Consequences of the Current Financial Crisis on the
Capital Structure of Non-Financial Firms Listed on the
NYSE.
Louk Kroeze
1Thesis MSc Finance
University of Groningen
Abstract
This study investigates the change in the common firm specific determinants of non-financial firm caused by the financial crisis. Using data of non-financial firms listed on the New York Stock Exchange in the period of 2002-2012. This study examines tangibility, profitability, growth prospects, size, non-debt tax shields and volatility. I find that the impact of profitability have dropped significantly during the financial crisis, while the impact of size increased significantly. In addition, I provide evidence that volatility does not have influence on a firm’s debt level. Lastly, I find evidence that smaller firms do not use more short-term debt relatively than bigger firms.
JEL classification: G15, G32
Keywords: Capital structure determinants; Financial crisis; Debt finance; Non-financial firms.
1 Master student at the University of Groningen; Faculty of Economics and Business.
-3
-In the next paragraph reviews the existing literature. First on, the explanation of different kinds of leverage is given, followed by a clarification of the firm specific determinants of capital structure. The last part of the next paragraph explains the hypothesis. The third paragraph of this research shows how the data is gathered and describes the methodology used. Paragraph four provides the findings of the research and the final paragraph gives a conclusion of this study.
2. Review of Literature
4 It holds that the use of leverage will reduce the overall cost of capital and hence also increase the value of a firm. Thus, the trade-off theory is essential in explaining the several firm-specific determinants of capital structure.
2.1 Capital Structure
The capital structure is the main interest of this research. There are several measures of capital structure, looking at different kinds of debt, maturities and market-based or book based value of equity. Rajan and Zingales (1995) apply four ratios of leverage. The first ratio is total liabilities over total assets; this could be considered as a proxy for what is left for shareholders in case of liquidation. They argue that it does not provide a good indication whether the firm is at risk of default in the near future. The second ratio is total debt to total assets, but this ratio is still influenced by the gross level of trade credit. Therefore the third ratio is total debt to net assets. Net assets are total assets less accounts payable and other liabilities, to eliminate the influence of trade credit. Although it eliminates this influence, it is still influenced by non-financing matters such as assets held against pension liabilities. Therefore Rajan and Zingales (1995) conclude that the best measure of leverage is the ratio of total debt to capital.
Titman & Wessels (1988) comment on the distinction between market and book leverage. There is a possible bias not including both leverages. They state that if managers set debt levels in terms of book value rather than market value ratios, then differences in market values, such as growth prospects, across firms will not necessarily affect the total amount of debt they issue. Since these differences do affect the market value of their equity, this will have the effect of causing firms with higher market/book value ratios to have lower debt/market value ratios.
2.2 Firm-Specific Determinants of the Capital Structure
5
2.2.1 Asset Structure
Borrowing decisions of a firm depends on the collateralization of its assets. Jensen & Meckling (1976) call this the agency cost theory; lenders take actions to protect themselves by demanding collateral as tangible assets. Tangible assets are easy to collateralize and thus they reduce the agency costs of debt. Rajan and Zingales (1995) argue that a stronger bank-firm relationship should imply a lesser role of tangibility due the better informed creditors by monitoring. In addition Hall (2011) shows that the magnitude of the relationship between asset tangibility and leverage varies substantially among countries, which depends on the restriction of collateral. In most studies, this is measured by the ratio of tangible assets over total assets which are proven to be positively related to the capital structure.
2.2.2 Profitability
Myers and Majluf (1984) argue that firms use a pecking order for financing their activities. First, by using internal funding which depends on the firm’s profitability. Followed by external funding in the use of debt, they find that equity is funding of last resort. Based on this theory, highly profitable firms with slow growth rate will end up with lower leverage ratio which means a negative correlation between profitability and capital structure. Then again, an unprofitable firm will end up with a relatively higher debt level. From another point of view, higher profitability implies lower expected costs of financial distress which results that firms are able to hold more debt based on the trade-off theory. Charalambakis and Psychoyios (2012) find for both USA and UK a significant negative relationship with leverage. Rajan and Zingales (1995) tried to look deeper in this variable and find a big difference in relationship between big and small firms; they conclude that the negative effect of earnings on leverage is considerably more important for large firms. Profitability is measured as EBITDA over total assets, which is highly significant in the survey of Titman and Wessels (1988).
2.2.3 Growth Prospects
6 Another argument is that firms prefer to issue equity instead of debt when overvaluation leads to higher expected growth. Wu and Yeung (2012) recently investigated the relationship between growth and capital structure. They find that firms with low growth prospects have significantly high leverage and firms with high growth prospects have a low leverage. This complements the traditional trade-off view on leverage. Rajan and Zingales (1995) and Wu and Yeung (2012) used the Tobin’s Q as a proxy for firm’s growth opportunities which is based on market to book ratio.
