Abstract
In contrast to the capital structure propositions in Modigliani & Miller (1958; 1963), prior research has not been able to find a clear-cut positive association between stock returns and financial leverage. Several reasons have been proposed for this, including the mispricing of financial risk in stock markets. However, the results might also be due to the statistical models in prior research failing to properly capture the financial risk associated with leverage. Return regressions investigating the impact of leverage on stock returns must explicitly accommodate both the (positive) compounding operating business risk due to leverage and the (negative) interest cost of financial debt. Disentangling both these effects and controlling for unexpected cash flow and macro news in stock returns, we report tests of the association between expected stock returns and leverage with US data from 1966 - 2017. With the enterprise book-to-price indicating operating business risk, we find a positive compounding return effect of leverage and a negative effect of leverage itself. Including the enterprise earnings yield and other indicators of operating business risk, the compounding risk of leverage becomes weaker, and leverage affects returns negatively only for commercially unviable firms. Our analyses also reveal a mitigating risk effect associated with financial leverage, likely stemming from the disciplining of agency risks through debt contracting. Even though the composite return effect of leverage is positive on average, the observed mitigating effect of leverage contradicts a core assumption in Modigliani & Miller (1958, 1963). In turn, this means that commonly used leverage formulas in economics and professional investment practice might overstate the effect of leverage on expected stock returns.