Functional form issues in the regression analysis of corporate capital structure

Abstract

Linear models are typically used in the regression analysis of capital structure choices. However, given the proportional and the bounded nature of leverage ratios, models such as the tobit, the fractional regression model and its two-part variant are a better alternative. In this article, we discuss the main econometric assumptions and features of those models, provide a theoretical foundation for their use in the regression analysis of leverage ratios and review some statistical tests suitable to assess their specification. Using a dataset previously considered in the literature, we carry out a comprehensive comparison of the alternative models, finding that in this framework the most relevant functional form issue is the choice between a single model for all capital structure decisions and a two-part model that explains separately the decisions to issue debt and, conditional on the first decision, on the amount of debt to issue.