2.2.4 Size
According to the trade-off theory larger firms probably issue debt at lower cost which causes a positive relation with leverage. On the other hand Fama and Jensen (1983) argue that there may be less asymmetric information about large firms, which should increase the preference for equity by larger firms and less debt. González and González (2012) comment on the conflicting theories and state that assuming no information asymmetry the pecking order theory does not survive. Therefore they argue that the trade-off theory applies, but the greater the information asymmetry, the greater the validity of the propositions of the pecking order theory. In addition, Chung (1993) argues that larger firms may have lower agency costs associated with the asset substitution and underinvestment problems. Rajan and Zingales (1995) do not observe any correlation between size and leverage for most countries, nor is it true that large firms issue more informational sensitive securities. They therefore conclude that they do not really understand why size is correlated with leverage. Size is measured as the logarithm of sales.
2.2.5 Non-Debt Tax Shields
Modigliani and Miller (1958) state that interest tax shields create incentives for firms to increase leverage since interest expenses are tax deductible. On the other hand DeAngelo and Masulis (1980) argue that tax deductions for depreciation and investment tax credits are substitutes for these tax benefits of debt financing. The empirical evidence of Miguel and Pindado (2002) and Ozkan (2001) show a significant negative relationship, Titman and Wessels (1988) show a small insignificant negative relationship. They examine this by the ratio of depreciation over total assets as measurement of non-debt tax shield. However, Ozkan (2001) note that deprecation may also proxy other things than non-debt tax shield.
2.2.6 Volatility
7 In other words, high volatility increases the costs of financial distress; firms with higher volatility should have lower leverage. Titman & Wessels (1988) use the standard deviation of the percentage change in operating income over total assets, which resulted in a negative but insignificant relationship. Korterweg (2010) confirms this relationship in his study.
3. Hypothesis
Kantor and Holdsworth (2010) conclude in their vision on the financial crisis, that managing capital structure is the primary function to avoid default. Firms should give more focus on to what the market tells them about their capital adequacy. It’s important to understand a firm’s capital structure during times of financial distress caused by a financial crisis so firms can adapt to the market. The current financial crisis may have led to changing impact on the determinants of the capital structure or possibly made some determinants not relevant anymore.
Norden and van Kampen (2013) show that property, plant and equipment are important drivers of the collateral channel. Moreover, Chaney et. al. (2012) argue that business downturns will weaken assets values, thus reducing the capacity to borrow debt. Hall (2011) argues that the willingness of creditors rely on collateral procedures in the event of bankruptcy, this seems to affect the ability of managers to obtain credit associated with financing tangible fixed assets. Considering these studies and with the increasing leverage during the financial crisis, the expectations is that positive impact of tangibility as determinant of the amount of debt should increase due to the financial crisis.
Table 1: Impact and hypothesis of firm specific factors
This table presents the current interpretations on the firm specific factors that have impact on a firm’s capital structure and the expected change in this factor caused by the financial crisis.
8 However, the financial crisis has shown that even tangible assets, for example real estate, are not reliable as they used to be.
Hypothesis 1: The positive relationship between tangible assets and the capital structure will increase during the current financial crisis.
Another important determinant whose impact might have changed during the financial crisis is profitability. The empirical evidence for testing the pecking order theory shows a negative relation between profitability and leverage. However, it is possible to argue that as a consequence of the financial crisis that, average profits of firms listed on the NYSE have decreased. Ozkan (2001) argues that an unprofitable firm will end up with a relatively higher debt level. In addition, Murray and Vidhan (2003) show a decrease in the coefficient profitability during the 1980s which is possibly caused by the recession. The combination of the increasing debt levels during the financial crisis and the declined average of earnings of firms is fundamental to my expectation of a decline in the impact of the estimated coefficient of this variable.
Hypothesis 2: The negative relationship between profitability and leverage will reduce during the financial crisis.
9 It should be even harder for smaller firms to issue debt than for bigger firms in crisis periods because of this financing constraint. Therefore I expect an increased positive relationship with leverage during crisis periods for size. Additionally, I test the finding of Chan et al. (1985) that smaller firms borrow more short term debt relative to bigger firms.
Hypothesis 4: The positive relationship between size and leverage increases during the financial crisis.
