Determinants of Capital Structure
The Impact of the Financial Crisis of 2008 on German Listed Firms
S.J. van der Laan
Bachelor Thesis
10578609
31-1-2017
BSc of Finance and Organization Field: Finance
Thesis Supervisor: Dhr. Dr. J.E. Ligterink
Index
INDEX ...
ABSTRACT ... 1
1. INTRODUCTION ... 2
2.1
T
HE GLOBAL FINANCIAL CRISIS AND THEE
UROPEAN DEBT CRISIS... 4
2.3
T
HEORIES OF CAPITAL STRUCTURE... 5
2.3.2 The agency costs theory ... 6
2.3.3 Pecking order theory ... 7
2.4.1 Measures of capital structure (book and market leverage) ... 8
2.4.2 Tangibility ... 9
2.4.3 Firm size ... 9
2.4.4 Profitability ... 10
2.4.5 Growth ... 10
2.4.6 Crisis ... 11
3. METHODOLOGY, HYPOTHESIS AND DATA ... 11
3.1
H
YPOTHESES... 11
3.1.2 Firm size ... 12
3.1.3 Profitability ... 13
3.1.4 Growth ... 13
3.2
D
ATA... 13
3.3
M
ETHODOLOGY... 14
3.3.1 Fixed effects regression ... 14
3.4
D
ESCRIPTIVES
TATISTICS... 15
4. RESULTS ... 16
4.1
F
IXED EFFECTS REGRESSION COMPARED WITHR
AJAN ANDZ
INGALES(1995) ... 16
4.2
T
HE FIXED EFFECTS MODELS... 17
4.3
R
OBUSTNESS... 20
5. CONCLUSION AND DISCUSSION ... 21
REFERENCES ... 22
1
Abstract
This study investigates whether the financial crisis of 2008 did have an influence on the determinants of the capital structure of German listed firms. The determinants, which have been used, are growth, size, profitability and tangibility. A dummy variable for the crisis period is added; the crisis period is defined as the period of 2008 till 2011. Each determinant has his influence on leverage, which is used as a proxy for capital structure. Leverage is defined in this study as market leverage, book leverage and long-term book leverage.
To analyze the data a fixed effects regression is used as a technique on the panel data retained over the period 2011. Forty-four German companies that were listed on the DAX or MDAX in the period 2004-2011 have been used in the sample.
The results from the sample give different results for different calculations of capital structure. First, the results of Rajan and Zingales (1995) are compared with the regression during the whole period (2004-2011) used in this study. Comparing the results of both models shows that during the period of this study profitability becomes more important and tangibility becomes less important. When leverage is defined as market leverage, no significant results come up. When leverage is defined as book leverage or market leverage firm size has a significant positive effect, which is in line with the agency theories and the trade-off theory. Profitability has a significant negative effect, which is in line with the pecking order theory. When the effect of the crisis is measured, only one significant result has been revealed: the effect of tangibility on book leverage and long-term book leverage is significantly larger during a crisis period.
Statement of Originality
This document is written by Student Sara van der Laan who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
2
1. Introduction
Organizations and corporations fund their investments with a combination of debt capital and equity capital. The sources, which they use to finance these investments, constitute their capital structure. According to Bradley, Jarrell, and Kim (1984), an optimal capital structure is one of the most contentious issues in finance. The issue of an optimal capital structure starts with Modigliani and Miller (1958). Their theory is regarded as the base for research in capital structure. After the Modigliani and Miller theory other theories came up; such as the pecking order theory, trade-off theory and agency cost theories nevertheless, in spite of these theories capital structure represents one of the main unsolved issues in finance (Haron, 2014).
The study of Modigliani and Miller (1958) assumes perfect capital markets in which the debt-to-equity ratio has no effect on the market value of a firm. This theory is also called the irrelevance theory. The pecking order theory believes that firms prefer to finance their new investment with internal funds and will only raise debt capital if internal recourses are insufficient; when they choose for external financing, they prefer debt above issuing equity (Antoniou, Guney, & Paudyal, 2008). According to the trade-off theory, the optimal capital structure implies balancing the tax advantage of debt against the present value of bankruptcy costs (Bradley et al., 1984). In the case of agency costs, the theories of Myers (1984), Jensen and Meckling (1976) and Jensen (1986) are discussed. Myers (1977) underinvestment theory states that there is a conflict between bondholders and stockholders, which may result in firms having incentives to reject a project with a positive net present value if the benefits are for the bondholders. Another agency cost theory originates from Jensen and Meckling (1976). They claim that managers and shareholders have different interests and when they both want to maximize their utility, there is enough evidence to believe that the agent not always acts in the interest of the principal, due to their different interests. At last, it is interesting to mention the free cash flow theory of Jensen (1986), which states that increasing the leverage ratio can reduce agency costs by limiting the ability of mangers to engage in wasteful spending. In this study, the financial crisis will be linked to the theories mentioned above.
The financial crisis of 2008 started in the United States because the US subprime mortgage market collapsed in 2007 and quickly spread to Europe and other parts of the world (Aubuchon & Wheelock, 2009). Firms all around the world were challenged with stricter lending conditions and economic downturn (IMF, 2014). A recession started and consequently influenced the capital structure of different firms. The financial market was disrupted and the available amount of equity and debt capital financing was reduced (Fosberg, 2012). The main purpose of this study is to ascertain the effect of the financial crisis on firm capital structure of German non-financial firms.
All theories declare different things about the reaction of leverage on variables. To determine which theory holds the most weight during a period of financial distress, we answer, the following main question; What is the impact of the financial crisis of 2008 on the influence of the firm-specific determinants of the capital structure of German listed non-financial firms?
There are two reasons why Germany is chosen for this study:
1. According to Brun et al. (2013) Germany is a developed country with a typical bank-based economy with lower transparency and investor protection.
2. German firms employ higher levels of debt than other developed European countries. An influence of the crisis on book and market leverage is for that reason expected to occur. According to empirical
3
research of Mandaci (2009) debt levels are high during a crisis period and for that reason, higher leverage ratios are expected. Empirical research is used to investigate this.
Earlier research on the determinants of capital structure in Germany is rare, but Dang (2013) tested for evidence of the existence of the trade-off theory and the pecking order theory in Germany with empirical models. He found evidence for a better explanation of the trade-off theory in Germany than the pecking-order theory. Iqbal and Kume (2014) investigated the impact of the financial crisis on the capital structure of Germany; they found that both; equity and debt levels change during the crisis years. Besides that, they have found that firms experience an increase in their leverage during the crisis period. Fosberg (2012) investigated the capital structure of firms during firms in general during the crisis; he found that the amount of debt significantly increased.
Deesomsak, Paudyal, and Pescetto (2004) and Mandaci (2009) also researched the effects of the crisis on capital structure. Deesmosak et al. (2004) found a significant effect of the Asian crisis of 1997 on capital structure and Mandaci (2009) investigated the Turkish financial crisis and the influence on capital structure and he also found a significant effect of the crisis on capital structure decisions. In research of Akbar, Rehman, and Ormrod (2013) is found that the crisis did not influence the long-term financing channels for UK financing firms.
