ساختار سرمایه و عملکرد شرکت: روش جدیدی برای آزمون تئوری نمایندگی و یک برنامه کاربردی برای صنعت بانکداری

Corporate governance theory predicts that leverage affects agency costs and thereby influences firm performance. We propose a new approach to test this theory using profit efficiency, or how close a firm’s profits are to the benchmark of a best-practice firm facing the same exogenous conditions. We are also the first to employ a simultaneous-equations model that accounts for reverse causality from performance to capital structure. We find that data on the US banking industry are consistent with the theory, and the results are statistically significant, economically significant, and robust.

Agency costs represent important issues in corporate governance in both financial and nonfinancial industries. The separation of ownership and control in a professionally managed firm – one source of agency conflicts – may result in managers exerting insufficient work effort, indulging in perquisites, choosing inputs or outputs that suit their own preferences, or otherwise failing to maximize firm value. In effect, the agency costs of outside ownership equal the lost value from professional managers maximizing their own utility, rather than the value of the firm. Theory suggests that the choice of capital structure may help mitigate these agency costs. Under the agency costs hypothesis, high leverage or a low equity/asset ratio reduces the agency costs of outside equity and increases firm value by constraining or encouraging managers to act more in the interests of shareholders. Since the seminal paper by Jensen and Meckling (1976), a vast literature on such agency-theoretic explanations of capital structure has developed (see Harris and Raviv, 1991 and Myers, 2001 for reviews). Greater financial leverage may affect managers and reduce agency costs through the threat of liquidation, which causes personal losses to managers of salaries, reputation, perquisites, etc. (e.g., Grossman and Hart, 1982 and Williams, 1987), and through pressure to generate cash flow to pay interest expenses (e.g., Jensen, 1986). Higher leverage can mitigate conflicts between shareholders and managers concerning the choice of investment (e.g., Myers, 1977), the amount of risk to undertake (e.g., Jensen and Meckling, 1976 and Williams, 1987), the conditions under which the firm is liquidated (e.g., Harris and Raviv, 1990), and dividend policy (e.g., Stulz, 1990). Whereas increased leverage may reduce the agency costs of outside equity, the opposite effect may occur for the agency costs of outside debt arising from conflicts between debt holders and shareholders. When leverage becomes relatively high, further increases may generate significant agency costs of outside debt from risk shifting or reduced effort to control risk that result in higher expected costs of financial distress, bankruptcy, or liquidation. These agency costs result in higher interest expenses for firms to compensate debt holders for their expected losses. As pointed out by Jensen and Meckling (1976), the effect of leverage on total agency costs is expected to be nonmonotonic. At low levels of leverage, increases will produce positive incentives for managers and reduce total agency costs by reducing the agency costs of outside equity. However, at some point where bankruptcy and distress become more likely, the agency costs of outside debt overwhelm the agency costs of outside equity, so further increases in leverage result in higher total agency costs. Agency costs and capital structure issues raise particularly important research and policy questions regarding the banking industry. This industry plays crucial roles in providing credit to nonfinancial firms, in transmitting the effects of monetary policy, and in providing stability to the economy as a whole. Agency costs may be particularly large in this industry because banks are by their very nature informationally opaque – they hold private information on their loan customers and other credit counterparties. In addition, banks’ access to government deposit insurance and other safety net protections may increase incentives for risk shifting or lax risk management, potentially increasing the agency costs of outside debt. However, to offset these incentives, bank regulators directly affect capital structure by setting minimums for equity capital and other types of regulatory capital. As well, regulators conduct examinations and take other actions to keep the expected costs of financial distress, bankruptcy, or liquidation relatively low, which may reduce the agency costs of outside debt. The marginal effects of changes in leverage on total agency costs may be expected to vary significantly in the banking industry, and we are able to test for such differences in our analysis below. The empirical evidence on the agency costs hypothesis in the banking literature and in the finance literature as a whole is mixed (see Harris and Raviv, 1991 and Titman, 2000; and Myers, 2001 for reviews). Tests of the agency costs hypothesis typically are based on regressions of measures of firm performance on the equity capital ratio or other indicator of leverage plus some control variables. In