Determinants of Capital Structure - Essay Example

Determinants of capital structure In order to be able to do the regression analysis, key financial indicators had to be determined. The financial indicators, which have been used for the purpose of conducting the analysis, contributes significantly towards ensuring the financial stability of a company and helps analysts and investors to value a company appropriately. The following table includes the descriptive statistics of different financial parameters. The database that has been used for this empirical analysis and calculation has been retrieved from Data Stream. The database includes statement of financial position, cash flow statement as well as statement of income of many existing and extinct companies based all over the world. Initially, a great deal of effort was made to define the independent and dependent variable required for the purpose of regression analysis. The regression analysis that has been done in this particular study is based upon gearing measures. Thus, in order to conduct this analysis, alternative definitions of gearing have been explained in the following paragraphs. Non-equity liabilities to total assets: The book value of this gearing ratio is the ratio between the total debt plus trade credit and equivalent to total assets (equation 1). The market value of this gearing measure can be calculated by adjusting the value of the total assets, deducting the book value of equity and adding the market value of equity (equation 2). The equation can be represented as follows: (1) (2) In the above equations, TD refers to total debt; TTCE is total trade credit and equivalent; TA is total assets; ECR is book value of equity capital and reserves and MV is the market value of equity. According to Rajan and Zingales (1995), the gearing measure serves as a proxy for the liquidation value of a company. The authors also argued that the value of this indicator may be significantly inflated, as it may only represent financial transactions, instead of assets. Debt to Total Assets: This gearing measure is the ratio between the total debts to total assets (equation 3). The market value of this multiple is determined by adjusting the asset, by deducting the book value of equity and adding the market value of equity (equation 4) (Phillips, Libby and Libby, 2011; Fridson and Alvarez, 2011). The equation can be represented as follows: (3) (4) Debt to Capital: This gearing measure is the ratio between the total debts to capital. The capital in the denominator represents the total debt plus the equity, which includes the preference shares as well (equation 5) (Rose and Hudgins, 2008). The market value of this gearing measure is calculated by adjusting market value of equity, instead of adjusting the book value of equity (equation 6). The equation can be represented as follows: (5) (6) In the equation given above, PS is the book value of preference shares. The three gearing measures given above will serve as the dependent variables, as suggested by Bevan and Danbolt (2002). The gearing measures, which have been included in this analysis, produced well-specified distributions and henceforth, required minimal outlier elimination. As far as the explanatory variables are concerned, choosing them is controversial (Harris and Raviv, 1991). Following the steps mentioned by Rajan and Zingales (1995), we have adopted the following four independent variables. Market-to-book (MTB) ratio: The market to book value ratio is, “the ratio of the book value of total assets minus the book value of equity plus the market value of equity, to the book value of total assets” (Bevan and Danbolt, 2002, p. 13). The equation can be represented as follows: (7) Log sales: This value is calculated as the natural logarithm of sales (equation 8). The equation can be represented as follows: (8) Profitability: Profitability is