This study starts with an introduction of the topic, followed by a literature review; the literature review explains the global financial crisis, the European sovereign debt crisis, the different theories explaining capital structure, their determinants and their link to the financial crisis. In the third chapter the methodology, hypotheses and data are explained as well as the descriptive statistics. In the fourth chapter the results of the model are presented, discussed and analyzed. The conclusion and discussion can be found in chapter 5. The regression used in this study is based on the paper of Rajan and Zingales (1995). They investigated capital structure and their determinants in the period of 1987/1990, a similar regression is provided in Appendix C the variables of this study and their study are compared.
4
2. Literature review
The knowledge about capital structure is limited. It is one of the most fundamental questions in corporate finance. The capital structure of a firm refers to the ratio of equity and debt, which a firm uses to finance its operations. Several theories describe the optimal capital structure of a firm and suggest that firms select their choice of capital structure depending on its effects on costs of equity and debt financing (Titman & Wessels, 1988). In this chapter the causes of the financial crisis and an explanation of the different theories on capital structure will be provided. First, the global financial crisis and the European sovereign debt crisis will be discussed. Secondly, the impact on Germany will be provided. After that, the theories of capital structure and later their determinants will be presented.
2.1 The global financial crisis and the European debt crisis
The financial crisis of 2008 started in the United States. The financial crisis quickly spread to Europe and other parts of the world (Aubuchon & Wheelock, 2009). A recession started in the U.S. and other countries, and the financial market was disrupted; the available amount of equity and debt capital financing decreased (Fosberg, 2012). Banks had lack of confidence in each other's financial institutions, which created an increase in interbank lending rates. Hence, the supply of loans to non-financial institutions was reduced (Iqbal & kume, 2014). Yet what caused the financial crisis? The financial crisis was caused by debt instruments that were backed by subprime mortgage loans (Mizen, 2008). Before the crisis the subprime mortgage market in the United States experienced high growth because nonstandard mortgages were offered to individuals with nonstandard income or credit profiles. Consequently; a liquidity shortage among financial institutions was created. In July 2007, Bear Stearns (a major financial institution) announced that their assets held by two of its subprime hedge funds had become almost worthless. This was one of the first indications that defaults on subprime mortgages were going to create problems; in March 2008 Bear Stearns approached bankruptcy. The next collapse was the failure of the Lehmann Brothers (the fifth largest investment bank) and became the largest corporate collapse in American history (Fosberg, 2012). Subsequently, many financial institutions reduced their lending.
These events (collapse of Bear Sterns and Lehman Brothers) are widely agreed to have triggered the crisis in Europe. At the end of 2009, many European countries announced a higher than expected deficit on its GDP. The biggest shock was Greece with a budget deficit of 12.7% of GDP, which was more than double, than the expected value (Lane, 2012). The European sovereign debt crisis began. Greece, Ireland and Portugal all received official funding to recapitalize their banking systems. The required funding far exceeded the common IMF lending levels. As a result of both crises (the US crisis and the European crisis), the lending conditions became stricter for non-financial, firms which might have influenced their capital structure (IMF, 2014).
2.2 The financial and European crisis in Germany
The German economy is the fifth largest economy in the world and Europe’s largest. The crisis affected the European and German economy. In 2009 in all countries in the European Union GDP fell in real terms except Poland (Karanikolos et al., 2013). In 2008 the annual economic growth rate in Germany decreased to 1%, in 2009 the growth in 2009 was -4,7%. Surprisingly the consumption has been stable and over the entire crisis
5
period, the German experience was not worse than in comparable economies (Deutsche Bundesbank, 2010). The financial crisis has had the most impact on the supply side of debt. According to Fosberg (2012), the lack of confidence of banks led to an increase in the interbank lending rates and as a result of that to a reduction in the supply of loans to non-financial firms. Additionally it also had impact on the demand side of debt; taking into account agency costs, Frank and Goyal (2009) concluded that if a free cash flow decreases in downturns, the need for managerial discipline in the form of debt is reduced, which is another factor suggesting a reduced utilization of debt.
2.3 Theories of capital structure
A lot of research has been done and a lot of papers have been written about capital structure, but still the theory of Modigliani and Miller (1958) is regarded as the base for research on capital structure. They assume perfect and frictionless capital markets (no transaction costs, no agency costs, full competition and no existence of arbitrage opportunities) and proved that the value of a firm is unaffected by choice of capital structure (Myers, 2001). Thus, capital structure does not affect the total cash flow that a firm can divide to its equity and debt holders. This theory is also called the irrelevance theory (Berk & DeMarzo, 2007). Derived from these
requirements, the following equation can be formulated: !"= !$ !"= &'()* +, '- )&(*&*.*/ ,0.1
!$= &'()* +, ' (*&*.*/ ,0.1
The first proposition of Modigliani-Millers says: ‘In a perfect capital market, the total value of a firm is equal to the market value of the total cash flows generated by its assets and is not affected by its choice of capital structure’ (Berk & DeMarzo, 2007, p. 489.)
Modigliani and Miller (1958) show that this proposition is valid under the set of the following three conditions: 1. Investors and firms can trade the same set of securities at competitive markets prices equal to the present value of their future cash flows.
2. There are no taxes, transaction costs or issuance costs associated with security trading.
3. A firm’s financing decisions do not change the cash flows generated by its investments, nor do they reveal new information about them (Myers, 2001).
After Modigliani and Miller, more theories came up. In the following paragraph the trade-off theory, the agency costs theories and the pecking order theory will be discussed and compared.
2.3.1 The trade-off theory
Kraus and Litzenberger (1973) state that you have to consider a balance between the tax saving benefits of debt and the dead-weight costs of bankruptcy. So the trade-off theory says that there is an optimum in capital structure; where the increase in debt tax-shield equals the increase in the present value of bankruptcy costs.
Modigliani and Miller (1958) demonstrated that capital structure does not influence the value of a company. So the conditions for a perfect market must hold. However according to the trade-off theory of Kraus
6
and Litzenberger (1973), while Modigliani and Miller ignore corporate tax and bankruptcy costs, but corporations and investors have to pay taxes on their revenues. They pay taxes after a deduction of interest expenses so a higher level of debt can reduce tax payments. This implies that a firm can enhance its value by having more debts. The so-called tax shield (Kraus & Litzenberger, 1973), the amount by which taxable income is reduced, plays an important role in this Theory.
