326 B.E. Eckbo et al.
average market reaction to SEOs. Convertibles are a hybrid between debt and equity,
and a convertible debt offering may be viewed by the market as a delayed equity is-
sue (Stein, 1992). The uncertainty hypothesis is also supported by the finding that the
market reaction to equity issues by regulated utilities is much smaller (though still sig-
nificant) than the average market reaction to industrial issuers. The regulatory process
required for a utility to issue equity reduces the issuer’s discretion to time the issue to
periods where the market is overvaluing the stock.
The evidence on the effect of issue size on the market reaction is mixed. WhileJung,
Kim, and Stulz (1996)find no relationship to issue size,Masulis and Korwar (1986),
Korajczyk, Lucas, and McDonald (1990), andBayless and Chaplinsky (1996)find a
significantly negative relation between the announcement-induced abnormal return and
the size of the offer.
As is evident from equation(1)in Section4.2, the firm’s incentive to issue is greater
the greater the investment project’s NPV (b). Ifbis sufficiently large relative to the
value of assets in placea, then the firm will issue even it the shares are undervalued by
the market. Ifbis sufficiently large relative toafor all firms, there is no adverse selec-
tion (pooling equilibrium) and no adverse market reaction to the issue announcement.
However, in a separating equilibrium (with adverse selection), the market reaction will
be more favorable the greater the ratiob/a. Since the value ofbis unobservable to the
econometrician, studies have used the issuer’s B/M ratio or Tobin’sQas a proxy for
“growth”. The evidence is mixed: whileJung, Kim, and Stulz (1996)report a signifi-
cantly positive relation between the market reaction to equity announcements and B/M
ratios, several studies fail to find a significant relation (Barclay and Litzenberger, 1988;
Dierkens, 1991; Pilotte, 1992; Denis, 1994).
Shareholder takeup.InEckbo and Masulis (1992), shareholder takeupksimply acts
like financial slack. The greaterk, the smaller the issue sold to the market, and the lower
the scope for wealth transfer from outside investors. Thus, the greaterk, the smaller the
market reaction to the issue announcement. In the notation ofTable 16, the prediction is
ARfc<ARsr<0 andARur≈0. This prediction is supported by the evidence on U.S.
offerings:ARfc=− 2 .2%,ARsr=− 1 .3% (both significantly different from zero and
significantly different from each other), andARur=− 0 .6% (not significant). There is
also direct evidence that the takeup parameterkis highest in uninsured rights offerings,
lowest in firm commitments, with standbys in between. Thus, the evidence supports the
hypothesis that expected shareholder takeup affects the flotation method choice under
adverse selection.^39
Quality certification. In the vernacular ofEckbo and Masulis (1992), Bøhren, Eckbo,
and Michalsen (1997), andEckbo and Norli (2004), the significantly negative market
(^39) Bøhren, Eckbo, and Michalsen (1997)andCronqvist and Nilsson (2005)provide direct evidence onk.
Generally speaking, the value ofkdepends on shareholder (personal) wealth constraints and demand for
diversification by risk-averse investors. Moreover,kis likely to reflect the presence (if any) of individual
shareholders’ private benefits of control.