calculated as the ratio of EBITDA (earnings before interest, taxes, depreciation and amortization) to the book value of total assets (equation 9) (Peterson and Faboozi, 2012). The equation can be represented as follows: (9) Tangibility: Tangibility is the ratio of book value of the fixed assets that are depreciated to the book value of total assets (equation 10) (Eisen, 2007). The equation can be represented as follows: (10) Table 1: Descriptive Statistics Year BV of Non-Equity Liabilities MV of non-equity liabilities total debt to total assets MV of total debt to total assets Debt to capital MV of debt to capital mean mean mean mean mean mean median median median median median median 2000 0.33 0.21 0.22 0.13 0.40 0.18 0.32 0.19 0.18 0.10 0.29 0.13 2001 0.33 0.22 0.22 0.15 0.20 0.20 0.33 0.20 0.20 0.11 0.32 0.15 2002 0.33 0.23 0.23 0.15 0.46 0.21 0.32 0.21 0.19 0.12 0.32 0.18 2003 0.33 0.21 0.24 0.14 0.36 0.19 0.32 0.19 0.20 0.12 0.32 0.16 2004 0.33 0.18 0.24 0.12 0.38 0.15 0.30 0.17 0.17 0.08 0.29 0.10 2005 0.32 0.17 0.21 0.11 0.37 0.15 0.29 0.14 0.18 0.08 0.32 0.10 2006 0.32 0.17 0.22 0.11 0.38 0.15 0.31 0.16 0.19 0.09 0.33 0.12 2007 0.33 0.19 0.23 0.13 0.36 0.17 0.32 0.17 0.20 0.11 0.33 0.14 2008 0.35 0.27 0.25 0.19 0.36 0.28 0.34 0.25 0.23 0.16 0.36 0.24 2009 0.33 0.24 0.23 0.17 0.36 0.25 0.30 0.22 0.22 0.14 0.34 0.21 2010 0.30 0.21 0.20 0.14 0.31 0.19 0.28 0.19 0.17 0.11 0.28 0.15 2011 0.64 0.21 0.27 0.13 0.42 0.19 0.27 0.18 0.16 0.10 0.25 0.12 In order to eliminate any potential reverse causality from this analysis, that might be present between the dependent and independent variables, the former ones have been lagged. The independent variables have been averaged over a period of one year and in order to complement the independent variable in the regression analysis, the dependent gearing measure variable has also been averaged over a one year period. The regression analysis has been formed for both averaged and non-averaged data and the results have been presented in the following tables. Interpretation of results will only be done for the average values. Although the assembled data used in the regression analysis is considerably clean, the biggest outliers were eliminated by setting the confidence level of independent and independent variables at the 5% level. The equation that has been implemented in the regression analysis is as follows: Gearingi,t = ß1 + ß2Market-to-Booki,t +ß3Logsalei,t+ß4Profitabilityi,t + ß5Tangibilityi,t + εi,t In the above equation, the right hand side represents the independent variables, whereas the left hand side represents the dependent variable. The term gearing refers to the individual gearing measures, ‘i’ represents the individual companies and t refers to the time period of the independent and dependent variables. εi,t is the residual return which is assumed to have a mean value of zero and standard deviation σi (Bevan and Danbolt, 2004). Result Interpretation Table 2 and 3 presented below depicts the result of our pooled ordinary least squared regression analysis. The dependent variable, in this case, is the gearing ratio. As is evident from table 2 and 3, the results are insignificant on an average, therefore, suggesting that the null hypothesis is true. The market to book ratio, log sales, profitability and tangibility do not seem to have much effect on the gearing measure of the company. Table 2: Regression Analysis (Gearing Ratio) SUMMARY OUTPUT Regression Statistics Multiple R 0.034041705 R Square 0.001158838 Adjusted R Square 3.90606E-05 Standard Error 17.30214245 Observations 3573 ANOVA   df SS MS F Significance F Regression 4 1239.226751 309.8066878 1.03488245 0.387613783 Residual 3568 1068131.228 299.3641334 Total 3572 1069370.455   Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept -2.835707714 1.604791243 -1.767025915 0.077309341 -5.982108008 0.31069258 -5.982108008 0.31069258 X Variable 