some cases a negative relationship is found, but opposite results are documented as well. The absence of clear-cut evidence could be partly explained by the intrinsic difficulty in defining a measure of performance that is close to the theoretical definition of agency costs. The measures of firm performance are usually ratios fashioned from financial statements or stock market data, such as industry-adjusted operating margins or stock market returns. These measures usually do not net out the effects of differences in exogenous firm-specific conditions that may affect firm value, but are beyond management’s control and therefore cannot reflect agency costs. Thus, the tests may be confounded by factors that are unrelated to agency costs. As well, these studies generally do not set a separate benchmark for each firm’s performance that would be realized if agency costs were minimized. We address these measurement issues by using profit efficiency as our indicator of firm performance. The conceptual link between productive efficiency and agency costs was first suggested by Stigler (1976), and profit efficiency represents a refinement of the efficiency concept developed since that time.1 Profit efficiency evaluates how close a firm is to earning the profit that a best-practice firm would earn facing its same exogenous conditions. This has the benefit of controlling for firm-specific factors outside the control of management that are not part of agency costs. The financial ratios, stock market returns, and similar measures used in the literature typically at most are industry adjusted, and do not account for important differences across firms within an industry – such as local market conditions – as we are able to do with profit efficiency. We also argue that the performance of a best-practice firm under the same exogenous conditions is a reasonable benchmark for how the firm would be expected to perform if agency costs were minimized. Although the use of profit efficiency addresses some of the difficulties in other performance measures employed in the literature, the profit efficiency measures are also imprecise and embody measurement error. At a minimum, the use of profit efficiency provides alternative measures to those based on financial ratios and stock market values to provide a more complete picture. The mixed results in the prior research may also be due to the possibility of reverse causation from performance to capital structure. If firm performance affects the choice of capital structure, then failure to take this reverse causality into account may result in simultaneous-equations bias. That is, regressions of firm performance on a measure of leverage may confound the effects of capital structure on performance with the effects of performance on capital structure. We address this issue by allowing for reverse causality from performance to capital structure based on two hypotheses for why firm performance may affect the choice of capital structure. Under the efficiency-risk hypothesis, more efficient firms tend to choose relatively low equity ratios, as higher expected returns from the greater profit efficiency substitutes to some degree for equity capital in protecting the firm against distress, bankruptcy, or liquidation. Under the franchise-value hypothesis, more efficient firms tend to choose relatively high equity ratios to protect the future income derived from high profit efficiency. We construct a two-equation structural model and estimate it using two-stage least squares (2SLS). An equation specifying profit efficiency as a function of the firm’s equity capital ratio (the ratio of equity to gross total assets) and other variables is used to test the agency costs hypothesis, and an equation specifying the equity capital ratio as a function of the firm’s profit efficiency and other variables is used to test the net effects of the efficiency-risk and franchise-value hypotheses, i.e., the extent to which the substitution versus income effects dominate the equity capital choice. Both equations are econometrically identified through exclusion restrictions that are consistent with the theories. In addition to the important research and policy questions regarding the banking industry, the application to banking is also an advantageous laboratory for testing agency theory because of the abundance of quality data available on firms in this industry. In particular, we have detailed financial data for a large number of firms producing comparable products with similar technologies, and information on market prices and other exogenous conditions in the local markets in which they operate. The availability of quality data and importance of the questions involved have drawn many researchers to study agency problems and related issues in the banking industry, although none specifically analyze the effects of the capital ratio or leverage on profit efficiency or used a structural model to address the problem of reverse causality. Nonetheless, some studies find evidence of links between bank efficiency and variables that are recognized to affect agency costs, such as the governance and ownership structures, suggesting that our approach may be fruitful (e.g., Mester, 1993, Pi and Timme, 1993 and DeYoung et al., 2001). Although banking is a regulated industry, banks are subject to the same type of agency costs and other influences on behavior as other industries. The banks in the sample are subject to essentially equal regulatory constraints, and we focus on differences across banks, not between banks and other firms. Most banks are well above the regulatory capital minimums, and our results are based primarily on differences at the margin, rather than the effects of regulation. By way of preview, our empirical results are consistent with the agency costs hypothesis that higher leverage or a lower equity capital ratio in banking is associated with higher profit efficiency. The effect is both economically significant and statistically significant, and continues to hold through a number of robustness checks. The data do not suggest that that the relationship is reversed when leverage is high, but there is some evidence of a smaller marginal effect of equity capital when leverage is very high, which may reflect some offsetting effects from the agency costs of debt. Finally, we find that neither hypothesis of reverse causality from efficiency to equity dominates the other over the range of the data. The paper is organized as follows. Section 2 illustrates how we measure performance and how we address the issue of reverse causality. Section 3 specifies the simultaneous equations model for testing the hypotheses, and Section 4 describes the data and variables employed in the model. Section 5 describes the empirical results, and Section 6 concludes. Appendix A details the efficiency estimation.

We test the agency costs hypothesis, an empirical description of a well-known hypothesis in corporate finance. Under this hypothesis, higher leverage reduces the agency costs of outside equity and increases firm value by constraining or encouraging managers to act more in the interests of shareholders. However, when leverage becomes relatively high, elevating the expected costs of financial distress, bankruptcy, or liquidation, the agency costs of outside debt may overwhelm the agency costs of outside equity, so further increases in leverage result in higher total agency costs. In this study, we use profit efficiency as an indicator of firm performance to measure agency costs and employ a two-equation structural model that also takes into account reverse causality from firm performance to capital structure. Our main findings are consistent with the agency costs hypothesis – higher leverage or a lower equity capital ratio is associated with higher profit efficiency over almost the entire range of the observed data. The effect is economically significant as well as statistically significant. For example, an increase in leverage as represented by a 1 percentage point decrease in the equity capital ratio yields a predicted increase in standard profit efficiency of about 16 percentage points at the sample mean in our main findings. The data are not consistent with reversal of the relationship when leverage is high, although we find some evidence of a smaller marginal effect of equity capital when leverage is very high, possibly reflecting some offsetting effects from the agency costs of debt. With respect to the reverse causality from efficiency to capital structure, we offer two competing hypotheses with opposite predictions, and we interpret our tests as suggesting which hypothesis may empirically dominate the other in affecting the equilibrium capital structure. Under the efficiency-risk hypothesis, the expected high earnings from greater profit efficiency substitute for equity capital in protecting the firm from the expected costs of bankruptcy or financial distress, whereas under the franchise-value hypothesis, firms try to protect the expected income stream from high profit efficiency by holding additional equity capital. We find that neither hypothesis of reverse causality from efficiency to equity dominates the other over the range of the data. The findings in the paper are generally robust to a number of tests. These include the use of different functional forms (linear, quadratic, spline), different measures of performance (standard and profit efficiency – each measured different ways – and return on equity), different samples (ownership sample, full sample, ranges of inside ownership, excluding those with high growth), different sample periods (1990s and 1980s), and different modeling approach (structural model versus Granger causality tests). The results in this paper – which are consistent with the agency costs hypothesis – stand in contrast to the mixed empirical evidence in the banking literature and in the finance literature as a whole. A likely reason for the difference in our view is our different methodology, including the specification of profit efficiency as an inverse measure of agency costs and the use of a structural model to address the problem of reverse causality. Future studies using these methods would be needed to clarify this issue. Finally, we note that the approach developed here can be built upon to test other hypotheses with predictions for agency costs using data from virtually any industry.