on leverage of public limited liability companies for the ease comparability. For the purpose of this paper, I use tangibility, profitability, size, Business risk and growth as the factors that determine the leverage of the company. I find considerable similarity between the correlation of determinants with leverage between public and private equity firms. This consistency in correlation shows that there are underlying Page 5 of 37

6 forces that affect the financing decisions of management of different companies and Industries in somewhat similar manner. The data set that I chose is quite international consisting of financial data of private equity firms of Europe s largest economies. Choosing an international data set allows me to not only observe the determinants of leverage in these countries but also gives me an opportunity to observe and analyze the cross country differences in these determinants of leverage. I organize the rest of the paper as follows. In Section II, I explain all the major theories in literature that explain the financing decisions of companies. In Section III, I construct hypothesis on the basis of the Literature Review that will be tested in the empirical study. Section IV describes the data set and collection methodology. Section V describes the detailed methodology that I follow in order to undertake the empirical study. In, Section VI, I discuss in detail the results that I obtained from my regressions and analysis. I discuss whether or not my hypothesis and expectations have been verified. Moreover, I try to answer the main questions of my paper in this part. Lastly, I give my conclusions in Section VII. Section IX contains the detailed tables and Appendices. II. LITERATURE REVIEW The literature explaining the determinants of Capital structure is spread over five decades. Some of the most prominent theories that have been put forward are Static Trade-off Theory, Pecking- Order theory and theories based on Agency Cost. The boundaries between these theories are quite blurred and they often overlap in explaining the affects of different factors in determining the financing decisions of a company. I would like to give a brief introduction to these theories and their implication on capital structure of private equity firms before I move on to my hypothesis as my hypothesis have been constructed on the bases of arguments presented by these theories. a. AGENCY COST THEORY Agency costs arise due to conflict of interest between Shareholders, Debtholders and managers. Both equity and debt can result in agency cost and a relative tradeoff between these two agency Page 6 of 37

7 cost will result in optimal capital structure for the firms. Jensen and Meckling (1976) introduced two kinds of conflict of interests that give rise to Agency Cost. i. Conflict between Shareholder & Manager Managers normally do not own 100% percent of the firm. In such cases the manger may be less interested in the well being of the firm and may have an incentive to use the company s resources for their personal gain. The reason for that is that the management doesn t capture 100% of the gains from the profit enhancement activities but they bear the entire cost of refraining from pursuing self-interest (Haris and Raviv 1991). In addition to this, as pointed out by Jensen and Meckling (1976), such management may be more interested in empire building and may invest in value destroying projects instead of investing in what is optimal for shareholders wealth. This would normally occur when the company has extra funds available after its normal operations which company can either use to pay back to stakeholders or invest in company to further help it grow. Since the dividend payment is not an obligation on the management, the management may use the surplus cash of the company to give high bonuses or perks to the top executives or may investing in negative NPV Projects, simply for the sake of empire building. Lang, Stulz and Walkling (1991) concluded in his paper that management prefers to invest all funds instead of paying it out to investors in the form of dividends. This results in overinvestment by the firms which increases the firm s size beyond what is optimal for investors. Levering up the company with debt will impose fixed periodic payments on the company which will resolve the free cash flow and the overinvestment problem. By imposing financial discipline on the management the shareholders can ensure that management utilizes the company s cash more diligently (Jensen and Meckling, 1976). Jung, Kim and Stultz (1996) concluded in their paper that there are two kinds of firms that seek to raise quity. Firstly, the firms that have high growth opportunities and they need capital to finance those projects. Secondly, there are firms that don t have valuable investment opportunities and have debt capacity. The second kind of firms give rise to agency cost as the equity is raised to benefit management rather than shareholders. Thus if a firm doesn t have high growth opportunities, raising debt can curb the agency problem. However, if the firm has high growth and investment Page 7 of 37