So the expectation is that firms prefer to fund their operations with debt. As expected there are limits to the tax benefits of debts. To receive the full benefits a firm should not use 100% debt. A firm will only receive tax benefits if it has to pay taxes. This restriction may limit the amount of debt needed as a tax shield. As stated before, debt financing puts an obligation on a firm; when a firm is not able to pay the required interest payments on the debt the firm is in default. When this happens, the debt holders are given certain rights to the assets of a firm. In the ultimate case the debt holders take legal ownership of the firm assets, which is called bankruptcy. Taking into account that equity financing does not carry this risk; the firm is not legally obliged to pay the dividends to equity holders. So levered firms risk incurring financial distress costs that reduce the cash flows available to investors (Berk & DeMarzo, 2007).
So in imperfect markets the optimal capital structure should consider the costs of financial distress and the benefits from the interest tax shield. The analysis, which is presented in this case, is called the trade-off theory. The trade-off theory refers to the idea that a company chooses its debts and equity by balancing costs and benefits. Kraus and Litzenberger (1973) state that you have to consider a balance between the tax saving benefits of debt and the dead-weight costs of bankruptcy. So the trade-off theory says that there is an optimum in capital structure; when the debt to equity ratio rises there is a trade-off between the interest tax shield and bankruptcy (Kraus & Litzenberger, 1973).
According to the trade-off theory ‘the total value of a levered firm equals the value of the firm without leverage plus the present value of the tax savings from debt, less the present value of the financial distress costs’. According to this, the following equation can be formulated.
!$= !"+ 3! 0-4*.*54 6'7 8ℎ0*(/ − 3! ;0-'-<0'( =054.*55 >+545
!"= &'()* +, '- )-(*&*.*/ ,0.1
!$= &'()* +, ' (*&*.*/ ,0.1
3! = ?.*5*-4 &'()*
This equation shows that leverage has benefits but also costs. When firms have too much debt they are more likely to risk default and incur financial distress costs. The presence of financial distress costs explains why firms choose lower debt levels than the optimal level in combination with the amount of taxes payable and differences in the magnitude of financial distress costs. The optimal ratio of leverage is accomplished when the marginal tax benefits are equal to the marginal costs of debt (Myers, 1984). The conclusion of the trade-off theory is that the optimal capital structure involves balancing the tax advantage of debt against the present value of bankruptcy costs (Bradley et al., 1984).
2.3.2 The agency costs theory
Several researchers developed theories about the agency costs of a company. Jensen and Meckling (1976) discuss conflicts between managers and shareholders (principals and agents), Myers (1977) discusses the underinvestment problem and Jensen (1986) speaks of a free cash flow theory.
7
The theory of Jensen and Meckling (1976) is based on the presence of information asymmetry because the shareholders are not able to observe the managers. This information asymmetry may prevent firms to achieve an optimal capital structure as predicted by the trade-off theory. But they both want to maximize their utility, when they both try to do this, there is enough evidence to believe that the agent not always acts in the interest of the principal, because of their different interests (Jensen & Meckling, 1976).
For example, a manager (agent) has strong incentives to invest in operations with a high pay-off expectation when this is successful, even if the chance of success is low. However, when the outcome is positive the manager will earn the gains, but when there is a negative outcome, the shareholders (principal) will suffer from most of the costs (Jensen & Meckling, 1976).
As a result agency costs will occur and influence the capital structure decisions of a firm. There are two different tools to control this agency problem; first bonding costs and a second solution can be issuing debt. By using these tools managers will be more monitored by outsiders (Barros & Ibrahimo, 2009) and the before mentioned undesirable situation will be prevented.
Another theory about agency costs is the underinvestment theory of Myers (1977). Myers states that an essential part of a firm’s value is arranged of intangible assets in the form of future investments opportunities. A firms which has outstanding bonds, might reject projects, which have a positive net present value if the benefit returns to the bondholders and not to the stockholder.
The theory of Jensen (1986) discusses the agency costs of debt and how to reduce them. He states that increasing the leverage ratio can reduce agency costs. Managers have control over future cash flows. They can promise to increase the dividend, but this is a weak promise because dividends can be reduced in the future. Instead of this action they can create debt and effectively bond their promise to pay out future cash flows. Thus, debt can be a substitute for dividends. 'By issuing debt in exchange for stock, managers are bonding their promise to pay out future cash flows in a way that cannot be accomplished by simple dividend increases. In doing so, they give shareholder recipients of the debt the right to take the firm into bankruptcy court if they do not maintain their promise to make the interest and principle payments' (Jensen, 1986, p.324). In conclusion, this theory of Jensen (1986) states that debt can reduce the agency costs of free cash flow because the amount available for spending is reduced. The probability of default on debt payments increases when managers spend the free cash flows on wasteful expenditures. Thus, managers will not invest in projects with a negative net present value and provides stimulation to utilize assets efficiently (Kochhar, 1996).
‘Managers of highly levered firms will also be less able to consume excessive perquisites since bondholders (or bankers) are inclined to closely monitor such firms. The costs associated with this agency relation may be higher for firms with assets that are less collateralizable since monitoring the capital outlays of such firms is probably more difficult. For this reason, firms with less collateralizable assets may choose higher debt levels to limit their manager's consumption of perquisites’ (Titman & Wessels 1988, p. 3).
2.3.3 Pecking order theory
The pecking order theory considers a situation in which the managers of a firm know more about the growth opportunities and its assets in place than the outside investors do (information asymmetry)(Myers, 1984; Myers & Majluf 1984). The theory explains the order in which of financing sources for an operation are used: first internal funds second debts and the last alternative rising equity (issuing shares). According to Myers and Majluf (1984) raising equity is the least preferred source to raise capital. The problem of asymmetric information occurs
8
when managers issue new shares, which gives external investors signs about the financial position of the company.
Myers and Majluf (1984) assume perfect financial markets in their investigation, except that investors do not know the real value of the existing assets or the new opportunities for investments. As a consequence, investors are not able to precisely value the securities issued to finance the new investment (Myers, 2001). So for example, when a firm announces an issue of common stock, this is good news for investors if this issue of shares makes a growth opportunity with a positive net present value visible. But it can also be bad news if this management decision is made because assets-in-place are overvalued by investors and it is tried (by the managers) to issue overvalued shares. These management decisions imply that the shares are issued at too low a price transfers value from existing shareholders to new investors. According to Myers and Majluf (1984) managers act in this situation in the interest of existing shareholders, this gives the pecking order theory of capital structure:
‘1) Firms prefer internal to external finance (Information asymmetries are assumed relevant only for external financing). `
2) Dividends are "sticky," so that dividend cuts are not used to finance capital expenditure, and so that changes in cash requirements are not soaked up in short-run dividend changes. In other words, changes in net cash show up as changes in external financing.
3) If external funds are required for capital investment, firms will issue the safest security first: debt before equity. If internally generated cash flow the surplus is used to pay down debt rather than repurchasing and retiring equity. As the requirement for external financing increases, the firm will work down the pecking order, from safe to riskier debt, perhaps to convertible securities or preferred stock, and finally to equity as a last resort.
4) Each firm’s debt ratio therefore reflects its cumulative requirement for external financing.’ (Myers, 2001, p 92).
