Disentangling Shareholder Risk Aversion from Leverage-Dependent Borrowing Cost on Corporate Policies

We study the optimal corporate policy of a risk-averse shareholder under leverage-dependent borrowing costs and other financial frictions. The firm’s objective is to maximize the risk-adjusted shareholder value by co-optimizing investment, dividend, and debt policies considering endogenous (leverage-dependent) leveraging costs, tax shield, as well as costs of equity issuance and asset fire sale. The resulting multistage stochastic linear program model is efficiently solved by the Stochastic Dual Dynamic Programming algorithm. After certifying that the risk-neutral results are consistent with previous studies, our model helps to resolve the well-known low-leverage puzzle, a dissonance between the empirical and structural modeling literature in corporate finance. Our case study results show that risk-aversion combined with leverage-dependent borrowing cost can significantly reduce the optimal leverage ratio as well as the firm size without significantly compromising dividend payments.