Macroeconomic Conditions and Capital Structure: Evidence ... · PDF fileMacroeconomic Conditions and Capital Structure: ... model with the financial constraint of over-leverage and
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Macroeconomic Conditions and Capital Structure: Evidence
dDR = debt ratio adjustment; DRt = the debt ratio at time t; DRt-1 = the previous debt ratio at time t; EC = 0 for economic trough and 1 for economic peak; BC7 = 0 for Business Cycles 6 and 7, and 1 for Business Cycle 7; BC8 = 0 for Business Cycles 6 and 7, and 1 for Business Cycle 8; IND13 = 0 for the textile and electronics industries and 1 for the plastics industry; IND14 = 0 for the plastics and electronics industries and 1 for the textile industry; lnS = natural logarithm of net sales; gTA = annual growth rate of total assets; OITA = net operating income/total assets; DEPTA = depreciation/total assets; INVFATA = inventory plus net fixed assets/total assets.
4.2 Regression Results
Based on Equations 6 and 7, the regression results for the determination of the debt ratio adjustment
and the actual debt ratio in the subsamples with a negative and a positive adjustment gap are shown
in Tables 2 and 3, respectively. As shown in Note 4 of Table 2, the explanatory power, i.e. adjusted R-
square, of the models for the determination of the debt ratio adjustment and the actual debt ratio in
the subsample with a negative adjustment gap is 43.41% and 79.61%, respectively. In addition, no
serious residual auto-correlation problems are found according to the Durbin-Watson D value that is
close to 2, i.e. 1.957 as also shown in Note 4 of the table. Further, as can be seen in the VIF column of
Table 2, the values of variance inflation factor (VIF) much less than the critical value of 10 are often
regarded as indicating no problematic multi-collinearity (Chatterjee & Price 1991). Furthermore, no
sample observations with values of DFFITS are greater than 1.02 that indicates no outlier effect in the
subsample with a negative adjustment gap (Belsey et al. 1980).
Table 2. Regression Results in the Subsample with a Negative Adjustment Gap
Variables
(1) Dependent variable: Debt Ratio Adjustment (dDR)
Notes: 1. dDRt = total debt ratio adjustment in the current year; DRt = total debt ratio at the end of the current year; DRt-1 = total debt ratio at the end of the previous year; EC = 0 for economic trough and 1 for economic peak; BC7 = 0 for Business Cycles 6 and 8 and 1 for Business Cycle 7; BC8 = 0 for Business Cycles 6 and 7 and 1 for Business Cycle 8;
13
IND13 = the dummy variable with a value of 1 for the plastics industry; IND14 = the dummy variable with a value of 1 for the textile industry; lnS = natural logarithm of net sales; gTA = annual growth rate of total assets; OITA = net operating income/total assets; DEPTA = depreciation/total assets; INVFATA = inventory plus net fixed assets/total assets; lnS×EC, gTA×EC, OITA×EC, DEPTA×EC and INVFATA×EC = interactions between firm-specific variables and macroeconomic conditions. 2. a, b and c indicate the significance level of 1%, 5% and 10%, respectively.
3. The value of the coefficient is the product of the rate of adjustment (ρ) and the regression coefficient (β) of each independent variable except the previous actual debt ratio (DRt-1) as shown in Equations 5-5 and 5-5A. 4. N Adj. R-square F value Durbin-Watson D value (1) 365 0.4341 18.56a 1.957 (2) 365 0.7961 90.05a 1.957 5. VIF: Variance Inflation Factor
Table 3. Regression Results in the Subsample with a Positive Adjustment Gap
Variables
(1) Dependent variable: Debt Ratio Adjustment (dDR)
Notes: 1. dDRt = total debt ratio adjustment in the current year; DRt = total debt ratio at the end of the current year; DRt-1 = total debt ratio at the end of the previous year; EC = 0 for economic trough and 1 for economic peak; BC7 = 0 for Business Cycles 6 and 8 and 1 for Business Cycle 7; BC8 = 0 for Business Cycles 6 and 7 and 1 for Business Cycle 8; IND13 = the dummy variable with a value of 1 for the plastics industry; IND14 = the dummy variable with a value of 1 for the textile industry; lnS = natural logarithm of net sales; gTA = annual growth rate of total assets; OITA = net operating income/total assets; DEPTA = depreciation/total assets; INVFATA = inventory plus net fixed assets/total assets; lnS×EC, gTA×EC, OITA×EC, DEPTA×EC and INVFATA×EC = interactions between firm-specific variables and macroeconomic conditions. 2. a, b and c indicate the significance level of 1%, 5% and 10%, respectively.
