Capital Structure Policy and Behavioral Influences
Malmendier, Tate, and Yan claimed that early life experience can play a large role in corporate finance decision making.18 They examined the capital structure policies of CEOs who had served in the military and those who had experienced an economic downturn early in their lifetime. They found that World War II veterans were more aggressive than other CEOs. In combat, soldiers manage stressful and risky situations that build their confidence in risk management. As a result, veterans tend to be willing to bear more risk. In contrast, CEOs who have experienced economic downturns tend to be more conservative. Experiencing an economic depression can have a deep and long-lasting impact that causes an aversion to financial risk taking. The study also looked at the effect of overconfidence on capital structure policy. CEOs who thought their firms were undervalued perceived external financing to be prohibitively expensive. When external financing was needed, these CEOs exhibited a marked aversion to equity. When issuing debt to finance a deficit, overconfident CEOs also tended to issue about 33 percent more debt than needed.
Robb and Robinson's examination of the capital structure decisions of new firms provides useful insight into behavioral characteristics and the capital structure choice.19 In particular, the study found that women were less likely to issue debt, and black-owned businesses and businesses started by people without high school degrees were less likely to obtain external financing of any kind. However, the authors also found that firms that obtained formal external financing were more successful, and that those that did not were more likely to fail within three years. This is presumably due to the additional management oversight provided by external stakeholders and financial markets.
Estimating the Costs and Benefits of Debt
Trade-off theory suggests that there is an optimal level of debt that balances the costs and benefits of debt. Various models suggest that the optimal capital structure consists of about 65 percent debt, whereas observed debt levels have been, on average, about 30 percent. Only recently, however, has the research begun to adequately quantify the costs and benefits of debt.
The Value of the Tax Shield
Graham examined the value of the tax shield and found that the aggregate tax benefits of the shield were significant. At their highest, tax benefits of debt for a sample of 6,087 firms totaled $114 billion in 1990.20 He also created a model to measure the value of the interest tax shield at the firm level and found that the tax benefits of debt amounted to 9.7 percent of firm value. However, despite the significant benefits of debt financing, even large, well-established firms tended to have conservative debt policies. More than 44 percent of the firms examined underutilized debt despite the fact that even after taking into account the costs of bankruptcy, firms could still benefit from the debt tax shield if they were to double their debt. Blouin, Core, and Guay claimed that Graham's study overestimated the tax benefits of debt, and that estimates of future tax-related cash flows were flawed since other tax benefits such as tax loss carry forwards and carrybacks had been ignored.21 Using an improved method to estimate marginal tax rates, they showed that although firms were not as underleveraged as previously thought, between 18.5 percent and 29.2 percent of them were underleveraged, and that an increase in leverage could have added 1-3 percent of value to the firm.22
The Costs of Debt
Studies of the cost of debt have historically focused on small samples of bankrupt firms. These studies estimated bankruptcy costs to be between 3.1 percent and 20 percent of firm value. Building on Graham's work, Van Binsbergen, Graham, and Yang developed a model to analyze the costs and benefits of debt.23 They used a model to develop an optimal capital structure, and compared the costs and benefits of this optimal (equilibrium) capital structure to observed capital structures. Their findings suggested that default costs amounted to 6 percent of book value for investment-grade firms and 17 percent for low-grade firms. The average benefit of debt at the equilibrium level was 10.4 percent compared to an observed benefit of 9.0 percent. The cost of debt at the equilibrium level was 6.9 percent compared to an observed level of 7.9 percent. The equilibrium net benefit of debt was 3.5 percent compared to an observed benefit of 1.1 percent, which suggests that firms are underleveraged.
Costs of Suboptimal Capital Structure
Van Binsbergen, Graham, and Yang quantified the costs of operating with suboptimal capital structures.24 Their results showed that the cost of having too much debt was disproportionately higher than the cost of having too little debt. If the average firms with equilibrium capital structures were to double their leverage, they would lose about 6.7 percent of firm value, whereas if the same firms were to eliminate their debt they would lose 4.5 percent of their value.