8 opportunities, raising equity will decrease Agency cost, as in this case Manager s interests are in line with the shareholders interest. ii. Conflict between Shareholder & Debt holders Debt may cure the conflict of interest between shareholder and management but excessive debt can be debilitating for the firm. Firstly, excessive debt may result in Debt Overhang (Jensen and Meckling, 1976) and may prevent management from making investments due to unavailability of resources or from the fear of getting bankrupt. Secondly, the shareholders may forgo positive NPV projects due to senior claim of debt holders (Myers 1977), as they would perceive debt holders as the one enjoying the benefits of investment. Thus, for companies with high investment possibilities, raising debt will reduce agency cost between management and shareholder but will increase the agency cost of debt. This kind of agency cost gives rise to the problem of underinvestment. Thus it is this tradeoff between the two agency costs that will determine the optimal capital structure for firms. One direct implication of Agency cost theory is that growth opportunities of the firms should be inversely related to capital structure of the firm. As high growth opportunities and high cash flow align managers and shareholders interest whereas debt restricts them from making favorable investments. Thus we can expect companies with low growth opportunities to have high debt ratios and companies with high growth opportunities to have low debt ratios. There is another agency cost of debt that arises due to limited liability characteristic of the publically traded companies. In case of default by limited liability companies, debt holders have a debt holders have huge downward potential as they can expect to lose their entire capital where as shareholders only lose their investment at most. In this case, the payoff of Equity holders is like a call option. This induces the shareholder to undertake risky projects or, in other words, gamble at the cost of debt holders. This phenomenon is described as asset substitution (Jensen and Meckling, 1976). Since, for the purpose of my paper, I am dealing with private non-quoted firms, we wouldn t expect private companies with unlimited liability to behave in this manner. b. PECKING ORDER THEORY Pecking order theory is based on the fact that there is an information asymmetry between corporate insiders and outside investors. Management has greater knowledge of the current Page 8 of 37

9 position of the company and also the future growth potential and opportunities of the company than the outsiders. The management s desire to raise capital through equity can be perceived as two ways by an outside potential investor. Firstly, they may see this as a signal of future growth opportunities of the firm. Secondly, a skeptical outside investors may also see it as an indicator of management s belief that the firm s equity is overvalued and my think that management is trying to take advantage of this overvaluation (Myers, 2001). Thus investors may put a lower value to the equity if management is trying to raise equity for financing. This implies that the management will have to issue equity at undervalued price to attract investors. This is one of the reasons why stocks in IPOs are normally undervalued. Undervaluing Equity would mean that the new investors would benefit at the risk of older investors and there would be a value transfer from older to new investors. Myers and Majluf (1984) suggested that managers have greater loyalty towards older shareholders and they would prefer maximizing their payoffs rather than issuing undervalued security that transfers value from older to newer shareholders. Debt has a lesser cost of Information Asymmetry and is less information sensitive of the two sources of capital. Raising debt by managers may be an indicator that the management considers its equity to be undervalued. Moreover, debt has a higher claim on assets then equity, so creditors would be less worried about the intrinsic value of the firm. Thus in case the financial condition of the firm is not too risky, raising debt will have a lesser negative impact on stock price than raising equity. Thus management, aware of the signaling effects to two major sources of capital, would consider the possibility of raising capital through debt before equity and they would go for equity only when debt is too expensive or it is unfeasible to raise further debt (Myers, 1984). Another reason why managers would prefer avoiding external financing is that it would expose them to scrutiny and monitoring of external investors. Myers (1984) noted in his paper that Firms Prefer Internal Financing over External Financing If there is a need of raising capital from outside sources, firms prefer to start with debt and then issue convertible bonds and issue equity as a measure of last resort. Now, let us see how well Pecking Order Theory predicts the choice of leverage in private equity firms. Pecking order theory suggests that greater the asymmetry of information between mangers Page 9 of 37

10 and outsiders, greater will be the cost of raising equity. As the private non-quoted firms are not required to publish their financial statements, we know there is greater opacity and hence greater information asymmetry between insiders and outside investors in case of private equity firm. Thus, we would assume that private equity firms would have a stronger inclination towards raising capital through debt than equity as compared to public equity firms. The above mentioned phenomenon is described as Level Effect by Brav (2009) which states that, for private firms relative cost of equity to debt capital is higher than that of public firms. Brav (2009) also concluded that, for private firms absolute cost of accessing capital market is higher than that of public equity firms which he dubbed as Sensitivity Effect. Private equity firms have concentrated ownership and mostly there isn t much separation between controlling shareholders and management. Thus raising equity for private equity firms would mean diluting the share of the controlling shareholders and giving up control over the business. This will also subject them to greater monitoring by the outside stakeholders. Therefore, the cost for accessing capital markets for private firms is more than that for public equity firms. This should not come as a surprise that the relative and absolute cost of equity is much more for private firms than that of public equity firms as one of the main reasons for private companies to go public is to reduce these asymmetries and get access to cheaper capital (Pagano, Paneta and Zingales, 1998). In the light of these arguments, I would expect that the financing decisions of private firms would be most accurately predicted by pecking order theory. c. STATIC - TRADEOFF THEORY Private firms do prefer debt over equity due to its less information-sensitive nature as discussed under Pecking Order Theory. In addition to that debt financing has other advantages over equity such as debt is tax detectable and a firm can have a significant tax savings if it is financed by debt. But all these benefits come at a cost and there is a limit to what a firm can lever up to. Raising debt increases financial risks of the company which may incur administrative and legal costs of bankruptcy. Tradeoff theory states that there exists an optimal capital structure where the benefit of debt such as tax savings is equally offset by financial distress caused by debt in the form of bankruptcy risk and agency cost of debt (Myers (2001)). Tradeoff theory implies that riskier firms would borrow less whereas safer firms would tend to borrow more as they can bear more debt and have greater potential of exploiting the benefit of Page 10 of 37

