REVISITING BOOKBUILDING VERSUS AUCTION IPOS: A PERSPECTIVE OF INFORMATIONALLY EFFICIENT PRICES IN THE AFTERMARKET
Date | 01 September 2019 |
Published date | 01 September 2019 |
DOI | http://doi.org/10.1111/jfir.12181 |
The Journal of Financial Research Vol. XLII, No. 3 Pages 491–523 Fall 2019
DOI: 10.1111/jfir.12181
REVISITING BOOKBUILDING VERSUS AUCTION IPOS: A PERSPECTIVE
OF INFORMATIONALLY EFFICIENT PRICES IN THE AFTERMARKET
Hyeongsop Shim
Gachon Univeristy
Junyoup Lee
Ulsan National Institute of Science and Technology
Ha‐Chin Yi
Texas State University
Abstract
We compare price efficiency between auction and bookbuilding initial public
offerings (IPOs). Our empirical results fail to support the prevailing conjecture that
bookbuilding IPOs are more price efficient than auction IPOs. We find statistical
insignificance between two IPO samples or weak evidence for the opposite
hypothesis. We add to the evidence that auctions yield aftermarket price efficiency
equal to that of bookbuilding or are statistically indifferent when measured by
market microstructure and Center for Research in Security Prices (CRSP) data. We
also examine whether underlying price efficiency forces reflect the relative presence
of informed institutional and retail investors, aftermarket price support, and
divergence of investor expectations.
JEL Classification: G14, G24, G32
I. Introduction
One of the interesting questions asked in recent initial public offering (IPO) studies is
why the bookbuilding method dominates other selling methods such as auctions. This
dominance is puzzling given the abuses associated with bookbuilding IPOs, such as
spinning and laddering (Hao, 2007; Jones and Yeoman, 2008; Liu and Ritter, 2010).
1
Wilhelm (2005) states, “If bookbuilding is so obviously flawed, anyone who believes
in the power of markets to wring out inefficiency must ask why this practice has not
only survived for over two centuries, but has in fact gained considerable market share
Junyoup Lee appreciates the support by the “Human Resources Program in Energy Technology”of the
Korea Institute of Energy Technology Evaluation and Planning (KETEP), granted financial resource from the
Ministry of Trade, Industry & Energy, Republic of Korea (No. 20184010201680). Ha‐Chin Yi would like to
thank the McCoy College of Business Foundation for financial support.
1
Regulators, the popular press, and many researchers have noted the abuses of bookbuilding IPOs in the
pricing and allocation of shares. For example, the recent class‐action lawsuit against Facebook alleged that the
firm and the lead underwriters (Morgan Stanley and Goldman Sachs) withheld material information from retail
investors while sharing that information with large institutional clients.
491
© 2019 The Southern Finance Association and the Southwestern Finance Association
worldwide in recent years”(p. 55). Chemmanur and Liu (2019) suggest a theoretical
model to present different properties of two methods, and Jones and Yeoman (2008)
highlight the potential of auctions. Despite the scandals and abuses related to
bookbuilding, the auction method has gained only modest popularity as an alternative.
Only 23 IPOs in the United States had adopted the auction method as of the
end of 2013, all of which were conducted by WR Hambrecht + Co. (hereafter WR
Hambrecht).
2
Most studies on firm choice between bookbuilding and auction IPOs have
used non‐U.S. data (e.g., see the France case studied by Degeorge, Derrien, and
Womack, 2007) because of the scarcity of auctionIPOsintheU.S.market.Thefew
U.S.‐focused studies tend to examine IPO valuation or allocation (e.g., Degeorge,
Derrien, and Womack, 2010). No study has investigated the two selling methods
from the perspective of price efficiency using a U.S. sample. We aim to fill this study
gap. We address the following research question: Is there any difference in price
efficiency level in the immediate aftermarket between bookbuilding and auc-
tion IPOs?
Price efficiency occurs when the intraday prices of a stock closely trace all
fundamental firm attributes, in that they precisely resemble a random walk of
prices. Lower price efficiency (i.e., a deviation from the fundamental value) implies
a noisy measure of the cost of capital, leading to poorly informed investment and
financing decisions (Harris, 2003). Price efficiency also suggests fast price
adjustments to new information. Informationally efficient securities prices are
especially important to IPO firms for two reasons: (1) in terms of follow‐up
financing for seasoned equity offers (SEOs) or mergers and acquisitions (M&As)
and (2) the resolution of the large information asymmetry between investors and
firm managers.