For non-debt tax shield I don’t expect any changes as long as the tax shield didn’t chance. Despite this fact, it could be possible that due the decreasing effect of other determinants the effect of non-debt tax shield increased. Furthermore, Barclay and Smith (1995) state there is a possible relationship with non-debt tax shield and other determinants as firms with higher depreciation ratios are also more likely to have relatively fewer growth opportunities and more tangible assets.
Hypothesis 5: There is no change in the relationship between non-debt tax shield and leverage caused by the financial crisis.
According to the trade-off theory volatility negatively related with a firm’s capital structure. High volatility increases the costs of financial distress; firms with higher volatility should have lower leverage especially during a financial crisis. Hence, we can expect an increasing impact of the determinant volatility on leverage during the crisis with respect to the period before the crisis.
Hypothesis 6: The negative relationship between volatility and leverage increases during the financial crisis.
10
4. Methodology & Data
4.1 Data
Data of firms in the centre of the financial crisis are analysed, therefore the data is derived from the New York Stock Exchange (NYSE). NYSE is by far the world’s largest stock exchange by market capitalization with 1421 listings (NYSE, 2013) in the United States. Both the data of the levels of debt and the explanatory variables are exported from Thomson’s DataStream. Table 2 presents the breakdown of the sample in industry.
Table 2: Industry Breakdown
This table presents a breakdown of the 2003 to 2012 sample into industries defined by DataStream. The observed number of firms (# Firms) and firm-observations (N) are reported for each industry.
Industry Code # Firms N
Aerospace 1300 9 129 Apparel 1600 11 109 Automotive 1900 22 228 Beverages 2200 9 82 Chemicals 2500 66 601 Construction 2800 52 493 Diversified 3100 19 537
Drugs, Cosmetics & Health care 3400 76 363
Electrical 3700 19 170
Electronics 4000 138 817
Financial (real estate, lease) 4300 4 19
Food 4600 53 187
Machinery & Equipment 4900 49 453 Metal Producers 5200 53 205 Metal Product Manufacturers 5500 19 174 Oil, Gas, Coal & Related services 5800 124 1117
Paper 6100 50 171
Printing & Publishing 6400 18 178
11 Financial firms, like banks and insurance companies, are eliminated from the sample as their leverage is influenced by other aspects such as capital requirements. The sample period is 2003-2012 which will be split up in two parts, because the data between pre-financial crisis and during financial crisis need to be compared. The first part, preceding the financial crisis contains 2003-2007. The second part, during the financial crisis includes 2008-2012. Setting these limits of non-financial firms and erasing all the firms with incorrect data, has led to a sample of 1274 firms currently listed on the NYSE. Hereby, exit of firms are neglected in the final sample, this exclusion could affect the representativeness of the final sample. Even so, the period of the sample is relatively small and therefore I do not expect this survivor bias has a major impact.
4.2 Methodology
In order to determine the impact of the determinants of the capital structure, I use a panel data analysis. Greene (2003) defines panel data analysis as a technique which uses cross data of the time dimension to predict the economic relations. This analysis makes use of the data which has both time dimension and cross section dimension, which will lead to more accurate results. The regressions are controlled for year-fixed effects and industry effects. Year fixed effects should be used to control for changes of the average value of the variables over time (Brooks, 2008). Industry effects should be used to control for the average differences across industries.
Table 3: Descriptions and calculation of the data
This table presents the symbols, description and calculation of the variables used in this study.
Symbol Description Calculation
BLEV Book Leverage Total Debt/Book Value Equity + Total Debt MLEV Market Leverage Total Debt/Market Value Equity + Total Debt TAN Tangibility Property, Plant & Equipment/Total Assets PRF Profitability EBITDA/Total Assets
GOP Growth Prospects Market-to-Book Ratio
SIZ Size Logarithm of Sales
NTS Non-Debt Tax Shield Depreciation/Total Assets
12 The models are based on the following equation:
(1)
where i denotes the cross-sections and t denotes time-period. So i takes into account the number of firms and t the number of years. The vector Xit represents the explanatory variables, Dit is the level
of debt and µit is the error term. Adding the firm specific determinants:
(2)
where, leverage (D) is the dependent variable in this research. Two measurements of leverage are used in this study; book leverage and market leverage. Book leverage is defined as the book value of debt over the book value of debt plus the book value of equity, whereas for market leverage the market value of equity is used. Tangibility (TAN) is measured as the property, plant and equipment divided by the total assets. Profitability (PRF) is measured as the earnings before interest, taxes, depreciation and amortization over total assets. Growth Prospects (GOP) is proxied by market to book ratio. Size (SIZ) is defined as the natural logarithm of sales. Non-Debt Tax Shield (NTS) is measured as depreciation divided by total assets. Volatility (VOL) is the standard deviation of the change in operating profits with respect to the previous 5 years.
