How reliably do empirical tests identify tax avoidance?

How reliably do empirical tests identify tax avoidance?

By Lisa De Simone, Jordan Nickerson, Jeri K. Seidman, Bridget Stomberg
December 16,2016Working Paper No. 3446

Can common empirical tests reliably identify tax avoidance? This is an important question because our understanding of the determinants of tax avoidance largely depends on results generated using such tests. We seed Compustat data with three tax avoidance strategies and examine how reliably tests of tax avoidance models using effective tax rates and book-tax differences identify this incremental simulated tax avoidance, holding constant known economic determinants of tax avoidance. We find permanent tax avoidance is more easily detected than deferral strategies, and that financial reporting choices can reduce statistical power. We also conclude that, holding constant other research design decisions, power varies by which proxy is the dependent variable and with the magnitude and pervasiveness of tax avoidance in the sample. We further offer evidence on how research design choices such as sample selection and the approach to handling outliers affect power. We contribute to the literature by using a controlled environment to examine the effectiveness of existing empirical tests.