11 debt s tax saving characteristics. Graham (2000) showed that this is often not the case. He showed there were many large cap and safe companies that chose not to lever up in spite of the fact that they are very safe. Despite that Tradeoff theory provides a good intuitive way of thinking about the factors that might affect capital structure and we would discuss these effects in detail when I am constructing my hypothesis. III. HYPOTHESIS BUILDING Rajan and Zingales concluded a positive correlation of Leverage with Tangibility and inverse correlation with growth opportunities and Profitability, whereas the correlation with size is ambiguous. Many other studies have shown that Leverage depends on firm s characteristics such as Haris and Ranviv (1991) summarized that, Leverage increases with fixed assets, nondebt tax shields, investment opportunities and firm Size and decreases with volatility, advertizing expenditure, the probability of bankruptcy, profitability and uniqueness of the product. For the purpose of my paper I focus on the five factors; tangibility (ratio of fixed assets to Total assets), Profitability (ratio of EBIT to Total Assets), Size (Natural log of Total Assets), Growth Potential (ratio of CAPEX to total assets.) and Business Risk (Standard Deviation of ROA of the past four years). The reasons of limiting myself to only these factors is Firstly, the effect of these factors for public equity firms has already been established that would provide a good basis for comparison with my analysis of private equity firms. Secondly, using a data set of private equity firms enforces a strict constraint on availability of data; hence I am using the factors whose proxies can be established from the financial statements of private equity firms. a. TANGIBILITY Tangibility is an important factor amongst the determinants of capital structure and many scholars have shown a significant correlation between tangible assets and leverage of the firms. Harris and Raviv (1991), Rajan and Zingales (1995) and Titman and Wessels (1988) showed that the leverage of a firm is directly proportional to the tangible fixed assets in the firm. Jensen and Meckling (1976), in their paper on agency cost, suggested that raising debt encourages manager to exploit the option like behavior of equity and they tend to invest in riskier projects which give rise to agency cost of debt. Firms that have high tangible fixed assets can use their assets as collateral to get debt which reduces the agency cost of debt. In case of default, the collateralized assets provide a security to creditors. Thus creditors would be more willing to lend Page 11 of 37

12 capital to firms with high amount of tangible assets and such firms will have lower cost of debt. This makes debt cheaper and more attractive form of financing for the Firm. Scott (1977) also suggested that in order to achieve optimal capital structure a firm should issue secure (collateralized) debt as much as it can. He even went on to claim that issuing secure debt can increase the value of the firm even in absence of corporate taxes. According to his model, a firm that is seeking optimal capital structure would lever up to the extent of collateralizable assets. Hence, we would expect companies with high amounts of tangible assets to have more debt and higher leverage. Moreover, firms with high fixed assets are also able to bear more debt without the risk of bankruptcy as they are more resilient towards the economic downturns as compared to firms with intangible assets. Even in case of bankruptcy, firms with tangible assets would be worth more than the firms with intangible assets. Both Agency Cost and Static Tradeoff theory predict positive correlation of Tangibility with Leverage. The collateralizable nature of Tangible Assets helps firms to get long term financing more easily as compared to firms with intangible assets. Long Term debt is more risky for creditors as it entails greater uncertainty over longer periods. Tangible Assets provide a security to long term creditors that the company s value will not fluctuate drastically and even in case of bankruptcy, they will be worth enough for creditors to recover all or some part of their investment. Thus I would expect positive effect of Tangibility to get even stronger in case of Long Term Leverage. On the other hand firms with intangible assets will find it difficult to obtain long term debt and even if they do it will be at a higher cost. So, the firms with intangible assets will rely more on short term debt and we can expect tangibility to show a negative correlation with short term debt. Hence, according to my predictions, Tangibility should affect the firm s Long Term and Short Term Leverage ratios differently. Hypothesis 1a: Firms with greater proportion of Tangible assets will have higher Leverage. 1b: Firms with Tangible assets will have higher proportion of long Term Debt 1c: Tangibility is negatively correlated to Short Term Leverage Page 12 of 37