Apart from these two theories there are more ideas about the pecking-order theory, but they are outside the scope of this study.
2.4 Determinants
In this paragraph the different determinants of capital structure that may suggest the firm’s debt-equity choice are briefly discussed. Different determinants influence the capital structure. For this research tangibility, firm size, profitability and growth are used. The possible influence of the financial crisis to the determinants will be discussed. To test the influence of the crisis a dummy will be added. Rajan and Zingales (1995), Titman and Wessels (1988) and Johnson (1998) used the same determinants in their empirical research. The crisis dummy, is added to test the influence of the financial crisis.
2.4.1 Measures of capital structure (book and market leverage)
Leverage is the dependent variable in the model and used as a proxy for capital structure. First, it is important to describe the meaning of leverage in this study. The first definition of leverage is total debt (short-term and long-term debt) divided by total assets which is defined as book leverage and used in earlier research of Rajan and
9
Zingales (1995). According to Rajan and Zingales (1995) this definition fails to incorporate the fact that there are some assets that are offset by specific non-debt liabilities.
Because the first definition of leverage is not perfect for capital structure cannot two other definitions of leverage are also used in this study. The second definition used for leverage is long-term debt divided by a sum up of long-term debt and the market value of equity, which is called market leverage, this calculation of leverage is used in empirical research of Bradley et al. (1984) The third definition is long-term debt divided by total assets which is long-term book leverage used in research of Degryse, De Goeij, and Kappert (2012) and Chen and Jian (2001). In Appendix C a table is added with the three different leverage calculations and their averages in each year.
As mentioned before, Germany is a bank-based country and German firms employ higher levels of debt. Mandaci (2009) researched the Turkish crisis and found that debt levels are high during a crisis period. Market leverage and long-term book leverage both depend on long-term debt and book leverage depends on total debt. In research of Deesomsak et al. (2004) only total debt is used for the calculation of leverage and shows a change in leverage and significant effects of TANG, GROWTH, SIZE and PROFIT. The same calculation of leverage is used in the study of Iqbal and Kume (2014), which found significant results for Germany for SIZE and GROWTH.
2.4.2 Tangibility
A tangible asset refers to an asset, which has a physical form. It includes fixed assets (machinery, buildings and land) and current assets (inventory). Tangible assets can serve as collateral to diminish the risk of the lender. Besides that, assets retain more value in liquidation. Therefore, the bigger the value of the tangible assets on the balance sheet, the more willing lenders should be to supply loans, which will give the company a higher leverage (Rajan and Zingales, 1995).
According to Myers and Majluf (1984) managers are better informed than the outside shareholders which is called the asymmetry problem. When debts are secured with collateral this can be avoided because the value of collateral is fixed. The expectation is that firms who use this advantage are able to issue more debt because the relation between tangibility and leverage is positive (Myers & Majluf, 1985). Other investigators such as Alderson and Betker (1995), Flannery and Rangan (2006), Mao (2003), Myers (1977), Scott (1977), Titman and Wessels (1988) and Rajan and Zingales (1995) also conclude a positive relation between tangibility and leverage.
According to the agency theory high-levered firms tend to underinvest and thus, transfer wealth away from debt- and equity holders. For that reason, lenders might require more collateral because the use of secured debts can help to abate this problem (Deesomsak et al., 2014).
But what will change during a period of financial distress? First, during a period of crisis it might be harder to receive a bank loan because banks will be more critical about their conditions (Fosberg, 2012). Secondly, banks and outside investors are more certain when a company has got more tangible assets in a period of crisis. Therefore, the expectation is that during financial crisis tangible assets become more important.
2.4.3 Firm size
Firm size is also a determinant of capital structure. However, a lot of researchers have different opinions about the effect. For example, Ching (1993) and Ferri and Jones (1979) say that firm size does not have any correlation
10
with leverage at all. According to Johnson (1998) and Kester (1986) firm size has a negative relation with leverage, because investors may better be informed about larger firms, which would increase the preference for equity. Other researchers believe that firm size has a positive relation with leverage because larger firms are able to issue debt easier because they are more diversified (Johnson, 1998). Rajan and Zingales (1995) also confirm that firm size has a positive relation with leverage, except Germany. Germany has a negative relation in size and leverage. The reason is unclear but it notes that larges firms should have more leverage with gives a positive relation.
According to the agency theory the relation between leverage and size is positive as well. Because large firms are more complex and less transparent agency costs are often higher (Jensen & Meckeling, 1976). For this reason the monitoring of a small firm is easier. As mentioned earlier, many investigations have proven that the relation between size and leverage is positive.
During the financial crisis banks suffered from a liquidity lack and did not carry out as much loans as before. Firm size is regarded as a variable with a negative effect on the chance of bankruptcy. So the bigger the company, the smaller the chance of bankruptcy. For this reason, banks issued loans easier to bigger companies than to smaller firms during the crisis (Berg & Kirschenmann, 2010).
2.4.4 Profitability
The discussed theories do not agree on the relationship between profitability and leverage. According to the pecking order theory the relation with leverage is negative. The pecking order theory believes that firms prefer to finance their new investment with internal funds and will only raise debt capital if internal recourses are insufficient; firms believe that issuing equity is the least option (Antoniou et al., 2008). The capability of a company to retain earnings will depend on its profitability and for that reason the predicted relation is negative (Rajan & Zingales, 1995). But the tradeoff theory predicts a positive relationship for the reason that higher profits lower the risk of bankruptcy, which implies a higher debt level because of the tax advantage (Kraus & Litzenberger, 1973). The agency theory postulates a positive relationship as well.
How can we link one of these relations to the financial crisis? The stock market prices and GDP declined deep during 2007 and 2008 (Chairman, 2011). GDP can be seen as a meter of welfare for the economy (Gleditsch, 2002). A low GDP reflects less expenditure of consumers and firms. When consumers and firms decrease their expenditure the profitability of the firms will decline.
Besides that, the stock market is also related to profitability. A turndown in the stock market prices creates a decline of the profitability of firms (Chairman, 2011). A high profitability gives more certainty to creditors to receive their money back. This becomes more important in times of crisis, because the chance to receive their invested capital back is smaller.
2.4.5 Growth
According to theory the growth opportunities of a company are influenced by their capital structure. But there are different opinions about the influence on leverage. Following the trade-off theory, the relation is negative because:
• The costs of financial distress increases with expected growth forcing managers to reduce the debt in their capital structure
11
equity leads to higher expected growth (Antoniou et al., 2008). This means that the prediction for growing firms is lower leverage, more theories agree with this prediction, for example: Rajan and Zingales (1976) and Myers (1977).
But the relation can also be positive; when growing firms do not have enough internal resources they might not be capable to finance their investments, for this reason they have to use external capital. When firms raise their external capital, they will raise debt before they issue equity according to the pecking order theory. Thus, leverage and growth are positive related (Kremp & Gerdesmeier, 1999) and (MacKay & Phillips, 2005).