13
4.3 Summary Statistics
Table 4: Summary Statistics
The table presents the summary statistics of the variables used in this study. The sample contains 1274 unique firms and 10691 firm-year observations from 2003 to 2012. All the variables are winsorized 1th and 99th percentiles. Book leverage is defined as the value of debt over the value of debt plus the book value of equity, for market leverage the market value of equity is used. Tangibility is property, plant and equipment divided by the total assets. Profitability is measured as EBITDA over total assets. Growth is proxied by market to book ratio. Size is the natural logarithm of sales. Non-debt tax shield is depreciation divided by total assets. Volatility is the standard deviation of the change in operating profits with respect to the previous 5 years. In panel B the correlation between de independent variables are shown as a test for multicollinearity.
Panel A: Descriptive Statistics
Percentile N Mean Std. Dev. Median 25th 75th
Book Leverage 10285 0.37 0.25 0.36 0.19 0.52 Market Leverage 9935 0.26 0.22 0.22 0.09 0.40 Tangibility 10420 0.35 0.26 0.28 0.14 0.54 Profitability 10457 0.12 0.10 0.12 0.08 0.17 Growth 9633 2.58 2.35 1.91 1.30 3.00 Size 10878 14.31 1.75 14.36 13.29 15.45 Tax Shield 8229 0.04 0.02 0.03 0.02 0.05 Volatility 8927 2.05 4.36 0.63 0.24 1.89 Panel B: Correlation
Tangibility Profitability Growth Size Tax Shield Volatility Tangibility 1 Profitability 0.0686 1 Growth -0.1037 0.2909 1 Size -0.0958 0.1503 0.0749 1 Tax Shield 0.4928 0.1510 0.0169 -0.1009 1 Volatility 0.0244 -0.1920 -0.0772 -0.1044 0.0828 1
15
Table 5: Pre-Financial Crisis vs. During Financial Crisis
The table presents the descriptive statistics of the market leverage and the book leverage per year. The sample contains 1274 firms and 10691 firm-year observations from 2003 to 2012. Book and market leverage are winsorized at 1th and 99th. The period of pre-financial crisis contains 2003-2007, during pre-financial crisis contains 2008-2012. Panel C presents the average of each variable for both periods. In addition the difference between the two periods are tested by a t-test: two-sample assuming equal variances. *, **, *** indicate significance at 10%, 5% and 1% confidence level.
Panel A: Book Leverage 2003-2007 Panel B: Book Leverage 2008-2012
Percentile Percentile
Year Mean Std. Dev. Median 25th 25th Year Mean Std. Dev. Median 25th 75th 2003 0.38 0.24 0.37 0.21 0.52 2008 0.41 0.27 0.40 0.22 0.57 2004 0.36 0.24 0.34 0.20 0.49 2009 0.38 0.27 0.36 0.18 0.54 2005 0.36 0.24 0.34 0.18 0.49 2010 0.36 0.26 0.34 0.17 0.52 2006 0.36 0.24 0.35 0.19 0.50 2011 0.38 0.27 0.37 0.18 0.54 2007 0.36 0.24 0.35 0.19 0.50 2012 0.38 0.26 0.37 0.18 0.53
Panel C: Market Leverage 2003-2007 Panel D: Market Leverage 2008-2012
Percentile Percentile
Year Mean Std. Dev. Median 25th 75th Year Mean Std. Dev. Median 25th 75th 2003 0.26 0.20 0.22 0.10 0.38 2008 0.35 0.26 0.33 0.13 0.54 2004 0.22 0.18 0.19 0.09 0.34 2009 0.29 0.23 0.25 0.10 0.43 2005 0.22 0.19 0.18 0.08 0.33 2010 0.26 0.21 0.22 0.09 0.38 2006 0.22 0.18 0.18 0.08 0.33 2011 0.28 0.23 0.24 0.10 0.41 2007 0.23 0.19 0.19 0.08 0.34 2012 0.29 0.23 0.26 0.11 0.43
Panel E: Pre-Financial Crisis vs. During Financial Crisis
2003-2007 2008-2012 T-stat 2003-2007 2008-2012 T-stat BLEV 0.36 0.38 -3.29*** GOP 2.84 2.37 9.83*** MLEV 0.23 0.29 -14.09*** SIZ 14.27 14.35 -2.54** TAN 0.36 0.35 1.21 NTS 0.04 0.03 2.29** PRF 0.13 0.12 9.78*** VOL 1.94 2.15 -2.23**
16 In addition, Table 5 presents a significant decrease in profitability and growth prospects and significant increase in size, non-debt tax shields and volatility. Tangibility is the only variable that didn’t change between the two periods.
