Testing static and dynamic leverage models: A standardized leverage measure approach

Abstract

This paper proposes and validates a new measure of capital structure by normalizing firms’ leverage ratios by their business risks, assuming asset values follow a geometric Brownian motion. We empirically test the new measure with both static and dynamic leverage models, and we find that the standardized leverage (SDL) ratio: 1) performs similarly to book leverage in static leverage regressions, and 2) presents a more realistic speed of adjustment (SOA) in dynamic leverage adjustment models. Within the framework of SDL, we also compute individual firms’ leverage adjustment speeds, which is shown to have significant predictive power regarding firms’ future capital structure adjustments. Overall, the empirical results suggest that the SDL exhibits strong candidacy to serve as a more reliable measure for firms’ real capital structure.