Ch. 6: Security Offerings 329
reaction to the news of the offering price) on offer-specific characteristics, including the
discount and the flotation method. The estimated coefficient on the discount is insignifi-
cantly different from zero whether the issuer is an industrial firm or a public utility. The
lack of a significant impact of the discount holds whether or not they account for sub-
scription precommitments in uninsured rights. This is important because greater levels
of subscription precommitments lower the risk of rights offer failure, thus reducing the
signaling effect of the discount itself. Overall, they find no support for the proposition
that the rights offer discount signals information (positive or negative) to the market
about the true value of the issuer.
Second, using Norwegian standbys and uninsured rights offerings,Bøhren, Eckbo,
and Michalsen (1997)also examine the information content of the rights offer discounts.
In contrast to rights offerings in the U.S., Norwegian issuers are required to set the rights
offer price a minimum of three weeks prior to the beginning of the rights offer period.
With a minimum rights offer period in Norway of two weeks, this means that the issuer
(and standby underwriter) must forecast the issuer’s secondary market price at least five
weeks ahead when determining the optimal offer price. The longer prediction period
probably increases the risk of offering failure relative to the U.S., making the Norwegian
rights offers a relatively powerful laboratory for examining signaling effects. They fail
to find a statistically significant effect of the offer price discount on the market reaction
to rights offer announcements.
Third, with a sample of U.S. utility standby rights offerings,Singh (1997)report that
abnormal stock returns over the “rights settlement period” (i.e., the period from the day
before the offer price release day and the following six days) are positively correlated
with the offering price discount.^41 Since his sample includes fully guaranteed rights
offerings only, there are no failure costs, so the signaling argument of Heinkel–Schwartz
does not apply.
Finally, focusing specifically on dividend implications of rights issues,Bigelli (1998)
reports a dividend-yield increase in more than 80% of his sample of Italian rights offers.
He finds that the average market reaction to rights offer announcements is positive, and
positively related to the subscription price discount. This is inconsistent withHeinkel
and Schwartz (1986)but consistent with separating equilibria in which unanticipated
dividend increases have information content. Further research is needed to establish
whether dividend increases associated with rights issues have information content also
in other issue markets.
(^41) Singh (1997)reports that there are on average 30 trading days between the first public announcement of the
standby and the “price release date” (the date on which the market first learns of the actual subscription price).
The price release date typically coincides with the date of the price amendment of the offering prospectus,
which is also typically the start of the offering.