Figure 2. Capital Structure over time. This figure shows the movement of the book and market leverage over the full sample period. Book leverage is defined as the value of debt over the value of debt plus the book value of equity, for market leverage the market value of equity is used
5. Findings
5.1 Regression
17
Table 6: Regression Coefficient Estimates
This table reports estimates of panel least squares regressions controlling for year fixed effects. *, **, *** indicate significance at 10%, 5% and 1% confidence level. Book leverage is defined as the value of debt over the value of debt plus the book value of equity, for market leverage the market value of equity is used. Tangibility is property, plant and equipment divided by the total assets. Profitability is measured as EBITDA over total assets. Growth is proxied by market to book ratio. Size is the natural logarithm of sales. Non-debt tax shield is depreciation divided by total assets. Volatility is the standard deviation of the change in operating profits with respect to the
previous 5 years. All variables are winsorized at 1th and 99th percentile. Panel C present the difference between the coefficients using
the Wald’s restriction test.
Panel A: 2003-2007 Panel B: 2008-2012 Panel C: Wald's Restriction test BLEV MLEV BLEV MLEV BLEV MLEV
Constant -0.182*** 0.026 -0.182*** 0.026 Tangibility 0.246**** 0.249*** 0.290*** 0.31*** 2.87* 8.44*** Profitability -0.955*** -0.778*** -0.747*** -0.76*** 9,28*** 0,90 Growth 0.024*** -0.010*** 0.028*** -0.010*** 3,00* 0,12 Size 0.031*** 0.016*** 0.037*** 0.021*** 2,81* 2,45 Tax Shield -0.242 -0.376** -0.656*** -0.586*** 1,80 0,68 Volatility -0.001 0.000 0.003*** 0.004*** 6,87*** 10,90*** Number of Observations 2769 2890 2769 2769 5538 5659 Adjusted R² 0.22 0.28 0.22 0.28 0.80 0.82 F-Stat 82.48*** 120.33*** 82.48*** 120.33*** 28.28*** 28.71***
Table 6 represents the regression coefficients controlling for year fixed effects. Year fixed effects should be used in a model where the average value of the dependent variable changes over time. Table 5 and Fig. 2 show that the average level of debt changes during the sample period. From Table 6 panel A, for book leverage and market leverage tangibility, profitability, growth and size show a significant relation. However, growth prospects display a positive relation with book leverage. Tax shield lacks significance with book leverage and volatility with both book and market leverage. However, volatility shows a significant relation during the financial crisis in Table 6 panel B. Moreover, the period 2008-2012 show significance for all the variables. Table 6 Panel C shows there is not much significant difference between the coefficients of the two periods. The coefficients of tangibility and volatility deviate significantly in relation with market leverage with respect to the two sample periods. In relation with book leverage, profitability and volatility show difference at one percent confidence level and size and tangibility at ten percent confidence level.
18
Table 7: Regression Coefficient Estimates
This table report estimates of panel least squares regressions controlling for industry effects. *, **, *** indicate significance at 10%, 5% and 1% confidence level. Book leverage is defined as the value of debt over the value of debt plus the book value of equity, for market leverage the market value of equity is used. Tangibility is property, plant and equipment divided by the total assets. Profitability is measured as EBITDA over total assets. Growth is proxied by market to book ratio. Size is the natural logarithm of sales. Non-debt tax shield is depreciation divided by total assets. Volatility is the standard deviation of the change in
operating profits with respect to the previous 5 years. All variables are winsorized at 1th and 99th percentile. Panel C present the
difference between the coefficients using the Wald’s restriction test.
Panel A: 2003-2007 Panel B: 2008-2012 Panel C: Wald's Restriction test
BLEV MLEV BLEV MLEV BLEV MLEV
Constant -0.206*** 0.027 -0.206*** 0.027 Tangibility 0.199*** 0.156*** 0.228*** 0.192*** 1.17 2.49 Profitability -0.865*** -0.692*** -0.735*** -0.709*** 3.32* 0.09 Growth 0.024*** -0.011*** 0.027*** -0.010*** 1.06 0.27 Size 0.036*** 0.018*** 0.035*** 0.020*** 1.21 6.46** Tax Shield -0.138 -0.013 -0.518*** -0.374*** 1.51 1.96 Volatility 0.001 0.001 0.002*** 0.003*** 1.72 4.24** Number of Observations 2769 2890 2769 2769 5538 5659 Adjusted R² 0.25 0.31 0.25 0.31 0.25 0.31 F-Stat 65.58*** 90.44*** 65.58*** 90.44*** 65.58*** 90.44***
Table 7 presents the coefficients controlling for industry effects while using crisis dummies to control for the time effects. The first period, Table 7 panel A, provides evidence on a significant relationship of tangibility, profitability, growth and size with both book and market leverage. For the crisis period, Table 7 panel B, tax shield and volatility also show a significant relationship. Controlling for industry effects gives less significant differences in coefficients between the two periods. Profitability shows a significant difference in relation with book leverage at a ten percent confidence level. In relation with market leverage, size and volatility show a significant difference between the two sample periods at a confidence level of five percent. With respect to using only year fixed effects, the adjusted r-squared slightly increased.
