Capital structure decision making on the traditional basis of the Modigliani–Miller (American Economic Review 1958 and 1963) propositions I and II poses several unresolved questions for practitioners. The authors find that a firm’s cost of equity is positively related to leverage deviation (i.e., actual leverage minus target optimal leverage) for overleveraged firms. Overleveraged firms exhibit a smaller leverage deviation when their cost of equity is more sensitive to leverage deviation. Underleveraged firms do not exhibit the same link; thus, the increase in the cost of equity is important for overleveraged firms but not for underleveraged firms. The speed of adjustment toward target optimal leverage is not homogeneous across firms because firms whose cost of equity is more sensitive to leverage deviation adjust more rapidly.

Prior research on capital structure has focused on three primary theories: the static trade-off model, the pecking order model, and the market-timing model. These theories are effective at explaining broad patterns but not the observed cross-sectional heterogeneity in capital structure. The trade-off theory, however, can hold in significant subsamples, such as firms that are more sensitive to leverage deviation. The speed of adjustment of the subsample of firms whose cost of equity is highly sensitive to leverage deviation is significantly higher than that of the entire pooled sample. This more rapid speed of adjustment supports the trade-off theory. The results in the capital structure literature are mixed, and capital structure decisions are not equally important across all firms. Therefore, it is important to explore meaningfully chosen subsamples.

The authors use a sample of 12,147 firm-year observations over 1980–2011, obtained from CRSP, Compustat, and Federal Reserve Banks. The dependent and independent variables are winsorized at the first and ninety-ninth percentiles to reduce the impact of outliers, and observations with missing marginal tax rate data are omitted.

The authors’ three hypotheses concern the impact of leverage deviation on the cost of equity, the impact of cost-of-equity sensitivity on leverage deviation, and the impact of cost-of-equity sensitivity on the speed of adjustment. The authors use regression models to test the three hypotheses. In testing the first and second hypotheses, they split the full sample into a subsample of overleveraged firms and a subsample of underleveraged firms. In testing the third hypothesis, they compare results from the full sample with results from subsamples ranked by the sensitivity of the cost of equity to leverage deviation. The authors conduct robustness tests by checking the results obtained using alternative measures of the corporate leverage ratio, an alternative set of determinants in target leverage estimation, and an alternative sample period. Although the results for the three hypotheses are robust, support for the second and third hypotheses is weak under the alternative market leverage measure.

The Modigliani–Miller propositions I and II suggest, among other things, a positive correlation between the cost of equity and leverage. The Modigliani–Miller theory is one of the most famous in finance, despite the fact that additional research is needed to reconcile weaknesses arising from the offsetting effect of heterogeneous firms and the overall mixed results in the empirical literature. The authors successfully expand on the Modigliani–Miller theory by demonstrating how deviations from the target optimal leverage affect the cost of equity.

Practitioners who make financing decisions (e.g., company treasurers and chief financial officers) can use the authors’ research in setting their own firms’ capital structure. Understanding to what extent other firms adhere to the Modigliani–Miller theory can provide a meaningful frame of reference for capital structure practitioners. We know from the literature that a firm’s capital structure is often comparable to that of similar firms in the same industry. Academics will appreciate the extension of the Modigliani–Miller theory, which has yielded many empirical disappointments and requires additional research to reconcile theory with practicality.