Haas School of Business University of California at Berkeley
Abstract This paper argues that matching expenses to revenues increases earnings’ usefulness to investors by providing an accounting rate of return (ARR) closer to current economic proﬁtability. To test this, I estimate a proxy for a ﬁrm’s internal rate of return (IRR) in order to approximate the distortion between ARR and IRR. Results show that the magnitude of the diﬀerence is reduced for ﬁrms with a higher degree of matched expenses. The median magnitude is 2.3% for the highest matching quintile, suggesting that matching expenses provides a reasonable proxy for economic proﬁtability. Secondly, it is argued that reducing the gap between ARR and IRR leads to more timely information included in current earnings. This is evidenced by a higher return reaction to current earnings (ERC) and a lower return relation with future earnings (FERC) for ﬁrms with a smaller gap between ARR and IRR and higher degrees of matched expenses. Keywords: Matching Principle, internal rate of return, timeliness, earnings quality
This paper is still in an early stage and I would greatly appreciate any feedback you might have. I thank Patricia Dechow, Sunil Dutta, Alexander Nezlobin, Richard Sloan, Ed Johnson, Alastair Lawrence, Panos Patatoukas, Malachy English, Aydin Usal and Eric Allen for many helpful discussions.
October 8, 2012
1. Introduction Economists and academics such as Peasnell (1996) or Danielson and Press (2003) have often advocated that a ﬁrm’s accounting must be informative about a ﬁrm’s current underlying economic proﬁtability in order to be useful in measuring proﬁtability, pricing shares or assessing managerial performance. However, while it is easy to see why a better measure of current economic performance leads to a more accurate assessment of managerial performance, the link between current performance and its importance in valuing a company is less clear. After all it is future proﬁtability not current profitability that matters for future value. This study aims to contribute to the debate by empirically testing (1) how well accounting numbers reﬂect economic proﬁtability in the cross-section and (2) whether a close approximation is associated with more timely value relevant information.
Theoretically, the only accounting that equates ARR and IRR is neutral accounting (See Solomon (1966); Hotelling (1925)) based on economic depreciation. Even full matching of expenses does not generally equal neutral accounting and it is conceptually diﬃcult to calculate the magnitude of divergence between the two. Therefore, it is of great interest to regulators, academics and practitioners alike to get an estimate of the magnitude of divergence in practice and its correlation with value relevance. This is the main motivation of the paper.
Hence the study ﬁts into the literature about accounting rates of return as a measure of economic proﬁtability around Fisher and McGowan (1983) as well as the classic earnings quality literature on what accounting rules best reﬂect sustainable and value relevant earnings. The analysis accordingly is in the spirit of prior work such as Penman and Zhang (2002), who examine how unconditional conservatism aﬀect the persistence of earnings and investors’ judgment of sustainable earnings. My contribution to this 2
literature is twofold: First, I empirically measure the link between the accounting principle of matching expenses to revenues1 with the economic property of providing a closer approximation of the current economic rate of return. The economic rate of return is hereby deﬁned as the internal rate of return on a ﬁrm’s current projects.2 To achieve this, two new empirical constructs, matching precision as well as an alternative IRR estimator, are introduced. With these I examine the magnitude of divergence...