19
Table 8: Regression Coefficient Estimates
This table report estimates of panel least squares regressions controlling for industry effects and year fixed effects. *, **, *** indicate significance at 10%, 5% and 1% confidence level. Book leverage is defined as the value of debt over the value of debt plus the book value of equity, for market leverage the market value of equity is used. Tangibility is property, plant and equipment divided by the total assets. Profitability is measured as EBITDA over total assets. Growth is proxied by market to book ratio. Size is the natural logarithm of sales. Non-debt tax shield is depreciation divided by total assets. Volatility is the standard deviation of the change in operating profits with respect to the previous 5 years. All variables are
winsorized at 1th and 99th percentile. Panel C present the difference between the coefficients using the
Wald’s restriction test.
Panel A: 2003-2007
STBLEV LTBLEV STMLEV LTMLEV Constant -0.083*** -0.207*** 0.008 -0.057* Tangibility -0.030** 0.188*** -0.030*** 0.332*** Profitability -0.231*** -0.802*** -0.158*** -1.115*** Growth 0.015*** 0.020*** -0.001 -0.015*** Size 0.013*** 0.029*** 0.006*** 0.039*** Tax Shield -0.229* 0.334 -0.190*** -0.384 Volatility -0.001 0.001 0.000 0.003 Panel B: 2008-2012
STBLEV LTBLEV STMLEV LTMLEV Constant -0.083*** -0.207*** 0.008 -0.057* Tangibility -0.003 0.218*** 0.002 0.364*** Profitability -0.247*** -0.546*** -0.256*** -0.760*** Growth 0.009*** 0.023*** -0.003*** -0.011*** Size 0.009*** 0.037*** 0.005*** 0.055*** Tax Shield -0.346*** 0.091 -0.448*** -0.789*** Volatility -0.001 0.003*** 0.000 0.003***
Panel C: Wald's Restriction test
STBLEV LTBLEV STMLEV LTMLEV Tangibility 3.50* 1.25 8.10*** 0.99 Profitability 0.19 13.40*** 11.28*** 19.55*** Growth 19.00*** 1.33 1.54 1.92 Size 4.40** 5.70** 0.79 15.01*** Tax Shield 0.50 0.43 3.87** 1.25 Volatility 0.81 3.45* 1.18 4.39** Number of Observations 6258 6308 6473 6522 Adjusted R² 0.11 0.21 0.12 0.30 F-Stat 18.72*** 41.74*** 22.10*** 68.82***
20 Additionally, volatility is only significant in relation with long-term debt during the financial crisis which is consistent with the previous results. Table 8, strengthens the previous findings of
significant drop in the coefficient of profitability. Moreover, in terms of short term leverage the determinant size loses impact in the crisis period but increases its impact in terms of long term leverage. Furthermore, Titman and Wessels (1988) show that small firms tend to use significantly more short-term financing than large firms. This difference mirrors the high transaction costs that small firms face when they want to issue long-term debt or equity. Also, Chan et al. (1985) argue that smaller firms borrow more short term debt; as a consequence they state that smaller firms are particularly sensitive to temporary economic downturns. However, Guedes and Opler (1996) argue smaller and riskier firms rarely issue short-term debt and never issue debt with a term-to-maturity of more than 29 years. In addition, they find that firms with greater growth options tend to issue debt that is shorter in maturity than other firms.
Table 9: Short-term Leverage with Size and Growth
This table presents the relation of short-term debt with size and growth. Comparing the average
observations of the 1-20th percentile and 80-99th percentile. To test the statistical difference a two
sample t-test is conducted.
Size (percentile) Growth (percentile) 1-20th 80th-99th 1-20th 80th-99th Short-term Book Leverage 0,048 0,094 0,068 0,087 Short-term Market Leverage 0,040 0,061 0,098 0,027
Statistical difference (t-test)
Short-term Book Leverage -11,94*** -4,14*** Short-term Market Leverage -6,53*** 17,21***
21
5.2 Hypothesis
In this paragraph the hypothesis are examined which are stated in Table 1 and explained in the literature review. The empirical evidence will be clarified by the underlying theories and compared to existing studies. The findings in table 6-8 will be used as a guidance to test the hypothesis.
