of a CB and to a significant positive effect for the announcement of a WB. These results are
contrary to results for the United States. We attribute these remarkable results to the fact that firms
"package" these announcements with other (good) firm specific news. Using regression analysis we
also investigate whether shareholders react differently to announcements of CBs or WBs. In these
regressions issue specific characteristics are taken into account. This analysis shows that the market
reacts more positively to announcements of WBs than to announcements of CBs and that this
difference can not be explained by differences in issue specific characteristics. The more positive
effect of WB announcements is in line with Billingsley et. al. (1990) who also find that sharehold-
ers react more positively to announcements of WBs in relation to CBs.
The remainder of this paper is structured as follows. In section 2 the event study analysis
is presented. Section 3 investigates the different announcement effects of WBs and CBs using a
regression analysis. Finally, in section 4 a summary and conclusions are presented.
2. Event study analysis
Data
We use the announcements of CBs and WBs that were issued by Dutch companies from January
1976 to December 1994 and which have been listed on the Amsterdam Stock Exchange (ASE).
During this period a total of 59 CBs and 22 WBs were issued. From this sample we eliminated 11
CBs and 3 WBs because the announcement of these issues was accompanied by the announcement
of the issuance of shares or warrants (for cash). One CB was eliminated because not enough data
in the pre-event period were available. Therefore we are able to use 47 announcements of CBs and
19 announcements of WBs. The announcement date is the first date at which the announcement
appears in the Dutch daily financial newspaper Het Financieele Dagblad. Stock price data are
derived from Datastream and Het Financieele Dagblad.
Event study methodology
In order to test whether the announcement effect of WBs differs from the effect of CBs we use a
standard event study methodology as described in Brown and Warner (1985). Excess returns are
measured using an Ordinary Least Squares market model regression:
au ( r1,. - »<U - ⅛m ∙ (1)
where Ai,t is the excess return for firm i at day t and where Ri,t denotes the return on security i at