Review of Quantitative Finance and Accounting, 10 (1998): 39–58
© 1998 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.
The Nature and Persistence of Initial Public Offering
Aftermarket Returns Predictability
DOUGLAS A. HENSLER
Engineering Management Program, College of Engineering and Applied Science, Campus Box 435,
University of Colorado at Boulder, Boulder, CO 80309-0435
Abstract. This paper examines the predictability of monthly aftermarket returns of initial public offerings
during the ﬁrst six years of trading. Predictability is tested under the null hypothesis of random walk using a
Markov chain analysis. The evidence shows that excess returns of IPOs (adjusted for the return on the equally
weighted NASDAQ index) demonstrate non-random walk behavior through the ﬁrst ﬁve years of trading and
random walk behavior in the sixth year. This is accompanied by predictability of monthly excess returns
conditioned on the two previous months’ excess returns. A trading strategy is offered to capitalize on the
predictability patterns. Implementing the trading strategy is not possible due to institutional barriers, providing
additional explanation for why IPOs do not reach their intrinsic values for extended periods of time.
Key words: Initial public offerings, returns predictability, Markov chain
In contrast to the signiﬁcant returns potential of investing in initial public offerings (IPOs)
at the offering price and selling them within three months, a long-term investment in IPOs
is a wealth “hazard” with short-horizon returns predictability. This paper shows that
monthly excess returns of IPOs are predictable conditioned on the previous two months’
excess returns and the nature of the predictability changes over time. A trading strategy
exists which can potentially capitalize on the predictability patterns of monthly IPO
excess returns and arbitrage away aftermarket pricing deviations from intrinsic values.
However, institutional barriers prevent the use of the strategy resulting in the pattern of
prolonged negative excess returns in the aftermarket.
Ritter (1991) exposes the hazards of investing in initial public offerings in the after-
market using a number of benchmarks. Ritter reports that IPOs underperform in the
aftermarket over the three year period from the ﬁrst trading day.
For example, when using
matched seasoned stocks to adjust monthly aftermarket returns of IPOs, Ritter ﬁnds that
the average cumulative abnormal return continuously decreases after an initial trading
period of approximately three months. Loughran (1993) and Loughran and Ritter (1995)
extend the period of study to six years and ﬁve years, respectively, of aftermarket trading
and show that there is a continual degradation of cumulative abnormal returns in the
aftermarket until after the ﬁfth year of trading when it appears to cease.
The persistent continuous degradation of cumulative returns raises questions about
market efﬁciency and about the predictability of short-horizon returns in the aftermarket.
@ats-ss3/data11/kluwer/journals/requ/v10n1art3 COMPOSED: 11/04/97 2:12 pm. PG.POS. 1 SESSION: 10