The Role of Analysts’ Forecasts in Accounting-Based
Valuation: A Critical Evaluation
QIANG CHENG firstname.lastname@example.org
Sauder School of Business, The University of British Columbia, 2053 Main Mall, Vancouver, BC, Canada
Abstract. This paper critically evaluates the use of analysts’ forecasts in accounting-based valuation.
Speciﬁcally, I assess the usefulness and the limitation of analysts’ forecasts in predicting future earnings
and in explaining the market-to-book ratio, in light of a comprehensive set of 22 explicit information
items, including: economic rent proxies, conservative accounting proxies, earnings quality signals, tran-
sitory earnings proxies, industry characteristics, and risk and growth proxies. While analysts’ forecasts
capture 45–83% of the information from these sources depending on model speciﬁcations, they do not
appear to fully incorporate certain information items. In particular, proxies for conservative accounting
and transitory earnings are incrementally useful in predicting future earnings; proxies for economic rents,
conservative accounting, and risk are incrementally useful in explaining the market-to-book ratio. Col-
lectively, these results validate the use of analysts’ forecasts as a parsimonious proxy for forward-looking
information in accounting-based valuation and suggest how to improve on their use.
Keywords: accounting-based valuation, earnings, analysts’ forecasts, market-to-book ratios
JEL Classiﬁcation: D4, G12, M4
The use of analysts’ earnings forecasts has become common in empirical studies
involving accounting-based valuation models.
The popularity of these forecasts is
understandable, because they capture forward-looking information in a form that
can be easily incorporated into various models.
However, the use of analysts’
forecasts is subject to at least two important limitations. First, most prior studies
treat these forecasts as a ‘‘black box’’, and prior research has provided little assur-
ance as to what information these forecasts are based on (Beaver, 1999). Second,
analysts’ forecasts might not fully incorporate certain types of public information.
Thus, simple reliance on these forecasts can yield noisy value estimates or biased
inferences. While prior studies recognize these limitations, they provide little evi-
dence on the eﬃcacy of analysts’ forecasts for valuation purposes.
This paper evaluates the use of analysts’ forecasts in accounting-based valuation
by investigating the extent to which these forecasts incorporate a comprehensive set
of 22 explicit information items that have been shown to be useful in valuation.
These explicit information items fall into six categories: economic rent proxies,
conservative accounting proxies, earnings quality signals, transitory earnings prox-
ies, industry characteristics, and risk and growth proxies (Lev and Thiagarajan,
1993; Fairﬁeld et al., 1996; Abarbanell and Bushee, 1997, 1998; Cheng, 2005). My
overall goal is to critically assess the usefulness and the limitation of analysts’
forecasts in valuation in light of these other information sources.
Review of Accounting Studies, 10, 5–31, 2005
Ó 2005 Springer Science+Business Media, Inc. Manufactured in The Netherlands.