Die relative Steuerbelastung mittelständischer Kapitalgesellschaften: Evidenz von handelsbilanziellen Mikrodaten

Official tax statistics do not provide sufficient disaggregated data to determine the effective relative tax burden of SMEs. Due to secrecy of tax matters, there are also no tax microdata. The only available data are commercial balance sheets and profit-and-loss statements. We discuss the issues of measuring effective tax rates derived from these commercial financial statements and estimate the relative effective tax burden of SME and large corporations in Germany. We use a backward-looking measure via bootstrapped trimmed quantile regression, to guarantee robust results. Backward-looking average tax rates derived from corporate data on actual tax payments can be robust indicators of the corporate tax burden. In contrast to forward-looking measures, our approach controls for the firm-specific tax optimisation strategy, the quality of the tax administration and potentially different degrees of leniency of tax audits. We find evidence for an inverted U-shaped pattern of burden: small enterprises face a significantly lower tax burden than medium-sized enterprises, and medium-sized enterprises face a significantly higher tax burden than large companies. That is, taxation is not size-neutral and medium-sized enterprises face a tax disadvantage. Given the evidence that SMEs represent employment and growth engines, taxation of SMEs in Germany may hamper growth and employment. We also provide empirical evidence that the big German tax reform at the beginning of this millennium („Steuerreform 2000“) did significantly reduce the effective tax burden. Again, the tax relief was different for small, medium and large corporations.