320 B.E. Eckbo et al.
Ta b l e 1 4
(Continued)Country Study Sample
size
Sample
periodAR
(%)(d) Firm commitments:N= 1 ,064;ARfc= 1. 10 ∗
Japan Kang and Stulz (1996) 185 1985–1991 0. 51 ∗
Japan Cooney, Kato, and Schallheim (2003) 555 1974–1991 0. 72 ∗d
France Gajewski and Ginglinger (2002) 18 1986–1996 − 0. 33
Hong Kong Wu, Wang, and Yao (2005) 306 1989–1997 1. 93 ∗
In the panel headings,Nis the aggregate sample size across all studies in the panel, andARis sample-weighted
average market reaction. The superscript*indicates that theARis significantly different from zero at the 1%
level. The table is restricted to studies that (1) use daily stock return to measure the SEO announcement effect
ARand (2) report the announcement effect by individual f lotation method. For example, studies that pool
uninsured and standby rights in one sample are excluded. Some studies measureARover the two-day window
[− 1 ,0] while others use a three-day window [− 1 ,+1], and the table does not make a distinction between
these. Some studies also separate out industrials from utilities, and when they do, we report result averaged
across both issuer types.
aThe authors do not indicate whether their rights sample is standbys or uninsured rights. However, judging
from the sample frequency, and the information inSlovin, Sushka, and Lai (2000), we placed this study in the
standby category.
bIn 111 of the 200 cases, shareholder takeup is greater than 90%. For these cases, the announcement period
return is reported to be− 0 .33% and statistically insignificant.
cThe authors refer to these as “placings” or “bought deals” that increases shareholder dispersion.
dThe event day is the board meeting date.
for Greece and Norway, respectively, whileCronqvist and Nilsson (2005)report an
insignificant market reaction to uninsured rights offers in Sweden.Slovin, Sushka, and
Lai (2000)reports a significantly negative market reaction to U.K. uninsured rights
offers, whileGajewski and Ginglinger (2002)report a significantly negative market
reaction in France also. The sample-weighted cross-country average is however a non-
negative and statistically insignificantARur= 0 .70%.
It should be noted that the cross-country average may hide important country-specific
institutional effects, which often motivates a study of foreign issues. Thus, although
Slovin, Sushka, and Lai (2000)report results for a relatively small sample (20) of unin-
sured rights, the significantly negative market reaction may emanate from economically
important unique institutional characteristics of the London Stock Exchange. A similar
argument goes for the negative effect for the 57 uninsured rights offers in France studied
byGajewski and Ginglinger (2002). We return to this issue below.
Second, Panel (b) ofTable 14shows that standby offering are met with a positive
market reaction in Japan (Kang and Stulz, 1996), a neutral market reaction in Nor-
way and Sweden (Bøhren, Eckbo, and Michalsen, 1997; Cronqvist and Nilsson, 2005),
and a negative market reaction in the U.K. (Burton, Lonie, and Power, 1999; Slovin,