The Attributes of a Costly Recall: Evidence
from the Automotive Industry
NICHOLAS G. RUPP
Department of Economics, East Carolina University, Greenville, NC 27858-4353, USA
Abstract. While researchers have extensively documented the equity response to product
recalls and subsequent shareholder losses, less attention in the literature has been given to
examining the damaging recall attributes. Using 1973–1998 automotive safety recall data, this
study identiﬁes the kinds of recalls that cause signiﬁcant shareholder losses. After constructing
an equally-weighted automotive market index to control for industry eﬀects and adjusting the
abnormal returns to account for the degree of surprise in the recall announcement, the study
estimates both percentage and real dollar abnormal returns. We ﬁnd that the indirect costs of
automotive recalls are likely larger than the direct costs.
Key words: abnormal returns, automobile defects, NHTSA, recall attributes
Product recalls have been extensively examined in the literature with
researchers ﬁnding that such actions are typically damaging for shareholder
While the equity response to product recalls has attracted consid-
erable attention, the recall attributes that cause these signiﬁcant share-
holder losses has received little attention. For example, Barber and
Darrough (1996) examine the market reaction over diﬀerent time periods
and whether this reaction diﬀers between Japanese and American compa-
nies. Hoﬀer et al. (1987) estimate the abnormal returns of severe safety
recalls. Jarrell and Peltzman (1985) consider only large automotive safety
recalls appearing in the Wall Street Journal (WSJ). Rupp (2001) compares
the equity responses of government-initiated and ﬁrm-initiated automotive
recalls. Recall attributes are potentially important given that Hoﬀer et al.
(1994) ﬁnd vehicle age, defect severity, and nationality of manufacturer all
Jarrell and Peltzman (1985), Hoﬀer et al. (1987) and Barber and Darrough (1996) all ﬁnd
signiﬁcant shareholder losses surrounding automotive recall announcements. Likewise, Pruitt
and Peterson (1986) and Davidson and Worrell (1992) have similar ﬁndings for non-
automotive recalls. Not every study reaches the same conclusion, since Hoﬀer et al. (1988)
revisit the Jarrell and Peltzman (1985) data and ﬁnd insigniﬁcant shareholder returns.
Review of Industrial Organization 25: 21–44, 2004.
Ó 2004 Kluwer Academic Publishers. Printed in the Netherlands.