Corporate tax competition: A meta-analysis

Abstract

Do jurisdictions compete over corporate taxes? The extent of uncooperative strategic government interactions in corporate tax policy remains unclear, because the results reported in relevant econometric studies vary considerably. This paper applies meta-analysis and meta-regression methods to a novel dataset consisting of 604 observations on corporate tax competition reported in 33 primary studies. The results point to the presence of corporate tax competition effects, and there is no consistent evidence for publication selection bias. The analysis, however, reveals that several data and specification choices systematically affect the results reported in the literature: the choice of the weighting scheme applied in estimating the corporate tax function and details of the econometric estimation strategy are significant when it comes to explaining variation in reported results. Furthermore, we find that accounting for partisan politics and country size matters.