Effective tax rate (ETR) measures: alternatives and their validity

Abstract

Corporate tax related empirical research often uses effective tax rates to investigate cross-sectional differences in the tax attributes of companies. Effective tax rates are often defined as (current or total) tax expense over financial accounting income before taxes, where the (current or total) tax expense is an estimate of the actual taxes paid. This approximation of the actual tax payment by using tax expense is often used, as the actual tax payment and taxable income were or are proprietary information, thereby rendering it difficult if not impossible to use this information to calculate effective tax rates. Due to changes in financial accounting standards more information on the actual tax payment and taxable income is now generally available in US financial statements. This paper argues that the use of an effective tax rate measure utilizing tax expenses has some inherent dangers which are to a lesser extent the case in an effective tax rate using actual taxes paid. Two inherent dangers of a tax expense based ETR measure are the quality of the tax expense measure as an estimation of the actual taxes paid and timing differences between tax expense and actual taxes paid. The differences between tax expenses and actual taxes paid ETRs are researched in an empirical study as well as other proposed ETR measures. Results from this empirical study indicate that in certain situations it is more appropriate to use an effective tax rate measure based on actual taxes paid rather than on tax expenses.