These two expectations are contradictory to each other.
According to Bailey and Elliott (2009) growth opportunities can be linked with the financial crisis. In time of crisis growth will stop or decline, there are less growth opportunities, so the expectation is that growth will be less important in time of crisis.
Table 2.1
A summary of the expected influences of the determinants of capital structure on leverage during a period of a stable economy according to three different theories
Trade-off Pecking-order Agency cost
Determinant Size Growth Asset tangibility Profitability + - + + - + + - + - + +
2.4.6 Crisis
The financial crisis has influenced the worldwide economy and for that reason the period of 2004-2007 is a period without financial distress and the period afterward is influenced by financial stress. The crisis is included in the model as a dummy variable, which equals 1 during 2008-2011 and equals 0 during 2004-2008.
3. Methodology, hypothesis and data
The different theories on capital structure give insights in the decisions taken on capital structure. Several economists have conducted empirical studies on this subject. The proxy, which is used for capital structure, is the leverage ratio. In this chapter the methodology, hypotheses and the data sample will be discussed. First, a brief description of the tested hypothesis and how they are formulated will be provided, secondly the data sample will be discussed (period, selection and received from) and as last, the used model will be explained.
3.1 Hypotheses
In the chapters before, the financial crisis, different theories of capital structure and their determinants were discussed. A clear summary of the influence of these determinants is given in table 1: the expected influences of the determinants of capital structure on leverage. Different theories expect different effects of determinants on
12
leverage. Each determinant will be discussed individually. The research of Rajan and Zingales (1995) showed the empirical evidence of Germany during a period of a stable economy. In this study the empirical results of Germany during a period of stable economy is showed sequential, a period of financial distress. The hypotheses are formulated based on the different theories and determinants of capital structure.
3.1.1Tangibility
Hypothesis 1: The positive relationship between tangibility and leverage becomes weaker during the crisis period which is in line with the agency costs theory and the trade-off theory
According to empirical evidence of Rajan and Zingales (1995) in Germany the relationship between asset tangibility and leverage is positive. The positive relationship is in line with the predictions of the trade-off theory and the agency cost theory.
As described before a tangible asset can serve as collateral to diminish the risk of the lender. According to the trade-off theory an optimal capital structure involves balancing the tax advantage of debt against the present value of bankruptcy costs (Bradley et al., 1984). Thus, a high amount of tangible assets lowers the costs of bankruptcy and following the trade-off theory will also mean a higher level of debt at which the higher value of tax advantage will be equal to the bankruptcy costs.
The different agency costs theory states different expectations. For example the underinvestment problem assumes that collateral gives the opportunity to invest in more positive net present value projects, this because collateral can diminish the risk of the lender and therefore reduces the financial constraints of a firm with high assets tangibility (Myers, 1977). And collateralized debt hinders shareholders to shift from low-risk to high-risk investments or change dividend policies. As result, debt holders claim a lower risk premium on debt which reduced the agency costs and enables higher levels of debt for firms with high asset tangibility (Smith & Warner, 1979).
The different theories all agree about collateral. High asset tangibility, thus high value of collateral lowers the risk premium required by lenders and gives companies the opportunity to have higher debt levels. Since the lending conditions became stricter during the crisis period it is expected that during the crisis lenders require more collateral for the same amount of debt and for that reason the relationship between asset tangibility and leverage becomes smaller.
3.1.2 Firm size
Hypothesis 2: The negative relationship between firm size and leverage weakens during the crisis period, which is in line with the trade-off theory and the agency costs theory
According to empirical evidence of Rajan and Zingales (1995) in Germany the relationship between size and leverage is negative which is in line with the pecking-order theory. Firm size is a determinant of capital structure, however, a lot of researchers have different opinions about the effect, which is mentioned before. During the financial crisis banks suffered from a liquidity lack and did not carry out as much loans as before. Firm size is regarded as a variable with a negative effect on the chance of bankruptcy. So the bigger the company, the smaller the chance of bankruptcy. For this reason, banks issued loans easier to bigger companies
13
than to smaller firms during the crisis (Berg & Kirschenmann, 2010). Because Germany is an exception the following hypothesis is formulated.
3.1.3 Profitability
Hypothesis 3: The negative relationship between profitability and leverage will become stronger during the crisis period in line with the prediction of the pecking order theory.
According to empirical evidence of Rajan and Zingales (1995) the relation between profitability and leverage in Germany is positive. But according to the pecking order theory, there is a negative relationship between profitability and leverage of a firm. The pecking order theory states that internal financing is always preferred to external financing When firms use external funds, debt is preferred above equity.
The capability of a company to retain earnings will depend on its profitability. When this is high, the company is less dependent on debt financing since they can use their internal financial funds first to finance their investments (Myers & Majluf, 1984). Since the lending conditions to non-financial firms have been strongly tightened since 2007 (IMF, 2014), the access to the debt market decreased which made it more difficult to increase debt capital. Firms are more dependent on internal financial sources. Thus, the dependency of firms on internal financial sources increased. The expectation is that the effect between profitability and leverage will become stronger in a period of crisis and the following hypothesis is formulated.
3.1.4 Growth
Hypothesis 4: The positive relationship between growth opportunities and leverage weakens during the crisis period in line with the prediction of the agency cost theory and the trade-off theory.
According to empirical evidence of Rajan and Zingales (1995) the relation between growth and leverage in Germany is negative which is in line with the agency cost theory and the trade-off theory. According to Bailey and Elliott (2009) growth opportunities can be linked with the financial crisis. In time of crisis growth will stop or decline, there are less growth opportunities, so the expectation is that growth will be less important in time of crisis.
3.2 Data
The used dataset is obtained on Datastream and is a panel data set, which is a given sample of observations of individual companies over time (Hsiao, 2014). The data was imported to Windows Excel and the formatted version was imported to STATA (statistical analysis program), which is used to produce the statistical output. The firms included in this research are 44 nonfinancial firms of Germany listed on the DAX and MDAX, I limited my attention to the largest indexes as there were sufficient representative firms to make relevant comparisons. The data is measured over the period 2004-2011.
In the regression the time period will be separated in two periods, the first period is 2004-2007 and the second period is 2008-2011, therefore a dummy is used. All the observations without a complete record on the variables were deleted. According to Taylor (2008) the financial crisis started in August 2007 in the United States and the European debt crisis started in the summer of 2009. Therefore, the choice has been made to use data from the beginning of 2008 till the end of 2011 as the crisis period in Germany.
14
According to Rajan and Zingales (1995) financial firms do have a different capital structure therefore no financial firms are included in the sample. The outliers are removed to prevent high errors. Three fixed effects regressions will be done with different definitions of leverage, the variables are explained in appendix C; a summary of the used variables in the regression and their definitions.
