R&D Reporting Biases and their Consequences
New York University - Stern School of Business
Rutgers, The State University of New Jersey - Accounting
University of Illinois at Urbana-Champaign - Department of Accountancy
Contemporary Accounting Research, Vol. 22, No. 4, Winter 2005
The immediate expensing of R&D expenditures is often justified by the conservatism principle. However, no accounting procedure consistently applied can be conservative throughout the firm's life. We ask the following questions: (a) When is the expensing of R&D conservative and when is it aggressive, relative to R&D capitalization? and (b) What are the capital market implications of these reporting biases? To address these questions we construct a model of profitability biases (differences between reported profitability under R&D expensing and capitalization) and show that the key drivers of the reporting biases are the differences between R&D growth and earnings growth (momentum), and between R&D growth and return on equity (ROE). Companies with a high R&D growth rate relative to their profitability (typically early cycle companies) report conservatively, while firms with a low R&D growth rate (mature companies) tend to report aggressively under current GAAP. Our empirical analysis, covering the period 1972-2003, generally supports the analytical predictions.
In the valuation analysis we find evidence consistent with investor fixation on the reported profitability measures: we detect undervaluation of conservatively reporting firms and overvaluation of aggressively reporting firms. These misvaluations appear to be corrected when the reporting biases reverse from conservative to aggressive and vice versa. This evidence is consistent with behavioral finance arguments about investor cognitive biases.
Keywords: R&D Accounting, Reporting Biases, Market Valuation, Mispricing
JEL Classification: M41, M44, O30, G14
Date posted: September 1, 2005
© 2015 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo5 in 0.360 seconds