13 I define tangibility as the ratio of tangible fixed assets to total assets. b. PROFITABILITY Different theories have predicted the effect of profitability differently on leverage. Since more profitable firms have more earnings to shield from tax, tradeoff theory predicts that profitability should be positively correlated with debt. So, as the earnings of the firm increase they should incur more debt to take full advantage of tax deductable nature of debt. Similarly, Agency cost theory also argues in favor of positive correlation. It argues that profitable firms have high cash flows and if the management doesn t have fixed obligations to pay out cash, they might divert this excess cash towards personal perquisites or even invest in suboptimal projects. This is defined as free cash flow problem by Jensen and Meckling (1976). Jensen (1986) and Williamson (1988) stated that this can be cured by levering up the firm so that the firm has fixed obligation to pay out cash to stakeholders and the management doesn t squander firm s resources for personal gain. In contrast to these theories, pecking order theory predicts a negative correlation between profitability and leverage and provides a much sounder rationale for such behavior which is backed by literature and empirical evidence. Rajan and Zingales (1995) and Booth et al. (2001) showed empirically for public equity firms that the profitability is negatively correlated to leverage. Myers (1984) proposed that companies prefer using their internal finances for operations and if there is a need to raise capital from outside resources, firms tend to start with the safer security i.e. debt. As discussed before, information asymmetry makes raising equity much more expensive for the firm and present shareholders of the firm, thus they prefer raising capital through debt. This issue of information asymmetry is even more severe in case of private equity firms and it makes it much more difficult and costly for them to raise equity. Private equity firms have concentrated ownership structure and normally there isn t much separation between managers and shareholders. So, for them, raising capital through equity would mean giving up their control over the company which adds cost to equity (Brav, 2009). Hence, profitable firms that have enough internally generated funds would like to utilize that and raise equity as an option for last resort. Page 13 of 37

14 Thus, in case of my data I would expect a strong negative correlation between profitability and leverage. Hypothesis 2: Leverage of the firm will decrease with increase in profitability. I use the ratio of EBIT to total assets as a proxy for profitability of a firm. c. SIZE Theories generally predict a positive effect of size on leverage. However, the empirical evidence has been quite ambiguous. Size is generally seen as an inverse proxy for the probability of default. Larger firms are generally more diversified and considered safer then the smaller ones. Creditors are more willing to lend to larger firms and thus they have lower cost of debt. Due to low risk of bankruptcy for larger firms, tradeoff theory would predict a positive correlation between size and leverage. On the other hand, larger firms have lower information asymmetry. Thus larger firms have better access to capital market and are more capable of issuing information sensitive security i.e. equity (Rajan and Zingales 1995). This would reduce the cost of issuing equity and firms would be enticed to issue more equity. Hence, we would expect size to have an inverse effect on leverage. Rajan and Zingales (1995) showed both positive correlation of leverage with size for most of G7 countries but a negative correlation with Germany. Gupta (1969) illustrated empirically that size has an inverse effect on leverage. He attributed this affect of size to the high cost of raising equity in the capital market for smaller firms and also to some psychological factors such as reluctance to dilute their equity shares by issuing equity to new equity owners. Smith (1977) showed empirically that smaller firms have to pay more to raise equity and long term debt. Thus we would expect that the smaller firms would be highly levered especially with the short term debt. Moreover, Marsh (1982) s study also indicated that large firms should have higher long term leverage whereas smaller firms rely mostly on short term debt. Page 14 of 37