3. The value of the coefficient is the product of the rate of adjustment (ρ) and the regression coefficient (β) of each independent variable except the previous actual debt ratio (DRt-1) as shown in Equations 5-5 and 5-5A. 4. N Adj. R-square F value Durbin-Watson D value (1) 262 0.7210 43.32a 2.138 (2) 262 0.9161 179.75a 2.138 5. VIF: Variance Inflation Factor
14
On the other hand, the explanatory power, i.e. adjusted R-square, of the models for the
determination of the debt ratio adjustment and the actual debt ratio in the subsample with a positive
adjustment gap is 72.10% and 91.61%, respectively, as shown in Note 4 of Table 3. In addition, no
serious residual auto-correlation problems are found according to the value of Durbin-Watson D that
is close to 2, i.e. 2.138, as also shown in Note 4 of Table 3. Further, no serious multi-collinearity
problem is found according to the values of variance inflation factor (VIF) shown in the VIF column of
Table 3 that are much less than the critical value of 10. Furthermore, no sample observations with
values of DFFITS are greater than 1.02 that indicates no outlier effect in the subsample with a positive
adjustment gap (Belsey et al. 1980). Further analyses on the regression results in the subsamples with
a negative and a positive adjustment gap over Business Cycles 6 to 8 shown in Tables 2 and 3 are now
discussed.
4.2.1 Macroeconomic Conditions and Their Interactions with Firm-Specific Variables
According to the t-value shown in the t-value columns in Table 2 in the subsample with a negative
adjustment gap, the dummy proxy for macroeconomic conditions, i.e. EC, in the subsample with a
negative adjustment gap is not significantly related to the debt ratio adjustment and the actual debt
ratio of the listed firms in the textile, plastics and electronics industries during the period from 1983
to 1995 over Business Cycles 6 to 8 in Taiwan. On the other hand, according to the t-value shown in
Table 3 in the subsample with a positive adjustment gap, the dummy proxy for macroeconomic
conditions is statistically significant and positively related to the debt ratio adjustment and the actual
debt ratio of the listed firms in the textile, plastics and electronics industries over Business Cycles 6 to
8 in Taiwan. The result in the subsample with a positive adjustment gap is consistent with the
conclusion of Stulz (1990) that firms finance with less debt at economic trough in response to future
investment and growth opportunities. In addition, the findings in the subsamples with a negative or a
positive adjustment gap suggest that the effect of macroeconomic conditions on the debt ratio
adjustment and the debt ratios varies according to whether the adjustment gap between the target
debt ratio and the previous actual debt ratio is positive or negative.
Further, according to the t-value shown in the t-value columns in Table 2, the interaction between
growth opportunities and macroeconomic conditions (gTA×EC) is statistically significant and
negatively related to the debt ratio adjustment and the actual debt ratio in the subsample with a
negative adjustment gap. The results show that the effect of growth opportunities on the debt ratio
adjustment and the actual debt ratio is augmented by macroeconomic conditions although
macroeconomic conditions do not have a direct effect on the debt ratio adjustment and the actual
debt ratio in the subsample with a negative adjustment gap. On the other hand, according to the t-
value shown in Table 3, the interaction between growth opportunities and macroeconomic conditions
(gTA×EC) is statistically significant and negatively related to the debt ratio adjustment and the actual
debt ratio in the subsample with a positive adjustment gap. As can be seen in Table 3, growth
opportunities have a positive main effect on debt ratio but the effect is diminished by the negative
15
interaction with macroeconomic conditions. In addition, the interaction between asset tangibility and
macroeconomic conditions is statistically significant and positively related to the debt ratio
adjustment and the actual debt ratio in the subsample with a positive adjustment gap. The findings on
the interaction effect in the subsamples with negative and positive adjustment gaps illustrate the
variation in the effect of interactions between firm-specific variables and macroeconomic conditions
on the debt ratio adjustment and the actual debt ratio.
As a whole, the findings above suggest that the effects of macroeconomic conditions and their
interaction with firm-specific variables vary according to whether firms have the financial constraint
of over-leverage or under-leverage, i.e. a negative or a positive adjustment gap between the target
debt ratio and the previous actual debt ratio.