There also will be a test to check if the independent variables are highly correlated (multicollinearity). The problem with multicollinearity is that it reduces the reliability of the coefficients (Bougie & Sekaran, 2010). The correlation matrix provided in Appendix A checks for the presence of multicollinearity. According to Wooldridge (2012) multicollinearity is a high correlation between the independent variable. As can be seen in the correlation matrix none of the correlation coefficients between the independent variable have a correlation higher than 0.7, which is assumed as the limit for multicollinearity (Anderson, Sweeney, Williams, Camm, & Cochran, 2013). Therefore, it can be concluded that multicollinearity is not a problem in this study.
3.3 Methodology
In this study a fixed effects regression has been carried out. The choice for this method and and not for a cross-sectional OLS-regression lies in the following reason. The cross-cross-sectional OLS regression is often used to estimate leverage as a function of firm-specific characteristics. ‘A cross-sectional study involves observations of a sample, or cross-section, of a population or phenomenon that are made in one point in time’ (Babbie, 2013, p.105). One of the assumptions in an OLS regression is that the independent variables are exogenous and thus; may not perfectly correlate with each other. When they do correlate (independent variables are endogenous) the estimators will be biased (Stock & Watson, 2003). There are ways to avoid this endogeneity for example
Deesomsak et al. (2004) lags the independent variables and calculate them as averages over the past years. However, in panel data there are better tools to avoid endogeneity. But studies, which investigate the crisis mostly, use a fixed effects panel data regression model (Akbar et al., 2013; Zarebski & Dimovski, 2012). In this chapter the fixed effects regression will be explained with a motivation.
3.3.1 Fixed effects regression
The fixed effects model tests if variables vary over time and not in company specific effects. The fixed effects model explores the relationship between predictor and outcome variables within companies. The individual company has its own individual characteristics that might influence the estimators (Torres-Reyna, 2007). The fixed effects model is used when the following assumptions are done:
1. Something within the individual company may impact or bias the estimator and we need to control for this (FE removes the effect of those characteristics so we can assess the net effect of the estimators)
2. The time-invariant characteristics are unique to the individual and should not be correlated with other individual characteristics. Each company is different and should not be correlated with the others. (Torres-Reyna, 2007).
Similar to the research of Lemmon, Roberts, and Zender (2008) is a fixed-effect panel data regression model used which includes a dummy for crisis.
The fixed effects regression used in this study is given below, all the variables are defined and explained in Appendix C; a summary of the used variables in the regression and their definitions. The results of the regression are provided in table 3.
15
@A!ABCDAE= F + GH8IJAE+ GK3BL;I6E+ GM6CNDE+ GODBLP6QE+ GR>BI8I8 + SE
To see the independent influence of the crisis and the determinants another regression is made with the crisis dummy as an interaction variable, the results of this regression are provided in table 4.
@A!ABCDAE= F + GH8IJAE+ GK3BL;I6E+ GM6CNDE+ GODBLP6QE+ GR>BI8I8 + GT8IJA ∗ >BI8I8
+ GV3BL;I6E∗ >BI8I8 + GW6CNDE∗ >BI8I8 + GXDBLP6QE∗ >BI8I8 SE
3.4 Descriptive Statistics
The descriptive statistics given in table 2 are used to describe the basic features of data in a study. The statistics are used to summarize and organize the data in an effective and meaningful way.
Table 3.1; Descriptive statistics
The data below consist of information of non-financial German firms in the period 2004-2011. The data are received from DATASTREAM. The first variable is LEVLDA which means long-term book leverage and if the first dependent variable, the second variable is LEVDA which means book leverage and is the second dependent variable, the third variable is LEVLDE which is market leverage; the third dependent variable. Correlation table can be found in Appendix A and the other variables are explained in Appendix B; a summary of the
used variables in the regression and their definitions.
Variable N Mean Median Std. Dev. Min Max
LEVLDA LEVDA LEVLDE SIZE PROFIT TANG GROWTH CRISIS SIZECRISIS PROFITCRISIS TANGCRISIS GROWTHCRISIS 344 344 344 344 344 344 344 344 344 344 344 344 0.157677 0.2175207 0.9425658 15.96306 0.1381176 0.2457851 2.245116 0.5 8.031758 0.0635736 0.1187777 1.140029 0.1542284 0.2008552 0.9969346 15.89933 0.1184313 0.2482472 1.665 0.5 6.329137 0 0.0102505 0.155 0.1065374 0.1355984 0.1784704 1.550141 0.084928 0.1245033 1.899247 0.5007283 8.115497 0.08544648 0.1458879 1.742145 0 0.0007431 0 12.3096 -0.0398798 0.0175559 0.31 0 0 -0.023793 0 0 0.5263956 0.6036628 0.9999158 18.88653 0.5972884 0.7265643 13.87 1 18.88653 0.4457988 0.6937048 13.87
16
4. Results
In this chapter the results of the same regression as Rajan and Zingales (1995) will be discussed shortly to create a clear abstract of their investigation and the differences with the current time period. After that the fixed effect regressions will be provided as well as a test of robustness. The errors used in this whole study are
heteroskedastic standard errors.
4.1 Fixed effects regression compared with Rajan and Zingales (1995)
To compare the results only one measure of leverage is used which is book leverage, the formula used is total debt divided by total assets, this for the reason that this calculation is the only one similar to their research. Rajan and Zingales (1995) used maximum likelihood and censored Tobit regression; this study uses a fixed effects regression. Besides that, the data is collected over different time periods. The equations show different results. According to Rajan and Zingales (1995) Growth, size and tangibility are significant for a firms capital structure. According to the two regression added, growth, size and profitability are significant and tangibility is not significant. During a period of crisis profitability becomes clearly more important and tangibility becomes less important. This outcome of tangibility is in line with the hypothesis stated about tangibility. During a period of crisis the positive relationship becomes smaller because more collateral is required for the same amount of debt. The dummy added for crisis does not show a significant effect, so this model does not prove a significant effect on the capital structure decisions of a firm.
Table 4.1; fixed effects-regression with the same variables as Rajan and Zingales (1995) and crisis added
Fixed effects-regression is performed with the same variables as Rajan and Zingales (1995) to show the difference with the current time period. In the second regression the variable crisis is added. Equation 1: @A!ABCDAE= F + GH8IJAE+ GK3BL;I6E+ GM6CNDE+ GODBLP6QE
and equation 2: @A!ABCDAE= F + GH8IJAE+ GK3BL;I6E+ GM6CNDE+ GODBLP6QE+ GR>BI8I8. Only one proxy for leverage is used because
this is the same as used in Rajan and Zingales (1995). Proxy for leverage is: total debt/total asset. The standard errors, reported in parentheses, are robust standard errors. *, ** And ***, significance at the 10%, 5% and 1% level, respectively. The data of Rajan and Zingales is collected in the period 1987-1990, the data of this study is collected over 2004-2011.