15 Hypothesis 3a: Size should generally affect size positively 3b: Long Term Leverage is directly proportional to Size 3c: Short Term Leverage should be inversely proportional to Size Many academic papers have used natural log of sales or natural log of assets as a proxy for size of the firm. Due to greater amount of data available on the total assets of the firm, I use natural log of total assets as a proxy. d. BUSINESS RISK Riskiness of a business can be measured by the volatility in its earnings. Many authors have used different proxies for Business Risk e.g. Titman & Wessels (1988) use standard deviation of percentage change in operating income and Booth et al. use standard deviation of return on assets and Bradely et al. (1984) used first difference in operation cash scaled by average total assets. In my paper, I use standard deviation of ROA for the past four years as a proxy for volatility and business risk. Many researchers such as Bradley, Jarrell and Kim (1984) and Titman & Wessels (1988) showed a significant and negative impact of firm s volatile earnings on the leverage of the firm. The result seems quite intuitive. As the companies need to pay fixed interest rates to debt holders in regular periods. Companies who have volatile earnings would find it difficult to pay interest consistently and would thus prefer lower debt ratios in order to avoid bankruptcy costs. Moreover, companies with volatile earning would come off as risky and unstable companies and would have higher cost of debt and thus make it difficult for them to raise debt, resulting in lower debt ratios. Thus I would expect countries to show inverse relation between Business Risk and Leverage. Hypothesis 4: Leverage is inversely proportional to the Business Risk. Page 15 of 37

16 e. GROWTH POTENTIAL Rajan and Zingales (1995) use ratio market-to-book value of equity as a proxy for future growth potential. Higher Market-to-Book ratio shows that the market attributes a high value to firm s equity due to its future growth potential and high investment opportunities. Since, for private equity firms, the market value of equity is not given, it is not possible to use this measure as a proxy for Growth. I use the ratio CAPEX to total assets as a proxy for future growth potential and assume that high investment in CAPEX in past year depicts a potential for high growth opportunities in the future. Literature normally shows negative correlation of growth opportunities of the firm with that of its leverage. The argument for this negative correlation resides in the Agency cost theory. As stated before, for the firms that have high growth potential and good investment opportunities, raising equity aligns the incentives of shareholders and managers. Whereas debt in such a firm may hinder the growth as manager pass on positive NPV project due to limited availability of cash or the risk of getting bankrupt (Jensen and Meckling, 1976). Thus Agency cost theory would predict the negative correlation of Growth opportunities with leverage. Tradeoff theory also argues in favor of negative correlation between growth and leverage. Fast growing companies have increased risk of bankruptcy. Moreover, they also have a higher bankruptcy cost due to loss of growth opportunities. Due to these high cost and probability of bankruptcy, fast growing companies tend to raise less debt. In the light of these theories and strong evidence in favor of negative relation of Growth potential with leverage in public companies, I would expect privately held companies to show similar correlation with growth potential of the firm. Hypothesis 5: High growth opportunities in the firm have negative correlation with leverage The following table shows the summary of signs of determinants of leverage as predicted by the three most prominent theories in capital structure. On the basis of these theories and present literature in capital structure, I build up my hypotheses and perform an empirical study to understand the effects of financing decision of Europe s private equity firms. Page 16 of 37

17 TABLE 1 Summary of Theories on Capital Structure and My Hypothesis The following table shows how the present theories on capital structure predict the effect of different factors on the leverage of the firm. The column on the extreme left shows the determinants of leverage. The column, Theory aggregates the prediction of different theories and past literature. On the basis of the past literature, my hypotheses are shown in the last three columns. Leverage is defined as the ratio of total interest paying debt to total assets. Tangibility is the ratio of tangible fixed assets to total assets. Profitability is EBIT divided by Book value of total assets. Size if the natural logarithm of assets. Business Risk is the standard deviation of Return on Assets of past four years. Growth is defined as the ratio of CAPEX to total assets. Agency Cost Pecking Order Static Tradeoff Theory HYPOTHESIS Leverage ST Lev LT Lev Tangibility Profitability / + - Size + - / / Business Risk Growth IV. DATA DESCRIPTION The data that I am using for my research is quite unique and provides me with the opportunity to carry out an analysis on scarcely available dataset of financial statements of private equity firms. The data for my research has been drawn from the Armadeus database, managed by Bureau van Dijk. This database contains comprehensive financial data of around 20 million European private equity firms. I focus my research on Europe s Big Four economies, namely Germany, France, Italy and Spain. I choose these countries for a number of reasons. Firstly, empirical studies for determinants of leverage for public companies of these countries have been done many times. So, that will provide a good basis for comparison between factors that affect capital structure of public and private equity firms. Secondly, these four, being the biggest economies in Europe, have the most number of private equity firms. Thus with millions of companies, the sample size is large enough for a meaningful analysis. Thirdly, considering that all these four economies are almost at the same level of development but with different laws for bankruptcy, creditor protection and taxes, it would be interesting to see if and how these institutional differences affect the determinants of capital structure in these seemingly similar economies. I prepared a 2007 cross sectional dataset. This year is appropriate as it is data set is quite recent and it is the time when the market is not much affected by the financial crisis that follows All the companies in my dataset are non-financial companies. This is a good thing because for Page 17 of 37

