338 B.E. Eckbo et al.
are disappointed when firms convey their post-issue earnings. That is to say, post-issue
earnings announcement on average should be associated with negative stock price reac-
tions. However, their results show no evidence of abnormal stock price reactions to the
earnings announcements.
Kang, Kim, and Stulz (1999)tests the overconfidence hypothesis using data on
Japanese public and private equity offerings. The non-negative announcement period
abnormal return to Japanese equity offerings supports the view that equity offerings
are regarded as good news in Japan. Nonetheless, they document post-issue negative
long run abnormal performance. Taken at face value, this is evidence goes against the
overconfidence hypothesis, but is consistent with investment based theories of equity
issuance.
5.2.2. Timing the market
Baker and Wurgler (2000)document that the proportion of equity in total new issues,
termed “the equity share”, is negatively correlated with future aggregate equity market
returns. For example, when the equity share was in its top historical quartile, the av-
erage market return in the following year was−6%. This could suggest that managers
are able to time the market component of their company’s returns. However,Baker,
Ruback, and Wurgler (2007)is cautious about this interpretation. They suggest that: “A
more plausible explanation is that broad waves of investor sentiment lead many firms
to be mispriced in the same direction at the same time. Then, theaveragefinancing
decision will contain information about theaverage(i.e., market level) mispricing, even
though individual managers are perceiving and responding only to theirownfirm’s
mispricing”.
Butler, Grullon, and Weston (2005a)question that timing ability or investor senti-
ment explain the predictive power of the equity share. They suggest that the apparent
ability to time the market can be understood as a form of aggregate “pseudo market
timing”. They point out that on an ex-post basis equity share value tends to be high
around market peaks and low around market troughs. Thus, it is the tendency to issue
equity when prices are high that leads to a spurious relationship between equity share
and future stock returns when measured ex post. They go on to argue that if equity tends
to be issued when current prices are high, then equity issuance activity should go down
during unexpected market declines—making pre-shock equity issuance look relatively
high and post-shock equity issuance look relatively low. Thus, aggregate pseudo market
timing should be most pronounced around market shocks. This prediction is supported
by evidence that the predictive ability of the equity share is driven by the Great Depres-
sion (1920–1931) and the 1973–1974 Oil Crisis.
The main point inButler, Grullon, and Weston (2005a)is that pseudo market timing
can appear as real timing ability in small samples.Baker, Taliaferro, and Wurgler (2004)
show that this problem extends to all time-series predictive regressions based on man-
agerial decision variables. Moreover, it is a special case of the small sample bias studied