A Dynamic Model of Corporate Capital Structure

Abstract

I present a dynamic empirical model for capital structure. The result of the dynamic panel data estimate is that fixed asset and the lagged-debt variables explain leverage. Conditioning the debt level on the lagged-debt level reveals that the significance of other parametes fall as compared to a static model, and can be interpreted as support for the pecking order hypothesis. The dynamic model is estimated by GMM methods that yield consistent parameter estimates. I have shown that the capital structure is changing at a slow pace. The firms change only about 28% of the total debt each year. The debt ratio of low and high level debt firms change more for each year, they are less dependent on past year debt ratio. This finding shows that the firms set a target level for their debt structure.

JEL Classification: G32