1 0.170471302 0.158188292 1.07764803 0.281263712 -0.139677256 0.48061986 -0.139677256 0.48061986 X Variable 2 0.197785083 0.117378446 1.685020456 0.092072075 -0.032350506 0.427920671 -0.032350506 0.427920671 X Variable 3 -1.50933466 1.747984821 -0.863471262 0.387936452 -4.936484439 1.91781512 -4.936484439 1.91781512 X Variable 4 0.926343849 1.192277009 0.776953545 0.437237684 -1.411269065 3.263956763 -1.411269065 3.263956763 Table 3: Regression Analysis using one year average (Gearing Ratio) SUMMARY OUTPUT Regression Statistics Multiple R 0.868308983 R Square 0.75396049 Adjusted R Square 0.613366484 Standard Error 0.393440497 Observations 12 ANOVA   df SS MS F Significance F Regression 4 3.320472544 0.830118136 5.362678763 0.026883298 Residual 7 1.083567972 0.154795425 Total 11 4.404040516         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 11.44160867 15.76122478 0.725933982 0.491431809 -25.82776568 48.71098301 -25.82776568 48.71098301 X Variable 1 1.914555488 0.491901805 3.892149752 0.005958853 0.751392551 3.077718425 0.751392551 3.077718425 X Variable 2 -0.374834836 0.943863815 -0.397128092 0.703103149 -2.606718102 1.85704843 -2.606718102 1.85704843 X Variable 3 -23.12422155 11.04157174 -2.094287127 0.074496963 -49.23338984 2.98494675 -49.23338984 2.98494675 X Variable 4 -22.86951538 10.78563652 -2.120367706 0.071682566 -48.37349306 2.634462308 -48.37349306 2.634462308 Result Interpretation The ordinary least squared regression analysis results depicted in table 4 and 5 consider non-equity liabilities as the dependent variable. As is evident from the values given in the following tables, it can be seen that the intercept (α) having a p value of 7.61 is statistically significant at 1% level of significance and thus, the null hypothesis can be rejected in this case. The p-value of β2, β 3, β4, and β5 are 2.07, 0.008, 0.29 and 0.004 respectively. The values indicate that β2, β 3 and β5 are statistically significant at 1% level of significance, whereas β4 is statistically insignificant. The gearing measure is positively related with the market to book ratio, but is negatively correlated with log sales, profitability and tangibility. This suggests that firms with higher growth prospects tend to hold more debt. However, this conclusion is not in alignment with the explanations provided by Michaelas, Chittenden and Poutziouris (1999), Chittenden, Hall and Hutchinson (1996) and Bevan and Danbolt (2002). Table 4: Regression Analysis (Non equity liabilities) SUMMARY OUTPUT Regression Statistics Multiple R 0.111495534 R Square 0.012431254 Adjusted R Square 0.011324114 Standard Error 0.392750188 Observations 3573 ANOVA   df SS MS F Significance F Regression 4 6.927958109 1.731989527 11.22826 4.71187E-09 Residual 3568 550.3736705 0.15425271 Total 3572 557.3016286         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 0.250215176 0.036427978 6.868763806 7.61E-12 0.178793425 0.321636927 0.178793425 0.321636927 X Variable 1 -0.015306782 0.003590797 -4.262781206 2.07E-05 -0.022347003 -0.008266561 -0.022347003 -0.008266561 X Variable 2 0.007063084 0.002664433 2.650876332 0.008064 0.001839118 0.012287049 0.001839118 0.012287049 X Variable 3 -0.041245412 0.039678402 -1.039492754 0.298646 -0.119040039 0.036549216 -0.119040039 0.036549216 X Variable 4 0.077820663 0.027064106 2.87541967 0.004059 0.024757991 0.130883335 0.024757991 0.130883335 Table 5: Regression Analysis using one year average (Non equity liabilities) SUMMARY OUTPUT Regression Statistics Multiple R 0.628805286 R Square 0.395396088 Adjusted R Square 0.049908138 Standard Error 0.087532357 Observations 12 ANOVA   df SS MS F Significance F Regression 4 0.035074921 0.00876873 1.144456958 0.409669728 Residual 7 0.053633395 0.007661914 Total 11 0.088708316         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 6.113030654 3.506545884 1.743319739 0.124803241 -2.178632779 