Book versus Taxable Income
Florida State University - College of Business
State University of New York (SUNY) at Buffalo - Department of Accounting
Income reported to shareholders (book income) and income reported to the U.S. Internal Revenue Service (taxable income) are alternative measures of U.S. corporate economic performance discussed in recent research, academic texts, and by U.S. legislators. In measuring economic performance, taxable income contains more 'mandated rule' error than book income because legislators generally write tax rules to raise revenue and motivate behavior and not specifically to measure performance. However, recent research suggests taxable income likely contains less 'managerial bias' error (i.e. error due to managers' biased application of rules) than book income. We decompose book income and taxable income into operating cash flows, book accruals, and tax accruals. We find that tax accruals have explanatory power for contemporaneous annual stock returns incremental to book accruals and operating cash flows, but that book accruals have relatively more explanatory power. However, the dominance of book accruals over tax accruals declines significantly and from 1997-2001, tax accruals explain returns as well as book accruals. In contrast to the conventional notion that low book relative to taxable income indicates high quality book income and vice versa, we find the opposite: tax accruals explain returns better than book accruals when the ratio of book to taxable income is low. Book losses partially, but do not fully, explain our results. Our results are relevant to policy debates concerning book-tax differences and to explanations for changes in the value relevance of book income during the 1990's.
Number of Pages in PDF File: 38
Keywords: taxable income, earnings, capital markets, financial reporting, earnings quality
JEL Classification: G12, G18, H25, K20, K34, M41, M43, M44
Date posted: October 15, 2004
© 2015 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo4 in 1.172 seconds