The IPO process is only the first step in a series of financing decisions taken
to fund growth, such as future M&A decisions. New public market entrants
presumably have greater asymmetric information than seasoned firms with prior
stock offerings. This is why regulators use a quiet period to mitigate the
information friction between insiders and outsiders (Bradley, Jordan, and Ritter,
2003) and why insiders agree to the use of lock‐up periods (Bradley et al., 2001).
Efficient stock prices reflect a firm’s fundamentals more accurately during the early
stage of the IPO process, allowing external investors, who are exposed to
potentially severe information uncertainty in the IPO market, to make informed
investment decisions.
3
2
More than 23 auctions were processed through WR Hambrecht, but several were not genuine auctions, as
we describe in the Data Description section.
3
We focus on price efficiency in the spirit of Subrahmanyam and Titman (2001), who demonstrate the
importance of an informationally efficient price of equity for generating firm cash flow: “Feedback from
financial market prices to cash flows arises when a firm’s nonfinancial stakeholders, for example, its customers,
employees, and suppliers, make decisions that are contingent on the information revealed by the price”(p. 2389).
492 The Journal of Financial Research
Thus, we examine the underlying forces that entrepreneurs consider when
choosing between the auction and bookbuilding approach as a way to go public and
investigate their viability in terms of aftermarket informational price efficiency. We
use three metrics as proxies for informational price efficiency. First, we use
autocorrelation, the variance ratio, the standard deviation of pricing error (SDPE)
from Boehmer and Kelley (2009); short‐term volatility (STVOL)fromO’Hara and Ye
(2011); and price delay (PD) from Hou and Moskowitz (2005). Second, we use
unexplained trading volume from Garfinkel (2009) and spread data to measure
divergence of opinion. Divergence of opinion is defined as expectations that differ
from those of heterogeneous investors. A larger divergence of opinion reflects a
greater information asymmetry, contributing to a lower price efficiency. Third, we
use the aftermarket support of institutional investors, market makers, and analyst
coverage.
Based on a thorough literature review, we develop and test the null
hypothesis that bookbuilt IPOs yield price efficiency higher than that yielded by
auctioned IPOs. Using the price efficiency measures described previously, we
compare price efficiency levels between bookbuilt and auctioned IPOs between 1999
and 2013. We examine the efficiency of stocks in the nine‐monthperiodaftertheIPO
date using rigorous methods based on price efficiency studies. This nine‐month
period corresponds to IPO lock‐up expiration dates, when stocks presumably reach
their equilibrium stage.
We are surprised to find that our empirical results fail to support the null
hypothesis. Bookbuilt IPOs do not yield superior aftermarket price efficiency. Instead,
we find either no difference or weak evidence of the opposite finding: we find that
auction IPOs have higher price efficiency according to some of the measurements used,
whereas the IPO types tend to show statistically insignificant differences when other
efficiency measurements are used. Our findings cast doubt on the merits of the
bookbuilding approach and support the notion that the auction approach can be a viable
complement in the IPO market.
Our study complements the growing debate regarding different IPO selling
mechanisms by examining the market‐micro aspects of the offering. Earlier studies
provide evidence that the bookbuilding method has become dominant around the world
and argue that its pricing and allocation mechanisms are superior. However, only
indirect evidence is available regarding whether the bookbuilding method leads to
more efficient stock prices. We build on these findings by directly investigating the
relation between IPO pricing techniques and the informational efficiency of actual
transaction prices.
Our study is similar to Degeorge, Derrien, and Womack (2010), although they
examine the bidding process of auctioned IPOs and do not examine price efficiency.
Using 19 auctioned IPOs in the U.S. market, the authors argue that the issuer can draw
valuable pricing information from investors’bids. Our study complements their work
by showing that the aftermarket pricing efficiency of auctioned deals is not inferior to
that of bookbuilt deals.
493Bookbuilding versus Auction IPOs
To continue reading
Request your trial