4.2.2 Adjustment Rate
Looking at the t-value shown in column t-value of Tables 2 and 3, regression coefficient of the
previous actual debt ratio (DRt-1) is significantly different from 0 in the case of a negative and a
positive gap over Business Cycles 6 to 8 for the listed firms in the textile, plastics and electronics
industries of Taiwan. The regression coefficient of the previous actual debt ratio in the empirical
model for the determination of actual debt ratio, i.e. Equation 7, is exactly equal to 1 minus the
adjustment rate of debt ratio adjustment (1−ρ). Therefore, the adjustment rate of debt ratio is
0.14831 (i.e. 1−0.85169) according to the regression coefficients of the previous actual debt ratio for
the determination of the debt ratio adjustment and the actual debt ratio in Table 2. On the other
hand, as shown in Table 3, the adjustment rate of debt ratio is 0.03474 (i.e. 1−0.96526) according to
the regression coefficient of the previous actual debt ratio for the determination of the debt ratio
adjustment and the actual debt ratio in the subsample with a positive adjustment gap. It is worth
noting that the effect of the explanatory variables, except the previous actual debt ratio, on the debt
ratio adjustment is a proportion of the regression coefficients of these variables in Equations 6 and 7.
The proportion depends upon the adjustment rate, as shown in Equations 6 and 7.
The findings above suggest that the adjustment rate of debt ratio for firms with a negative adjustment
gap, i.e. a financial constraint of over-leverage, is faster than for those with a positive adjustment gap,
i.e. a financial constraint of under-leverage. Firms with the financial constraint of over-leverage tend
to gear down their leverage due to the high risk of bankruptcy and adjust faster to rebalance their
debt ratio toward the target level. This suggests that firms adjust at a different rate of debt ratio
adjustment over time when they have a negative or a positive adjustment gap between the target
debt ratio and the previous actual debt ratio. This finding does not support the constant adjustment
rate of Flannery and Rangan (2006) but is consistent with Byoun (2008).
Moreover, in order to contrast the difference when negative and positive adjustment gaps are or are
not taken into account in the estimation of the modified partial adjustment model utilized in the
study, the regression results for the determination of the debt ratio adjustment and the actual debt
16
ratio in the full sample when negative and positive adjustment gaps are not taken into account are
shown in Table 4.
As shown in Note 4 in Table 4, the adjusted R-square for the model without negative and positive
adjustment gaps taken into account is much lower than that for the model used in the study, as
shown in Tables 2 and 3. There is residual auto-correlation problem due to the Durbin-Watson D
value (1.279) close to 1, as shown in Table 4. In addition, based on Equations 6 and 7, the regression
coefficient of the previous actual debt ratio is respectively equal exactly to the negative value of the
adjustment rate (−ρ) and to 1 minus the adjustment rate (1−ρ). The adjustment rate of debt ratio
(0.21824 or 1−0.78176) shown in Table 4 is overestimated without the adjustment gaps taken into
account. It is much higher than the adjustment rates of 0.14831 and 0.03474 shown respectively in
Tables 2 and 3 with negative and positive adjustment gaps taken into account. These additional
regression results reflect the fact that, in the application of the partial adjustment model for capital
structure adjustment, the estimation should take into account whether the adjustment gap between
the target capital structure and the previous actual capital structure is negative or positive.
Table 4. Regression Results in the Full Sample without Adjustment Gap Taken into Account
Variables
(1) Dependent variable: Debt Ratio Adjustment (dDR)
Notes: 1. dDRt = total debt ratio adjustment in the current year; DRt = total debt ratio at the end of the current year; DRt-1 = total debt ratio at the end of the previous year; EC = 0 for economic trough and 1 for economic peak; BC7 = 0 for Business Cycles 6 and 8 and 1 for Business Cycle 7; BC8 = 0 for Business Cycles 6 and 7 and 1 for Business Cycle 8; IND13 = the dummy variable with a value of 1 for the plastics industry; IND14 = the dummy variable with a value of 1 for the textile industry; lnS = natural logarithm of net sales; gTA = annual growth rate of total assets; OITA = net operating income/total assets; DEPTA = depreciation/total assets; INVFATA = inventory plus net fixed assets/total assets; lnS×EC, gTA×EC, OITA×EC, DEPTA×EC and INVFATA×EC = interactions between firm-specific variables and macroeconomic conditions. 2. a, b and c indicate the significance level of 1%, 5% and 10%, respectively.