5.2.1 Tangibility and leverage.
In previous studies tangibility have proven to be one of the main drivers behind a firm’s debt level. The financial crisis caused the average amount of debt in a firm to increase. On the other hand the average amount of the tangible assets remained on the same level, see Table 9. Table 6 provides evidence that the impact of tangible assets increased significantly with both market and book leverage during the financial crisis. Though, controlling for industry effects doesn’t indicate an significant increase in the coefficients of tangibility. Wu and Yeung (2012) explain that some industries always have more tangible than intangible assets. In Table 8, it appears that tangibility lost its influence on the amount of short-term debt during the financial crisis. In relation with long-term debt it shows similar results as table 7. So there is insufficient evidence that the financial crisis caused an increase in the amount of debt gained per tangible assets. In relation with short-term debt there is evidence that tangibility doesn’t have an impact on debt level anymore during the current financial crisis.
5.2.2 Profitability and leverage.
22
5.2.3 Growth prospects and leverage.
The financial crisis is responsible for a decline in the economy and therefore it causes reduced growth opportunities. The expectation is an increased impact of expected growth as a determinant of the capital structure because it should be harder for firms with high growth prospects to adopt debt. Table 6 and 7 show a significant relation between growth and leverage. Though, it shows a small negative relation with market leverage and a small positive relation with book leverage. In addition, there is no significant change caused by the financial crisis. Moreover, Table 8 displays a decreasing impact of growth opportunities in relation with short-term book leverage. Thus, growth prospects show contradicting evidence about the relation with leverage and doesn’t seem to change during the financial crisis.
5.2.4 Size and leverage.
In previous studies the role of size has been questionable due conflicting theories. González and González (2012) state that assuming no information asymmetry the pecking order theory does not survive and therefore the trade-off theory applies. The trade-off theory states that it should be harder for smaller firms to issue debt than for bigger firms. This effect should increase in times of economic downturns due higher costs of financial distress for smaller firms. Table 6 and 7 show a significant positive relation with book and market leverage. Controlling for year effects there is a small but significant positive change in relation with book leverage. Additionally, controlling for industry effects market leverage shows a significant positive change in the coefficient size. Furthermore, table 8, strengthens these findings with significant increased coefficients for long-term book and market leverage. The relation with short-long-term book leverage shows also a significant positive change. Interpreting these results it’s safe to say that it should be harder for smaller firms to issue debt than for bigger firms during the current financial crisis. However, the impact of the determinant size stays relatively small.
5.2.5 Non-debt tax shield and leverage.
23 Ozkan (2001) note that deprecation may also proxy other things than non-debt tax shield. Overall, provides this research no significant changes in the relationship of non-debt tax shields with leverage.
5.2.6 Volatility and leverage.
According to the trade-off theory, volatility should be negatively related with a firm’s capital structure. High volatility increases the costs of financial distress; firms with higher volatility should have lower leverage especially during a financial crisis. However, the empirical evidence for the period preceding the financial crisis doesn’t show any significant relationship with leverage. Although it does show a significant relationship for the period during the financial crisis, the values of the coefficients are close to zero. Titman and Wessels (1988) also did not provide support for an effect on debt ratios arising from volatility. The conclusion can be made that the financial crisis doesn’t show an increased negative relationship with volatility.
Overall, did the financial crisis have impact on a firm’s capital structure determinants. The determinants profitability and size seem to being influenced by the outbreak of a financial crisis. Less evidence is found for changes in tangibility, non-debt tax shields and growth opportunities. Volatility doesn’t seem to influence the level of debt in a firm although it shows a significant relationship with leverage for the period of the current financial crisis.
6. Conclusion and Further Research
The purpose of this research is to examine changes in the determinants of a firm’s capital structure caused by the current financial crisis. There are several previous studies on determinants of capital structure, but less focus on the consequences of the current financial crisis on the capital structure of non-financial firms. Nonetheless, Murray and Vidhan (2003) study the changes in determinants over the period 1960-2000, which included the early 1980s recession.
24 The evidence on growth prospects show contrary results; in relation with book leverage it indicates a positive relation and in terms of market leverage is indicates a negative relation. There is no change between the financial crisis and the period preceding the financial crisis. For non-debt tax shields I find that it is negatively related with leverage but there is no change between the two periods. Furthermore, I provide evidence that volatility does not have influence on a firm’s debt level. Lastly, I find evidence that smaller firms don’t use more short-term debt relatively than bigger firms.
