Tax Avoidance - Are Banks Any Different?

Abstract

While the public has noticed the need for the detection of potential tax loopholes and demands further improvement in the taxation of banks, there is scarce empirical evidence whether banks’ degree of tax avoidance actually differs from that of non-banks. We try to close this gap by investigating U.S. banks’ tax avoidance behavior for a sample period from 2004 to 2016. To identify banks’ tax avoidance, we use annual Cash ETRs and GAAP ETRs and compare them to the tax avoidance behavior of non-banks. As there are various channels of tax avoidance, we account for differences in several areas such as corporate fundamentals, the degree of multinationality and regulatory scrutiny. We provide cautious evidence that banks have significantly higher Cash ETRs than non-banks. Using quantile regression, we find evidence that the assocation between banks and ETRs is not constant over the whole tax avoidance distribution, with banks reporting significantly higher ETRs compared to nonbanks in those regions of the tax avoidance distribution which are regularly classified as “high tax avoidance”. In line with recent research, we provide some evidence that the difference in Cash ETRs between banks and non-banks is more pronounced for worse-capitalized, than for better-capitalized banks.

Keywords: tax avoidance, banks, non-banks, standardized coefficients, quantile regression

JEL Classification: G21, H26, M41