18 financial companies debt is not only a source of financing but it is also a part of their operations which makes such companies highly levered. It is very hard to differentiate the operating from the financing activities of the companies. The purpose of my papers is to focus on financing decisions of the firms, and inclusion of financial firms will tarnish results of empirical study. After dropping out the unnecessary financial data and years from my dataset, I filtered the data further by getting rid of the outliers and anomalies to make the dataset more appropriate for my research. The following paragraphs show the criteria that I use to filter my dataset. Since I wanted to focus on only private non-quoted firms, I dropped all the companies that were quoted. I limited my dataset to only medium sized companies that had total book value of assets of at least one million. Some of the companies in the data set had negative values of equity and debt due to accounting practices. These gave negative values of leverage which makes little sense. Hence, I dropped the companies that have negative values of debt or equity. Before, eliminating the outliers, I prepare the required variables for my regression analysis and then observe and eliminate the absurd values. I define the variable for regression as follows. Leverage: ST Leverage: LT Leverage: Tangibility: Profitability: Size: Business Risk: Growth: Interest Paying Debt / Total Assets Current Liabilities / (Equity + Current Liabilities) Long Term Debt / (Equity + Noncurrent liabilities) Tangible Fixed Assets / Total Assets EBIT / Total Assets ln (Total Assets) Standard Deviation of ROA of past four years (at industry level) Growth is defined as the ratio of CAPEX to total assets. After making my variables, I look into the outliers and try to eliminate any extreme values that may affect the results of the regression. For example some firms that have just been incorporated only recently may not have stable ratios and may show extreme ratios. I trim around 0.5 percentile of peripheral values. The final descriptive statistics are stated in Table 3. Page 18 of 37

19 There are some interesting things that can be observed from the dataset. The first thing that we observe is that all the countries have significantly higher short term leverage as compared to the long term leverage. I perform differences in means test to see if the means of two variables are indeed statistically different. I find that for my sample the firms indeed have high amount of short term leverage as compared to Long term leverage. This can be explained from the fact that the sample contains private equity firms and as we know private equity have lesser access to equity or long term debt relative to publicly held companies, this is in line with our expectations that private firms should be more reliant on short term debt. We would also see an affect of size on firm s choice of obtaining Short Term or Long Term debt. Smith (1977) noted that smaller firms should have more short term debt due to higher cost of equity and long term debt for them. It is interesting to see that Leverage in France is considerably low as compared to that of other countries. Moreover, tangibility in France is also quite low as compared to that of other countries. As tangibility is one of the main determinants of leverage, the considerably low leverage mean in France could be due to the fact that firms in France have comparatively less tangible assets. We have a huge sample size for Italian and Spanish firms. However, the sample size for Germany is severely limited by the variable standard deviation of ROA (Business Risk). Hence, I use this variable at an industry level in order to gain observations in Germany. I use industry mean SDROA for company which do not have data for business risk. V. METHODOLOGY After developing my hypothesis, I perform a cross sectional regression on my data set to observe how the five factors affect the leverage of the firm. In order to further investigate the effects of these determinants on short term and long term leverage ratios, I perform separate cross sectional regressions of these factors with Long Term and Short Term. Leverage = β 0 + β 1 Tangibility t + β 2 Profitability t + β 3 Size t + β 4 Business Risk t + β 5 Growth Potential + ε i,t Long Term Leverage = β 0 + β 1 Tangibility t + β 2 Profitability t + β 3 Size t + β 4 Business Risk t + β 5 Growth Potential + ε i,t Page 19 of 37