The exit of firms is neglected, this exclusion could affect the representativeness of the final sample. Even so, the period of the sample is relatively small and therefore I do not expect this survivor bias has a major impact. Moreover, it is possible that some variable may also proxy other things than exclusively the impact on debt . This research only used data of firm listed on the NYSE, therefore only firms located in the United States were used. Further research could include different countries to check if the financial crisis has the same effect on firms distant of the centre of the crisis. Additionally, it is interesting to check for difference between bank dependent countries and market oriented countries.
7. References
Baker, M., Wurgler, J., 2002. Market timing and capital structure. Journal of Finance 57, 1-32. Barclay, M., Smith, C., 1995. The maturity structure of corporate debt. Journal of Finance 50, 609-631.
Bayrakdaroglu, A., Ege, I., Yazici, N., 2013. A panel data analysis of capital structure determinants: empirical results from Turkish capital market. International Journal of Economics & Finance 5, 131-140.
Brooks, C. 2008. Introductory Econometrics for Finance, Cambridge University, Cambridge Chan, K., Chen, N., Hsieh, D., 1985. An Exploratory Investigation of the Firm Size Effect. Journal of Financial Economics 14, 451-471.
25 Charalambakis, E., Psychoyios, D., 2012. What do we know about capital structure? revisiting the impact of debt ratios on some firm-specific factors. Applied Financial Economics 22, 1727-1742. Chung, K., 1993. Asset characteristic and corporate debt policy: an empirical test. Journal of Business Finance & Accounting 20, 83-98.
Covitz, D., Liang, N., Suarez, G,. 2013. The evolution of a financial crisis: collapse of the asset-backed commercial paper market. Journal of Finance 68, 815-848.
DeAngelo, H., Masulis, R., 1980. Optimal capital structure under corporate and personal taxation. Journal of Financial Economics 8, 3-29.
Fama, E., Jensen, M., 1983. Agency problems and residual claims. Journal of Law & Economics 26, 327-350.
Murray, F., Goyal, K., 2009. Capital structure decisions: which factors are reliably important? Financial Management 38, 1-37.
Fosberg, R., 2012. Capital structure and the financial crisis. Journal of Finance & Accountancy 11, 46-55.
Guedes, J., Opler. T., 1996. The determinants of the maturity of corporate debt issues. Journal of Finance 51, 1809-1833.
Gonzalez, V. Gonzalez, F., 2012. Firm size and capital structure: evidence using dynamic panel data. Applied Economics 44, 4745-4754.
Hall, T., 2012. The collateral channel: evidence on leverage and asset tangibility. Journal of Corporate Finance 18, 570-583.
Jensen, M., Meckling, W., 1976. Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3, 305-360.
Kantor, B., Holdsworth, C., 2010. Lessons from the global financial crisis. Journal of Applied Corporate Finance 22, 112-122.
26 Kraus, A., Litzenberger, R., 1973. A state-preference model of optimal financial leverage. Journal of Finance 28, 911-922.
Maroney, N., Naka, A., Wansi, T., 2004. Changing risk, return, and leverage: the 1997 Asian financial crisis. Journal of Financial & Quantitative Analysis 39, 143-166.
de Miguel, A., Pindado, J., Torre, L., 2004. Ownership structure and firm value: new evidence from Spain. Strategic Management Journal 25, 1199-1207.
Modigliani, F., Miller, M., 1958. The cost of capital, corporation finance and the theory of investment. American Economic Review 48, 261.
Myers, S., 2001. Capital structure. Journal of Economic Perspectives 15, 81-102.
Myers, S., Majluf, N., 1984. Corporate financing and investment decisions when firms have information that investors do not have. Journal of Financial Economics 13, 187-221. New York Stock Exchange 2013, Listings Directory, accessed 13 December 2013, <http:// http://www.nyse.com >
Norden, L., van Kampen, S., 2013. Corporate leverage and the collateral channel. Journal of Banking & Finance 37, 5062-5072.
Ozkan, A., 2001. Determinants of capital structure and adjustment to long run target: evidence from UK company panel data. Journal of Business Finance & Accounting 28, 175-198.
Rajan, R., Zingales, L., 1995. What do we know about Capital Structure? some evidence from international data. Journal of Finance 50, 1421-1460.
Titman, S., 1984. The effect of capital structure on a firm's liquidation decision. Journal of Financial Economics 13, 137-151.
Titman, S., Wessels, R., 1988. The determinants of capital structure choice. Journal of Finance 43, 1-19.