14.40469409 -2.178632779 14.40469409 X Variable 1 0.076903276 0.109437958 0.702711181 0.504926575 -0.181876372 0.335682925 -0.181876372 0.335682925 X Variable 2 -0.341329565 0.209990139 -1.625455212 0.148092841 -0.837877339 0.155218209 -0.837877339 0.155218209 X Variable 3 -1.633951675 2.45652089 -0.665148699 0.527254885 -7.442700544 4.174797194 -7.442700544 4.174797194 X Variable 4 -4.183864648 2.399580609 -1.743581621 0.124755585 -9.857971148 1.490241851 -9.857971148 1.490241851 Result Interpretation The following tables depict the ordinary least squared regression analysis with total debt to total assets as the dependent variable. Going by the values given in table 6 and 7, it can be said that the p value of the intercept (0.05) is statistically significant at 5% level of significance and therefore, we can reject the null hypothesis. The value of β2, β3, β4, and β5 are 0.76, 0.087, 0.43 and 0.10, respectively. The values suggest that β3 and β5 are statistically significant at 5% and 10% level of significance and thus, the null hypothesis can be rejected for those values and it holds true for the other values. Table 6: Regression Analysis (total debt to total assets) Regression Statistics Multiple R 0.237498135 R Square 0.056405364 Adjusted R Square 0.055347523 Standard Error 0.23679432 Observations 3573 ANOVA   df SS MS F Significance F Regression 4 11.95920617 2.989801543 53.32119 1.05749E-43 Residual 3568 200.0632909 0.05607155 Total 3572 212.0224971         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 0.142415584 0.021962913 6.484366757 1.01E-10 0.099354459 0.185476709 0.099354459 0.185476709 X Variable 1 -0.011181123 0.002164939 -5.164635442 2.54E-07 -0.015425765 -0.00693648 -0.015425765 -0.00693648 X Variable 2 0.005294882 0.001606422 3.296070953 0.00099 0.002145284 0.008444481 0.002145284 0.008444481 X Variable 3 -0.129070391 0.023922637 -5.395324484 7.29E-08 -0.175973808 -0.082166974 -0.175973808 -0.082166974 X Variable 4 0.177797871 0.01631731 10.89627333 3.21E-27 0.145805679 0.209790063 0.145805679 0.209790063 Table 7: Regression Analysis using one year average (total debt to total assets) SUMMARY OUTPUT Regression Statistics Multiple R 0.641818144 R Square 0.41193053 Adjusted R Square 0.075890833 Standard Error 0.018378188 Observations 12 ANOVA   df SS MS F Significance F Regression 4 0.001656146 0.000414037 1.225838893 0.380804991 Residual 7 0.002364304 0.000337758 Total 11 0.004020451         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 1.715619793 0.736230121 2.330276559 0.05259248 -0.025287805 3.456527391 -0.025287805 3.456527391 X Variable 1 -0.007184525 0.022977461 -0.312677069 0.763635339 -0.061517586 0.047148536 -0.061517586 0.047148536 X Variable 2 -0.08752952 0.044089275 -1.985279195 0.087498163 -0.191784088 0.016725048 -0.191784088 0.016725048 X Variable 3 -0.431133623 0.515768147 -0.835905873 0.430812918 -1.65073149 0.788464244 -1.65073149 0.788464244 X Variable 4 -0.930319521 0.503813034 -1.846557071 0.107307818 -2.12164804 0.261008998 -2.12164804 0.261008998 Result Interpretation Table 8 and 9 given below depicts the ordinary least squared regression analysis, considering the debt to capital as the dependent variable. The year on year analysis as well as the one year average value regression analysis has been done. As is evident from the values in the following tables, it can be seen that debt to capital is significantly positively correlated with market to book ratio with a p value of 0.007, which is statistically significant at 5% level of statistical significance. The debt to capital is negatively correlated with the value of log sales, profitability and tangibility. However, the p values of the variables are statistically significant at 1% level of significance. Thus, in both the cases, the null hypothesis can be rejected. Table 8: Regression Analysis (Debt to Capital) SUMMARY OUTPUT Regression Statistics