Theoretical arguments for the tax sensitivity of capital structures are convincing. Empirical findings instead have for years been rather weak. Even today, despite a surge of studies providing point estimates for the tax effect on corporate capital structure, the empirical evidence remains ambiguous. Surprisingly, however, no study has ever quantitatively examined the factors which determine the variation in empirical evidence.

The contribution of this paper is to fill this gap. It provides a comprehensive quantitative review of the empirical literature on the impact of taxation on corporate debt financing. Synthesizing the evidence from over 1,000 primary estimates extracted out of 46 studies, we find that this impact is indeed quite substantial. Our results suggest that, in particular, the tax rate proxy used for identification determines the outcome of primary analyses. More refined measures like the simulated marginal tax rate suggested by Graham (1996, 1999) avoid a significant downward bias in estimates for the debt response to tax. Moreover, we find that debt characteristics, the econometric specification, the set of control-variables, and publication selection in primary studies exert significant influence on estimated tax effects. Accounting for all potential misspecification biases by means of meta-regression analyses, we predict a marginal tax effect on the debt ratio of 0.3.

Keywords

Capital Structure, Corporate Income Tax, Meta-Analysis