20 Short Term Leverage = β 0 + β 1 Tangibility t + β 2 Profitability t + β 3 Size t + β 4 Business Risk t + β 5 Growth Potential + ε i,t After identifying the factors that are statistically significant, I check for the economic significance of these variables and try to find which of these factors are most important in determining the financing decisions of firms? I also performed Analysis of Variance on these factors to see separately the explanatory power of all these variables. I further investigate as to how these factors compare with the determinants of public limited liability companies. a. DEPENDENT VARIABLES Different measures of leverage have been used over the years by different scholars in their empirical studies to find the determinants of Leverage. Leverage is measure that gives idea about what portion of firms assets are financed through debt and what proportion of firm is financed through equity. This effect can be captured through different ratios such as the ratio of total assets to total liabilities (Rajana and Zingales, 1995), ratio of debt to equity, the ratio of debt to total capital etc. The broadest definition of leverage is total liabilities over total assets as used by Rajan and Zingales (1995). This measure may overstate the leverage of the firm as total liabilities include accounts payable, trade credit, deferred taxes etc which are operating activities of the firm and not financing. These may be caused through unpaid bills or temporary mismatch in the balance of payment but they do not determine how the firm is financed. As the purpose of this paper is to find the factors that affect the financing decision of a company, I exclude any liability that is a part of operating activity from the total debt. I use the ratio of interest paying debt to total assets as a measure of Leverage. For Long Term Leverage, I use only long term debt in my ratios. I define long term leverage as the ratio of long term debt to sum of Equity and long term debt. Similarly, I exclude long term liabilities from total liabilities when calculating the short term leverage. I define short term leverage as the ratio of current liabilities to the sum of Equity and current liabilities. VI. RESULTS a. TANGIBILITY The regression in the Table 5 shows that Tangibility is positively and significantly correlated to leverage. This is in line with my expectation and the present literature on capital structure. This Page 20 of 37

21 makes intuitive sense as the firms with more tangible assets are able to use their assets as collateral and obtain more debt at lower cost as compared to firms with intangible assets. This is in coherence with studies done by Rajan and Zingales (1995), Lemon et al. (2008) and Haris and Raviv (1991). Tangibility has an even greater coefficient when regressed on Long Term debt as illustrated in Table 6. This shows that Tangibility impacts the long term debt even more strongly. Firms with more tangible assets are less risky and have higher value even in case of default. Thus creditors will be willing to give credit at better terms to firms with higher tangible assets as compared to firms with intangible assets. Moreover, tangibility is negatively correlated to short term debt at 1% significant level (Table 7), once again proving my hypothesis that firms with intangible assets rely more on short term financing as it is harder and more expensive for such firms to secure long term debt. Tangibility is one of the most important factors determining the leverage of any firm. This has already been shown in many studies. My results show that Tangibility is both statistically and economically significant determinant of leverage (Table 9). In case of Germany and France, one standard deviation increase in Tangibility increases Leverage by 34% and 35% respectively. Moreover it can also be seen that Firm s Tangibility has the most explanatory power in the regression as compared to all other factors, as shown in TABLE 10. Looking at part D of Table 10, we can see that Tangibility accounts for more than 50% of the variation in Leverage in case of Germany and France, and its explanatory power remains quite significant even in case of Italy and Spain. b. PROFITABILITY We can see in Table 5 that profitability is negatively correlated to Leverage for all countries. It remains significant and negatively correlated with short term and long term leverages as well, refer to Table 6 and Table 7. These results support the Myers s (1984) pecking order theory concluding that firms prefer internally generated funds as opposed to raising external capital. And when the external capital is needed, firms prefer to raise debt before equity. As more profitable firms have greater availability of internal funds, they have the lesser need to raise debt to finance their operations. Hence profitable firms have lower leverage as compared to less profitable firms. Page 21 of 37

22 The main reason given by Myers (1984) for firms preference for debt over equity was the Information asymmetry between insiders and outside investor. This causes the cost of equity to be high and makes raising equity an unattractive option. This case of asymmetry is even higher in case of private equity firms which would result in even higher cost of Equity. Brav (2009) concluded in his paper that private firms almost exclusively rely on debt due to high cost of equity both in relative and absolute terms. Cost of Equity is higher for private firms due to lesser access to the capital markets. Moreover, as private equity firms have concentrated ownership structure, raising equity means that the ownership of existing shareholders will be diluted. This further adds to the cost of equity as raising equity would mean giving up some control over the firm. Thus private firms would prefer using up internally generated funds and then issue debt before exercising the option to raise equity. More profitable the firm is, the greater it has the availability of internal cash flow and lesser will be the need to raise any external funds or equity. Thus it is no surprise that profitability shows strong negative correlation will Leverage and also with Short Term and Long Term Leverages. Profitability is also economically significant factor of leverage (Table 9). It has the second most important factor in explaining the variation in leverage in case of France and Italy (Table 10). c. SIZE Literature on capital structure has shown both positive and negative relation of size with Leverage. Gupta (1969) and Smith (1977) concluded that smaller firms incur higher cost to raise equity and long term debt and thus show a positive relation with equity and long term debt and a negative relation with short term debt. Spain and Italy most appropriately fit our hypothesis. They show a strong positive correlation of debt with Size which is both economically and statistically significant (Table 5 and Table 9). Moreover, they also show positive correlation with long term debt (Table 6) and negative correlation with short term debt (Table 7) which is in line with our expectations. The explanatory power of size is most prominent in case of Spain where size has higher explanatory power than any other factor in the model, even more so than tangibility, as seen in Table 10. France, on the other hand, shows a statistically significant negative correlation with leverage and long term leverage (Table 5 and Table 6) which is contradictory to my hypothesis. However, the explanatory power of size in explaining the variation in leverage is quite negligible and has little economic significance as well (Table 10 Page 22 of 37