Equation 1 Equation 2 Rajan and Zingales (1995)
Dependent variable is LEVDA (book leverage)
Intercept Independent variable Growth Size Profitability Tangibility Dummy variable Crisis -1.087*** (0.346) 0.012* (0.006) 0.085*** (0.022) -0.665*** (0.144) 0.037 (0.125) X -0.817** (0.333) 0.013* (0.007) 0.063*** (0.021) -0.532*** (0.112) 0.022 (0.117) 0.015 (0.011) -0.20*** (0.07) -0.07*** (0.02) 0.15 (0.52) 0.42** (0.19)
17
4.2 The fixed effects models
Table 4.2 and 4.3 represents the results of the fixed effects regressions. In addition, heteroscedasticity-consistent standard errors are used to prevent incorrect inferences due to heteroscedasticity and serial correlation in both regressions. The first regression is without the interaction variable. The dependent variable is measured in three ways. When leverage is defined as market leverage none of the results are significant. The other variables, which are significant, are constantly significant in the model when book leverage and long-term book leverage are used as the dependent variable. GROWTH, SIZE and PROFIT are significant; TANG and CRISIS are both not significant. These results are not in line with earlier research of Iqbal and Kume (2014), they found the dummy variable added for crisis positive statistically significant in Germany. They have found SIZE and GROWTH statistically significant, which is the same as these results, and they have found PROFIT and TANG not significant, which is not in line with the results showed in table 4.2. Research of Deesomsak et al. (2004) did only find a significant relation of TANG, GROWTH and PROFIT in one of the investigated countries and did find significant results for a positive impact of SIZE.
In table 4.3 the fixed effects regression is provided with interaction variables with CRISIS. These results show that SIZE, PROFIT and TANGCRISIS are significant in both models. For the independent variable GROWTH and TANG no consistent results have been found, the same holds for the dummy variable of CRISIS and the interaction variables SIZECRISIS, GROWTHCRISIS and PROFITCRISIS. These results are not in line with the earlier research of Iqbal and Kume (2014) because they did not find any significant results with their interaction variable.
The effect of TANG itself on leverage is not significant. The positive significant effect of TANGCRISIS on leverage is not in line with the hypothesis about tangibility formulated. This results states that when a company has more collateral, their leverage (debt) will increase during the period of crisis and that this effect becomes stronger during a period of crisis. According to the theory the expectation was that more
collateral was necessary for the same amount of debt and for that reason made the relation weaker. The positive significant effect of SIZE on book leverage and long-term book leverage is in line with the
trade-off and agency cost theory. Unfortunately, the interaction variable with SIZE is not significant.
The negative effect of profitability on leverage is in line with the pecking-order theory; the interaction variable is not significant either.
18
Table 4.2; Fixed effects results
This table presents the estimates from the fixed effects panel regression explaining the three different measures of leverage during 2004-2011. The variables are explained in Appendix C; a summary of the used variables in the regression and their definitions. The standard errors, reported in parentheses, are robust standard errors. *, ** And ***, significance at the 10%, 5% and 1% level, respectively.First used equation is:@A!ABCDAE= F + GH8IJAE+ GK3BL;I6E+ GM6CNDE+ GODBLP6QE+ GR>BI8I8 + SE, in the second equation PROFIT is omitted.
Model Fixed
effects
Equation 1 Equation 2
Dependent variable Market leverage
Book leverage Long-term book lev Market leverage Book leverage Long-term book lev Intercept Independent variable Growth Size Profitability Tangibility Dummy variable Crisis YZ [ 0.262 (0.643) -0.017 (0.019) 0.044 (0.042) -0.409 (0.248) 0.288 (0.237) 0.0028 (0.017) 0.1568 344 -0.817** (0.333) 0.013* (0.007) 0.063*** (0.021) -0.532*** (0.112) 0.022 (0.117) 0.015 (0.011) 0.1265 344 -0.817** (0.333) 0.013* (0.007) 0.063*** (0.021) -0.532*** (0.112) 0.022 (0.117) 0.015 (0.011) 0.1265 344 0.283 (0.616) -0.021 (0.020) 0.040 (0.040) 0.276 (0.233) 0.013 (0.015) 0.1388 344 -0.788** (0.382) 0.007 (0.008) 0.057** (0.025) 0.007 (0.121) 0.028** (0.012) 0.0917 344 -0.790** (0.382) 0.007 (0.008) 0.057** (0.025) 0.007 (0.121) 0.028** (0.012) 0.0917 344
19
Table 4.3; Fixed effects results
This table presents the estimates from the fixed effects panel regression explaining the three different measures of leverage during 2004-2011. The variables are explained in Appendix C; a summary of the used variables in the regression and their definitions. The standard errors, reported in parentheses, are robust standard errors. *, ** And ***, significance at the 10%, 5% and 1% level, respectively.
@A!ABCDAE= F + GH8IJAE+ GK3BL;I6E+ GM6CNDE+ GODBLP6QE+ GR>BI8I8 + GT8IJA ∗ >BI8I8 + GV3BL;I6E∗ >BI8I8 + GW6CNDE∗ >BI8I8
+ GXDBLP6QE∗ >BI8I8 SE
Model Fixed effects
Dependent variable F-value Market leverage 2.09 Book leverage 6.05
Long-term book lev 7.41 Intercept Independent variable Growth Size Profitability Tangibility Dummy variable Crisis Interaction variable
Size and crisis Growth and crisis Tangibility and crisis Profitability and crisis
YZ [ 0.124 (0.591) -0.034 (0.0245) 0.053 (0.039) -0.165 (0.104) 0.275 (0.282) -0.114 (0.171) 0.008 (0.010) 0.032 (0.017) -0.138 (0.126) -0.380 (0.411) 0.1684 344 -1.122** (0.379) 0.005 (0.008) 0.087*** (0.024) -0.505*** (0.166) 0.001 (0.134) 0.086 (0.127) -0.008 (0.008) 0.001 (0.001) 0.204* (0.112) -0.182 (0.208) 0.2062 344 -1.009*** (0.341) 0.004 (0.006) 0.076*** (0.022) -0.370*** (0.126) -0.028 (0.132) 0.130 (0.103) -0.010 (0.007) 0.011 (0.008) 0.195** (0.105) -0.227 (0.163) 0.1370 344
20
4.3 Robustness
In this study, various approaches have been used to provide robust results. First, the descriptive statistics mentioned in table 2 are used to check for potential outliers. Besides that, this study used three different measures as a dependent variable. In addition to that, this study tried to perform a Hausman test to choose for the use of a random or fixed effects model. Unfortunately, it was not feasible to perform such a test for the reason that this is not possible with heteroscedastic standard errors. On the other hand, there are various indication that strongly suggest the use of the fixed model instead of the random effects model. According to Baltagi (2005) the fixed effects model is a decent specification if study focusses on N firms and our inference is restricted to the behavior of these set of firms, which is the case in this study. Besides that, most recent studies that focused on the crisis used a fixed effects model (e.g. Iqbal & Kume (2014); Akbar et al. (2013); Deesomsak et al. (2014))
In this paper different calculations for leverage are used. Book leverage, long-term book leverage and market leverage. The market-based calculation of leverage is used in earlier empirical research (Titman & Wessels, 1988) and is consistent with the trade-off theory (Dang, 2011). Book leverage is used to examine the robustness of the results. The results of market leverage are not significant; the results of book leverage are. SIZE, and PROFIT become significant when the dependent variable is book leverage or long-term book leverage. Besides that, also the sign in front of GROWTH changes to a positive sign instead of negative. In the correlation matrix is found that PROFIT has the biggest correlation with the independent variable. For that reason, an extra regression is provided in table 4.2 with the variable PROFIT omitted to check for possible multicollinearity. In the results is shown that GROWTH loses significance and CRISIS becomes significant, but in the first regression PROFIT is significant so an omitted variable bias will occur in the second regression. Multicollinearity can be recognized on big standard errors (Stock & Watson, 2003), the robust standard errors slightly change so multicollinearity is not a problem in this case.
