Review of Quantitative Finance and Accounting, 13 (1999): 261±276
# 1999 Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.
The Time-Series Behavior of IPO Betas
JOHN D. NEILL
Argyros School of Business and Economics, Chapman University, Orange, CA 92866
STEVEN B. PERFECT
Sonat Marketing Company LP, Four Greenway Plaza, Houston, TX 77046
KENNETH W. WILES
Lloyd & Company, Renaissance Plaza, 230 N. Elm Street, Suite 1790, Greensboro, NC 27401
Abstract. We examine individual IPO betas and provide further evidence that the documented decline in IPO
betas results primarily from a seasoning or information effect and not from the delisting of high beta securities.
We employ stochastic coef®cient regression analysis which permits the estimation of individual IPO betas at all
points in time, and therefore avoids disadvantages associated with grouped cross-sectional beta estimates and
average individual time-series beta estimates. We ®nd that IPO ®rms with the lowest betas are more likely to
delist, and that individual IPO betas, on average, decline over time which provides support for the information
Key words: Initial public offerings, systematic risk, delisting hypothesis
The systematic risk of ®rms that have recently undertaken an initial public offering (IPO)
is dif®cult to estimate because there are no ®rm return data from which to estimate market
model parameters prior to the offering date, and because the risk estimates may contain a
time-dependent component. Despite these dif®culties, several studies estimate IPO betas.
Ibbotson (1975), Clarkson and Thompson (1990), Ritter (1991), Chan and Lakonishok
(1992), and Clarkson and Satterly (1997) estimate average cross-sectional IPO betas, and
conclude that these betas decrease, or decay, over time.
In this paper, we employ stochastic coef®cient regression (SCR) analysis to describe the
time-series behavior of ®rm-speci®c IPO betas over a 72-month post-offering period. SCR
permits the estimation of individual IPO betas at all points in time, and therefore avoids
disadvantages associated with grouped cross-sectional beta estimates and average
individual time-series beta estimates.
Our results reveal that IPO betas, and the patterns within which these betas evolve, vary
substantially across ®rms and across industries particularly during the ®rst 12 months of
trading. We then compare the SCR betas to betas estimated by the Value Line Investment
Survey and ordinary least squares (OLS) regressions. We ®nd that the average SCR betas
are approximately equal to the OLS betas, and that the Value Line betas appear to
converge with the SCR betas over time.