17
3. The value of the coefficient is the product of the rate of adjustment (ρ) and the regression coefficient (β) of each independent variable. 4. Case N Adj. R-square F value Durbin-Watson D value (1) 627 0.2331 12.91a 1.279 (2) 627 0.7437 114.70a 1.279 5. VIF: Variance Inflation Factor
5. Conclusion
This paper utilizes the modified partial adjustment model to examine the adjustment behavior of
capital structure decisions for the listed firms in the textile, plastics and electronics industries across
years of economic trough and peak over Business Cycles 6 to 8 in Taiwan. With the controls for the
effects of firm characteristics and industry type, the findings show a significant effect of
macroeconomic conditions on the debt ratio adjustment and, in addition, the variation in the effect of
macroeconomic conditions on the debt ratio adjustment. Macroeconomic conditions have a positive
effect on the debt ratio adjustment and the actual debt ratio for firms with the financial constraint of
under-leverage relative to the target debt ratio in the case of a positive adjustment gap. No significant
effect, however, appears on the determination of the debt ratio adjustment and the actual debt ratio
for firms with the financial constraint of over-leverage relative to the target debt ratio in the case of a
negative adjustment gap. In addition, the interactions between macroeconomic conditions and firm-
specific variables also vary according to whether firms have a negative or a positive adjustment gap.
Further, the adjustment rate of debt ratios varies according to the financial constraint of over-
leverage or under-leverage relative to the target debt ratios in the adjustment gap between the target
debt ratio and the previous actual debt ratio for the listed firms in the textile, plastics and electronics
industries over Business Cycles 6 to 8. The adjustment rate is negatively related to the debt ratio
adjustment for the listed firms with the financial constraint of over-leverage in the case of a negative
adjustment gap. This indicates that the greater the adjustment rate, the greater is the decrease in the
debt ratios of firms with the financial constraint of over-leverage relative to the target debt ratios. On
the contrary, the adjustment rate is positively related to the debt ratio adjustment for the listed firms
with the financial constraint of under-leverage relative to the target debt ratio in the case of a positive
adjustment gap. This shows that the greater the adjustment rate, the greater is the increase in the
debt ratios of firms with the financial constraint of under-leverage relative to the target debt ratios.
As a whole, firms tend to adjust at a faster rate when firms have a negative adjustment gap with the
financial constraint of over-leverage to gear down leverage due to the high risk of bankruptcy. The
evidence on the variation in the adjustment rate of debt ratio is consistent with Byoun (2008) but
does not support the constant adjustment rate over time as Flannery and Rangan (2006) conclude.
Finally, Taiwan has a most successful record of economic transition from being a less-developed
country to becoming a newly industrialized country within few decades only. Within this context, the
findings of this study provide new evidence on the effect of macroeconomic conditions on the debt
18
ratio adjustment and the determination of the actual debt ratios for firms with financial constraint of
over-leverage and under-leverage relative to the target debt ratios. Future research might address the
adjustment behaviour of capital structure decisions across the Asian countries to provide further
evidence on the effect of macroeconomic conditions and on the variation in the adjustment rate of
capital structure decisions. In particular, further evidence on the differences in the adjustment
behavior of capital structure between developing and developed countries leaves for future research.
Acknowledgements
An earlier version of the paper was presented at a refereed international scholarly conference hosted
by the Accounting and Finance Association of Australia and New Zealand held at the Gold Coast,
Queensland, Australia from July 1 to 3, 2007 and in the Queensland University of Technology – Griffith
University Finance Symposium organized by Professor of Finance, Michael Drew, and held at Griffith
University in November 2006. The authors are grateful for the helpful comments and suggestions
from the conference and symposium participants.
19
References
Belsey, D. A., Kuh, E., Welsch, R. E. (1980) Regression Diagnostics: Identifying Influential Data and
Sources of Collinearity. John Wiley and Sons, New York.
Bevan, A. A., Danbolt, J. (2002) Capital Structure and its Determinants in the UK: A Decompositional
Analysis. Applied Financial Economics, 12, pp. 159-70.
Booth, L., Aivazian, V., Demirguc-Kunt, A., Maksimovic, V. (2001) Capital Structures in Developing
Countries. Journal of Finance, 56, pp. 87-130.
Byoun, S. (2008) How and When Do Firms Adjust Their Capital Structures toward Targets? Journal of
Finance, 63, pp. 3069-96.
