Taxation and Earnings Management: Natural Experiment Evidence from Staggered Changes in State Corporate Tax Rates

Abstract

We provide evidence on how statutory corporate tax rate changes affect income-shifting behavior. Our difference-in-differences identification strategy exploits the 112 staggered changes in corporate income tax rates across the 33 U.S. states during the period 1989-2012. We find that accounting earnings are more likely to be managed upward in response to state tax cuts and more significantly among firms with higher prior-year return on assets. Most firms do not manage earnings downward in response to tax increases, probably because significant tax increases usually occur when states are fiscally constrained and downward smoothing of reported earnings can certainly draw scrutiny from cash-strapped state governments. More importantly, firms had higher ROA but experienced lower stock returns prior to tax cuts do use discretionary accruals to increase earnings in the year of the tax cut and reduce earnings in the year of the tax rise, a strategy that improves returns to shareholders. This evidence suggests that firms often trade off the tax benefits of earnings management to maximize firm value for a lower risk of being detected by the regulators and other market participants.

Keywords: corporate income tax, earnings management, natural experiment

JEL Classification: G32, G38, M41