Multiple R 0.06998613 R Square 0.004898058 Adjusted R Square 0.003782473 Standard Error 1.044225338 Observations 3573 ANOVA   df SS MS F Significance F Regression 4 19.15004016 4.787510039 4.390573415 0.001528365 Residual 3568 3890.570594 1.090406557 Total 3572 3909.720634         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 0.06936278 0.096852958 0.716165842 0.47393582 -0.120529941 0.259255501 -0.120529941 0.259255501 X Variable 1 -0.012930249 0.009547039 -1.354372735 0.175703253 -0.03164845 0.005787952 -0.03164845 0.005787952 X Variable 2 0.022131252 0.007084068 3.124088156 0.001797879 0.008242023 0.03602048 0.008242023 0.03602048 X Variable 3 -0.068016838 0.10549503 -0.644739733 0.519137313 -0.274853456 0.138819781 -0.274853456 0.138819781 X Variable 4 0.11285224 0.071956746 1.568334404 0.116891785 -0.028228244 0.253932724 -0.028228244 0.253932724 Table 9: Regression Analysis using one year average (Debt to Capital) SUMMARY OUTPUT Regression Statistics Multiple R 0.817217998 R Square 0.667845256 Adjusted R Square 0.478042546 Standard Error 0.045574195 Observations 12 ANOVA   df SS MS F Significance F Regression 4 0.029232868 0.007308217 3.518628652 0.070486747 Residual 7 0.014539051 0.002077007 Total 11 0.043771919         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 6.713866194 1.825702061 3.677416122 0.007886554 2.396766826 11.03096556 2.396766826 11.03096556 X Variable 1 0.114062995 0.056979464 2.001826403 0.085389012 -0.020672027 0.248798018 -0.020672027 0.248798018 X Variable 2 -0.361151987 0.1093325 -3.303244559 0.013062297 -0.619682268 -0.102621705 -0.619682268 -0.102621705 X Variable 3 -4.011728824 1.279000874 -3.136611479 0.016456916 -7.036085308 -0.987372339 -7.036085308 -0.987372339 X Variable 4 -4.487081437 1.249354609 -3.591519496 0.008839201 -7.441335645 -1.53282723 -7.441335645 -1.53282723 Table 10: Descriptive Statistics Year Trade creditors Borrowings repayable < 1 year Total liabilities Total current liabilities Total debt mean mean mean mean mean median median median median median 2000 250747.11 232923.60 1645326.07 986361.02 658965.05 36326.00 11771.00 181547.00 112735.00 58240.00 2001 249358.15 269046.69 1787579.76 1006663.98 780915.77 37909.00 13750.00 205706.50 128976.50 67990.00 2002 256120.97 226998.19 1749058.50 984713.39 764459.25 32173.50 12616.00 199344.00 119437.00 64049.50 2003 257466.01 200142.34 1700769.11 944730.97 756042.35 30597.00 10370.50 210052.50 119050.00 60165.50 2004 280707.89 188831.45 1806500.07 1006910.72 799594.81 32648.00 12200.00 205349.00 131451.00 57700.00 2005 341892.58 227964.43 2169654.57 1264083.41 905571.16 34900.00 11100.00 246200.00 143900.00 64406.00 2006 356615.60 240664.90 2234479.85 1216498.19 1017981.65 39100.00 10717.00 279000.00 143500.00 84900.00 2007 414467.39 349024.50 2821410.80 1432597.56 1388813.23 44325.00 11856.00 311100.00 164153.00 118200.00 2008 513672.98 612731.78 4280331.56 2152719.72 2127611.84 51400.00 13872.50 353323.50 188512.00 150436.00 2009 488355.89 378196.03 3768411.47 1741308.15 2027103.32 49386.00 11482.50 342190.00 188523.00 124050.00 2010 550209.24 293060.87 3773368.32 1779210.44 1994157.88 50700.00 11520.00 345862.00 190000.00 112255.00 2011 142399.18 54414.47 829692.91 444464.57 385228.35 Table 11: Descriptive Statistics Year MTB Ratio Log sales Profitability Tangibility Gearing ratio mean mean mean mean mean median median median median median 2000 2.63 12.71 0.14 0.34 0.47 1.45 12.85 0.14 0.28 0.39 2001 1.80 12.81 0.13 0.34 -0.13 1.36 12.97 0.14 0.28 0.43 2002 1.47 12.76 0.08 0.33 0.14 1.23 13.00 0.13 0.27 0.36 2003 1.56 12.70 0.11 0.33 -0.70 1.28 13.00 0.13 0.27 0.39 2004 1.61 12.73 0.15 0.31 -1.16 1.40 13.03 0.14 0.26 0.33 2005 1.92 12.86 0.15 0.30 0.78 1.61 13.13 0.14 0.23 0.35 2006 2.14 13.03 0.15 0.28 0.70 1.76 13.21 0.15 0.22 0.43 2007 2.01 13.14 0.15 0.28 0.64 1.67 13.24 0.14 0.23 0.45 2008 1.39 13.27 0.13 0.27 -0.17 