Chatterjee, S., Price, B. (1991) Regression Analysis by Example. John Wiley & Sons, New York.
Chu, P. Y., Wu, S., Chiou, S. F. (1992) The Determinants of Corporate Capital Structure Choice: Taiwan
Evidence. Journal of Management Science, 9, pp. 159-77.
Cronbach, L. (1987) Statistical Tests for Moderator Variables: Flaws in Analysis Recently Proposed.
Psychological Bulletin, 102, pp. 414-17.
Downs, T. W. (1993) Corporate Leverage and Nondebt Tax Shields: Evidence on Crowding-Out.
Financial Review, 28, pp. 549-83.
Ferri, M. G., Jones, W. H. (1979) Determinants of Financial Structure: A New Methodological
Approach. Journal of Finance, 34, pp. 631-44.
Flannery, M. J., Rangan, K. P. (2006) Partial Adjustment toward Target Capital Structures. Journal of
Financial Economics, 79, pp. 469-506.
Glen, J., Singh, A. (2004) Comparing Capital Structures and Rates of Return in Developed and
Emerging Markets. Emerging Markets Review, 5, pp. 161-92.
Hackbarth, D., Miao, J., Morellec, E. (2006) Capital Structure, Credit Risk, and Macroeconomic
Conditions. Journal of Financial Economics, 82, pp. 519-50.
Harris, M., Raviv, A. (1991) The Theory of Capital Structure. Journal of Finance, 46, pp. 297-355.
Hovakimian, A., Opler, T., Titman, S. (2001) The Debt-Equity Choice. Journal of Financial and
Quantitative Analysis, 36, pp. 1-24.
Huang, G., Song, F. M. (2006) The Determinants of Capital Structure: Evidence from China. China
Economic Review, 17, pp. 14-36.
20
Jaccard, J., Turrisi, R. (2003) Interaction Effects in Multiple Regression. Sage, Thousand Oaks.
Jensen, M. C. (1986) Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers. American
Economic Review, 76, pp. 323-29.
Kim, W. S., Sorensen, E. H. (1986) Evidence on the Impact of the Agency Costs of Debt on Corporate
Debt Policy. Journal of Financial and Quantitative Analysis, 21, pp. 131-44.
Korajczyk, R. A., Levy, A. (2003) Capital Structure Choice: Macroeconomic Conditions and Financial
Constraints. Journal of Financial Economics, 68, pp. 75-109.
Levy, A., Hennessy, C. (2007) Why Does Capital Structure Choice Vary with Macroeconomic
Conditions? Journal of Monetary Economics, 54, pp. 1545-64.
Marsh, P. (1982) The Choice Between Equity and Debt: An Empirical Study. Journal of Finance, 37, pp.
121-44.
Miller, M. H. (1977) Debt and Taxes. Journal of Finance, 32, pp. 261-75.
Modigliani, F., Miller, M. H. (1958) The Cost of Capital, Corporation Finance and the Theory of
Investment. American Economic Review, 48, pp. 261-97.
Myers, S. C., Majluf, N. S. (1984) Corporate Financing and Investment Decisions When Firms Have
Information That Investors Do Not Have. Journal of Financial Economics, 13, pp. 187-221.
Narayanan, M. P. (1988) Debt versus Equity under Asymmetric Information. Journal of Financial and
Quantitative Analysis, 23, pp. 39-51.
Rajan, R. G., Zingales, L. (1995) What Do We Know about Capital Structure? Some Evidence from
International Data. Journal of Finance, 50, pp. 1421-60.
Ross, S. A. (1977) The Determination of Financial Structure: The Incentive-Signalling Approach. The
Bell Journal of Economics, 8, pp. 23-40.
Stulz, R. M. (1990) Managerial Discretion and Optimal Financing Policies. Journal of Financial
Economics, 26, pp. 3-27.
Taggart, R. A., Jr. (1977) A Model of Corporate Financing Decisions. Journal of Finance, 32, pp. 1467-
84.
Titman, S., Wessels, R. (1988) The Determinants of Capital Structure Choice. Journal of Finance, 43,
pp. 1-19.
Wald, J. K. (1999) How Firm Characteristics Affect Capital Structure: An International Comparison.
Journal of Financial Research, 22, pp. 161-87.
21
Wiwattanakantang, Y. (1999) An Empirical Study on the Determinants of the Capital Structure of Thai
Firms. Pacific-Basin Finance Journal, 7, pp. 371-403.