Review of Accounting Studies, 8, 493–530, 2003
# 2003 Kluwer Academic Publishers. Manufactured in The Netherlands.
Accounting Returns Revisited: Evidence of their
Usefulness in Estimating Economic Returns
MORRIS G. DANIELSON* email@example.com
Erivan K. Haub School of Business, Saint Joseph’s University, Philadelphia, PA 19131
ERIC PRESS firstname.lastname@example.org
The Fox School of Business, Temple University, Philadelphia, PA 19122
Abstract. Accounting information is used for measuring ﬁrm performance in various ﬁnancial
applications—a practice supported by empirical studies demonstrating the value relevance of
accounting numbers, but disputed by theoretical papers arguing that a ﬁrm’s accounting rate of return
(ARR) serves poorly as a proxy for its internal rate of return (IRR). We derive a new model of the ARR–
IRR relation, and describe how the conservatism of GAAP constrains a ﬁrm’s IRR to fall in a range
bounded by its historical growth rate and ARR. Using cross-sectional data, we demonstrate that economic
returns can be estimated from accounting numbers for many ﬁrms. We link empirical results to underlying
economic theory, and thus contribute to understanding why accounting information is value relevant.
Keywords: return on investment, economic proﬁt, measuring proﬁtability, value relevance, accounting
JEL Classiﬁcation: G3, L5, M2, M4
Analysts, managers, and researchers use ﬁnancial statement information to measure
ﬁrm performance in a variety of applications. Accounting numbers are used to
measure proﬁtability for compensation plans and in research studies, and often serve
as the starting point for analysts’ estimates of future cash ﬂows for valuation
purposes. To be meaningful in such applications, accounting information must be
informative about the internal rate of return (IRR) generated by a ﬁrm’s underlying
assets. However, a number of theoretical papers—including Fisher and McGowan
(1983), Fisher (1988), Salamon (1985, 1988), and Brief and Lawson (1991)—argue
that accounting rate of return (ARR) serves poorly as a proxy for a ﬁrm’s IRR.
Nevertheless, a wide range of studies demonstrates the value relevance of accounting
data, supporting their use in practice.
Given this conﬂict between theory and empirical evidence, it is unclear why and
when accounting information is relevant. In this paper, we address whether
accounting information can justiﬁably be used for measuring proﬁtability, assessing
managerial performance, or valuing a ﬁrm.
To do so, we derive a model linking accounting and economic returns, using
assumptions equivalent in most respects to previous models of the ARR–IRR
When applied in an unconstrained setting, the model shows that a ﬁrm’s