21
5. Conclusion and discussion
The aim of this thesis is to investigate the impact of the financial crisis of 2008 for German listed non-financial companies. This study is examining different theories of capital structure; the trade-off theory, agency costs theories and the pecking order theory. The following question had to be answered;
what is the impact of the financial crisis of 2008 on the influence of the firm-specific determinants of the capital structure of German listed non-financial firms?
The used technique for this study is the fixed effects method over panel data in a time period from 2004 till 2011. An extra fixed effects regression is performed just to compare the results with the earlier research of Rajan and Zingales (1995) and to check if the crisis dummy had an influence on that regression. The crisis period for Germany is determined as the period started in 2008 till 2011. This for the reason that the financial crisis started in the United States in 2007 and came a little bit later to Europe and subsequently; the European sovereign debt crisis. For the panel data sample 44 German listed (DAX and MDAX) companies have been used as well as four determinants of capital structure; growth, size, profitability and tangibility.
When the regression of Rajan and Zingales (1995) is performed again in another time period no significant effects for the crisis dummy show up. Profitability becomes more important and significant which was not the case during their research and the effect of tangibility becomes smaller and is not significant anymore. This is in line with the hypothesis about profitability and also in line with the pecking order theory because according to that theory firms prefer to finance their new investment with internal funds and will only raise debt capital if internal recourses are insufficient.
The results of the fixed effects regression performed for all the calculations of leverage are as follows. Unfortunately, when leverage is defined as market leverage no significant results are revealed and more research is necessary. When leverage is defined as book leverage or long-term book leverage significant results did show up. A positive relationship between firm size and leverage, which is significant and earlier empirical research, confirms that size becomes more important in relation with leverage during a crisis, a negative relation between leverage and profitability and a positive relationship between tangibility and leverage during the crisis. The effect of tangibility itself on leverage is (not significant) really small positive or negative. The positive relation of tangibility on leverage during the crisis is not in line with the formulated hypothesis about formulated. The results show that tangibility has a bigger positive effect on leverage during a period of crisis. This is not in line with hypothesis 1 and thus, the agency theories and the trade-off theory.
The positive effect of size on book leverage and long-term book leverage is in line with the trade-off and agency cost theory. The negative effect of profitability on leverage is what the pecking order theory expects.
Therefore, this study did not find evidence of an impact of the financial crisis of 2008 on the firm-specific determinants of capital structure of German listed non-financial firms. Only a small effect on tangibility is found. More research is necessary to show a significant effect of the crisis on capital structure decisions.
In every study there are limitations and recommendations for future research; which is also the case in this study. To start with, in this study the sample existed of forty-four companies, which is a small sample. Thus, the sample may not be representative for all the listed firms in Germany. It can also be useful to increase the amount of independent variable, for example; business risk and non-debt tax shields, which can increase the R2.
22
References
Akbar, S., ur Rehman, S., & Ormrod, P. (2013). The impact of recent financial shocks on the financing and investment policies of UK private firms. International Review of Financial Analysis, 26, 59-70.
Alderson, M. J., & Betker, B. L. (1995). Liquidation costs and capital structure. Journal of Financial Economics, 39(1), 45-69.
Antoniou, A., Guney, Y., & Paudyal, K. (2008). The determinants of capital structure: capital market-oriented versus bank-oriented institutions. Journal of financial and quantitative analysis, 43(01), 59-92. Aubuchon, C. P., & Wheelock, D. C. (2009). The global recession. Economic Synopses, 2009.
Baily, M. N., & Elliott, D. J. (2009). The US financial and economic crisis: Where does it stand and where do we go from here. Brookings Institution, Jun.
Baltagi, B. H., Bratberg, E., & Holmås, T. H. (2005). A panel data study of physicians' labor supply: the case of Norway. Health Economics, 14(10), 1035-1045.
Berg, G., & Kirschenmann, K. (2010). The impact of the US financial crisis on credit availability for small firms in Central Asia.
Berger, A. N., & Di Patti, E. B. (2006). Capital structure and firm performance: A new approach to testing agency theory and an application to the banking industry. Journal of Banking & Finance, 30(4), 1065 -1102.
Dang, V. A. (2011). Testing capital structure theories using error correction models: evidence from the UK, France and Germany. Applied Economics, 45(2), 171-190.
Bradley, M., Jarrell, G. A., & Kim, E. (1984). On the existence of an optimal capital structure: Theory and evidence. The journal of Finance, 39(3), 857-878.
Deesomsak, R., Paudyal, K., & Pescetto, G. (2004). The determinants of capital structure: evidence from the Asia Pacific region. Journal of multinational financial management, 14(4), 387-405.
Degryse, H., De Goeij, P., & Kappert, P. (2012). The impact of firm and industry characteristics on small firms' capital structure. Small Business Economics, 38(4), 431-447
De Jong, A., Kabir, R., & Nguyen, T. T. (2008). Capital structure around the world: The roles of firm-and country-specific determinants. Journal of Banking & Finance, 32(9), 1954-1969.
Deutsche Bundesbank (2010): Monatsbericht Oktober 2010, Deutsche Bundesbank
Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of management review, 14(1), 57-74.
Ferri, M. G., & Jones, W. H. (1979). Determinants of financial structure: A new methodological approach. The
Journal of Finance, 34(3), 631-644.
Flannery, M. J., & Rangan, K. P. (2006). Partial adjustment toward target capital structures. Journal of financial
economics, 79(3), 469-506.
Fosberg, R. H. (2012). Capital structure and the financial crisis. Journal of Finance and Accountancy, 11, 1. Halme, M., & Huse, M. (1997). The influence of corporate governance, industry and country factors on
environmental reporting. Scandinavian journal of Management, 13(2), 137-157.
Haron, R. (2014). Capital structure inconclusiveness: evidence from Malaysia, Thailand and Singapore.
International Journal of Managerial Finance, 10(1), 23-38.
