International Review of Finance, 11:1, 2011: pp. 1–17 DOI: 10.1111/j.1468-2443.2010.01125.x

Two Common Problems in Capital Structure Research: The FinancialDebt-To-Asset Ratio and Issuing Activity Versus Leverage Changes
IVO WELCH Brown University, RI and NBER
ABSTRACT
This paper points out two common problems in capital structure research. First, although it is not clear whether non-financial liabilities should be considered debt, they should never be considered as equity. Yet, the common financial-debt-to-asset ratio (FD/AT) measure of leverage commits this mistake. Thus, research on increases in FD/AT explains, at least in part, decreases in non-financial liabilities. Future research should avoid FD/AT altogether. The paper also quantifies the components of the balance sheet of large publicly traded corporations and discusses the role of cash in measuring leverage ratios. Second, equity-issuing activity should not be viewed as equivalent to capital structure changes. Empirically, the correlation between the two is weak. The capital structure and capital issuing literature are distinct.

I. INTRODUCTION
Leverage is defined as the sensitivity of the value of equity ownership with respect to changes in the underlying value of the firm. Empirically, leverage ratios are frequently independent variables (sometimes as part of a hypothesis, sometimes as a control). Leverage ratios are also the dependent variable in the empirical capital structure literature. This literature tries to explain variations in corporate leverage, both in the cross section of capital structure (i.e. why some firms have high leverage) and in the time series (how capital structures evolve). In the theory of capital structure, one common hypothesis derives directly from the equity-sensitivity channel: a firm with more leverage has both higherpowered incentives and (usually) a higher probability of