23 Table 9), showing that size doesn t have a major impact in explaining the capital structure choices in France. Most puzzling of the results is that of Germany. Germany shows a strong negative correlation with leverage and long term leverage which is both statistically and economically significant (Table 5, Table 6 and Table 9).. According to Rajan and Zingales (1995), Germany should have a strong positive correlation with leverage. They construct this hypothesis on the basis that size is seen as an inverse proxy for expected costs of Bankruptcy. Since larger firms are more diversified and have a lower probability of getting into financial distress, they can incur more debt without the risk of bankruptcy. Moreover, the bankruptcy laws in Germany dictate a higher cost of bankruptcy, we would expect German firms to be strongly positively correlated with leverage. The fact that we see a significant negative correlation of size with leverage in case of both public and private equity firms in Germany makes us believe that there are other institutional differences between European countries which induce them to show an opposite impact of the same factor. Further research in institutional differences between countries can help us answer this mystery. This correlation of size remains the same with the long term debt (Table 6) but the correlation with short term debt is negative and significant for all countries (Table 7). This makes sense because it would be more expensive for smaller firms to raise equity and long term debt; hence, they would rely mostly on the short term debt. Size is an important determinant of Leverage and firms show high sensitivity to this factor. It can be seen as one of the three most important determinants of leverage in our model (Table 10). d. BUSINESS RISK Table 5 shows that Business Risk is significantly negatively correlated to leverage for three countries and is positively correlated to leverage of Spain. The result of all countries but Spain is in line with my hypothesis. The reason for this negative correlation is that firms that have high volatility in earnings are less able pay regular interest on payments and hence have greater possibility of defaulting on their payments. Thus such firms should incur less debt in order to avoid default or bankruptcy. Similarly firms with stable earnings and cash flows can expect to lever up and enjoy benefits of debt without the risk of bankruptcy. Hence, it makes sense that Business Risk is inversely related to leverage. Page 23 of 37

24 Column D of Table 10, however, reveals that the explanatory power of Business Risk is very negligible for all countries other than Italy. Moreover, except Italy, countries don t even show a strong economic significance for this variable, refer to Table 9. Thus we come to a conclusion that business risk is not a very important determinant for managers in determining the leverage of the company. e. GROWTH POTENTIAL I expected company s growth potential to be negatively correlated to leverage and this is exactly what I obtain from my results in Table 5. The negative correlation between Growth potential is statistically significant. It remains negatively correlated even with short term and long term leverages. This effect of growth potential is explained by the agency cost theory. Levering up the firm is considered favorable in case the firm doesn t have many growth opportunities. Because excess cash, in absence of growth opportunities, leads to free cash flow problem as managers seek perks instead of increasing shareholder s wealth (Jensen and Meckling, 1976). However, Myers (1977) noted that in case firms have high growth opportunities, debt can result in Debt Overhang as firms even forgo investing in positive NPV Projects in this case. Hence, firms that have growth opportunities do not incur high debt as concluded by my results. Growth potential doesn t have high explanatory power in case of European countries other than Spain as shown in Table 10. This is in line with the argument by Brav (2009) who showed that firm s leverage is more dependent on firm s performance (Profitability) than firm s characteristics such as Growth. Looking at Table 9 it can also be observed that Growth potential is also economically significant factor of Leverage. In the light of above results, conclude that growth is an important determinant of capital structure and firms that have high growth potential try to keep their debt ratios at a low level. The following table shows a summary of my Hypothesis and the Results obtained through regressions. Page 24 of 37