Arqus-Diskussionsbeiträge zur quantitativen Steuerlehre 182
The theoretical literature suggests that when taking tax effects into account, debt ought to be preferable to equity. However, there are no uniform predictions of the size of this tax benefit (tax shield) in comparison to an opposing increasing cost of debt (especially insolvency costs). The vast body of empirical studies on the impact of taxation on capital structure only provides puzzling effects. We believe the German corporate tax reform in 2008, which introduced an interest barrier as a "quasiexperiment", is a promising opportunity to investigate the effects that arise from a reform of interest deductibility. We study capital structure adjustments empirically using financial statement data from German companies. We consider a study of German tax reform on the basis of German data of general interest because, first, similar tax reforms have been conducted in several countries. Second, the availability of single entity financial statements for German companies allow us to capture tax and capital structure details that have not been available in most prior studies. Third, the major characteristics of the German tax system can be regarded as representative for most European and the major Asian countries. All of this information enables us to contribute to solving the capital structure puzzle in a unique way. With significance at the 5% level, we find evidence that the companies that are affected by the interest barrier reduce their leverage by 3 percentage points more than companies that are not affected. We are the first to employ a detailed matching approach to the underlying rich dataset, which enables us to overcome some of the limitations of previous studies. While many prior empirical studies on capital structure have provided mixed results on capital structure reactions, we find robust evidence for the impact of tax reforms on corporations' financing decisions.