Discussion of ‘‘Separating Winners from Losers
among Low Book-to-Market Stocks using Financial
JOSEPH D. PIOTROSKI email@example.com
Graduate School of Business, University of Chicago, 5807 South Woodlawn Avenue, Chicago, Illinois, 60637
Abstract. The conference paper by Mohanram (2005) provides evidence on the success of contextual
ﬁnancial statement analysis in the low book-to-market (i.e., glamour) stock setting. The economic beneﬁts
of the strategy are concentrated in the identiﬁcation of glamour ﬁrms that will ultimately underperform the
market. In contrast to traditional accounting-based anomalies, Mohanram’s growth-based trading
strategy is stronger among large, heavily followed ﬁrms, suggesting that the mechanism behind the mis-
pricing of glamour ﬁrms is diﬀerent than the traditional information environment and dissemination
arguments found in other settings. Despite the robustness of the reported results, the strategy faces
implementation constraints due to (1) the relative costs associated with gathering industry-adjusted data
and (2) the frictions and costs associated with capitalizing on expected price declines over a long horizon.
Finally, the relative beneﬁts of contextual analysis need to be assessed against the predictive beneﬁts
accruing to traditional ﬁnancial statement analysis-based investment techniques.
Keywords: ﬁnancial statement analysis, growth stocks, accounting-based investment strategies
JEL Classfication: M41, G14
The conference paper by Mohanram (2005) examines the ability of contextual
ﬁnancial information to successfully diﬀerentiate glamour ﬁrms with strong future
return performance from glamour ﬁrms with weak future return performance.
Mohanram ﬁnds that glamour ﬁrms (i.e., low book-to-market ﬁrms) with a strong
likelihood of high continued growth outperform ﬁrms with weak growth attributes,
yielding a hedge return of approximately 20% annually. The predictive ability of his
GSCORE methodology is shown to be robust across numerous speciﬁcations;
however, the economic beneﬁts appear to be concentrated in the strategy’s ability to
predict subsequent glamour losers. Moreover, the intensity of GSCORE’s predictive
ability varies across subsets of the glamour portfolio, with the largest hedge returns
being generated for large ﬁrms, ﬁrms with analyst coverage, and ﬁrms with histori-
cally high growth rates. Combined with preceding work on the book-to-market eﬀect
(e.g., Piotroski, 2000; Griﬃn and Lemmon, 2002; Ali et al., 2003), the results shed
new light on the source of the underperformance of the low book-to-market portfolio.
The goals of this discussion are fourfold: ﬁrst, to outline the contribution of the
study in the context of extant research on the book-to-market eﬀect and contextual
ﬁnancial statement analysis; second, to interpret the reported results and discuss
several important questions raised by Mohanram’s analysis; third, to highlight
several of the research design choices made in the conference paper, and discuss their
Review of Accounting Studies, 10, 171–184, 2005
Ó 2005 Springer Science+Business Media, Inc., Manufactured in The Netherlands.