1.22 13.33 0.12 0.22 0.49 2009 1.48 13.33 0.11 0.29 0.07 1.26 13.33 0.11 0.23 0.47 2010 1.75 13.40 0.13 0.27 0.44 1.41 13.43 0.13 0.22 0.37 2011 2.12 12.94 0.16 0.27 0.95 1.30 13.20 0.15 0.19 0.23 Market to book value ratio As is evident from the regression analysis results presented above, we observe a significantly positive correlation between the gearing measures adopted for regression analysis and the market to book ratio, which contradicts the literatures provided by Chittenden, Hall and Hutchinson (1996), Barclay, Smith and Watts (1995) as well as by Bevan and Danbolt (2002). The values are by and large statistically significant, suggesting that companies in order to be efficient, in terms of productivity, tend to hold more debt. Even so, for some of the gearing measures, the value tends to be negatively correlated, but the values are not that significant. Size Having done the ordinary least squared regression analysis, we find the size of the company to be negatively correlated with all the gearing measures adopted for this analysis, which contradicts the results explained by Bevan and Danbolt (2002). Although the values are negatively correlated, yet they are statistically significantly by and large, which indicates that lenders tend to refrain from issuing loans to smaller company by limiting the loan maturity in order to minimize the risk of lending of companies of such size. Profitability As far as the value of profitability is concerned, it is evident from the regression analysis that it is negatively correlated with all forms of gearing multiples adopted in this research analysis. However, it should be noted that the values are not significant and thus, the null hypothesis cannot be rejected. This result also contradicts the results explained by Bevan and Danbolt (2002). Tangibility Very similar to the profitability, the statistical values of tangibility are by and large negatively correlated with the gearing measures. The values are also statistically significant and thus, in this case, we can reject the null hypothesis. This observation is in partial alignment with the one provided by Bevan and Danbolt (2002). Reference List Barclay, M. J., Smith C. W. and Watts, R. L., 1995. The determinants of corporate leverage and dividend policies. Journal of Applied Corporate Finance, 7(4), pp. 4-19. Bevan, A. A. and Danbolt, J., 2002. Capital Structure and its determinants in the United Kingdom: A decompostional analysis. Applied Financial Economics, 12(3), pp. 159-170. Bevan, A. A. and Danbolt, J., 2004. Testing for inconsistencies in the estimation of UK capital structure determinants. Applied Financial Economics, 14(1), pp. 55-66. Chittenden, F., Hall, G. and Hutchinson, P., 1996. Small firm growth, access to capital markets and financial structure: Review of issues and an empirical investigation. Small Business Economics, 8, pp. 59-67. Eisen, P. J., 2007. Accounting. 5th edn. New York: Barrons Educational Series. Fridson, M. and Alvarez, F., 2011. Financial statement analysis: A practitioners guide. New Jersey: John Wiley & Sons. Harris, M. and Raviv, A., 1991. The theory of capital structure. Journal of Finance, 46(1), pp. 297-355. Michaelas, N., Chittenden, F. and Poutziouris, P., 1999. Financial policy and capital structure choice in U.K. SMEs: Empirical evidence from company panel data. Small Business Economics, 12, pp. 113-130. Peterson, P. P. and Faboozi, F. J., 2012. Analysis of financial statements. 3rd edn. New Jersey: John Wiley & Sons. Phillips, F., Libby, R. and Libby, P. A., 2011. Fundamentals of Financial Accounting. 3rd edn. New York: McGraw Hill. Rajan, R. G. and Zingales, L., 1995. What do we know about capital structure? Some evidence from international data. Journal of Finance, 50(5), pp. 1421-1460. Rose, P. S. and Hudgins, S. C., 2008. Bank management and financial services. 7th edn. New York: McGraw Hill. Read More