Club Deals in Leveraged Buyouts (2010)

Journal of Financial Economics 98 (2010) 214–240
Contents lists available at ScienceDirect
Journal of Financial Economics

journal homepage: www.elsevier.com/locate/jfec
Club deals in leveraged buyouts$

Micah S. Officer a,, Oguzhan Ozbas b, Berk A. Sensoy c
a
College of Business Administration, Loyola Marymount University, Los Angeles, CA 90045, USA
b
Marshall School of Business, University of Southern California, Los Angeles, CA 90089, USA
c
Fisher College of Business, The Ohio State University, Columbus, OH 43210, USA
a r t i c l e i n f o abstract
Article history: We analyze the pricing and characteristics of club deal leveraged buyouts (LBOs)—those
Received 29 October 2008 in which two or more private equity partnerships jointly conduct an LBO. Using a
Received in revised form comprehensive sample of completed LBOs of U.S. publicly traded targets conducted by
26 August 2009
prominent private equity firms, we find that target shareholders receive approximately
Accepted 8 December 2009
10% less of pre-bid firm equity value, or roughly 40% lower premiums, in club deals
Available online 1 June 2010
compared to sole-sponsored LBOs. This result is concentrated before 2006 and in target
JEL classification: firms with low institutional ownership. These results are robust to controls for target
G34 and deal characteristics, including size, Q, measures of risk, and time and industry fixed
G38
effects. We find little support for benign motivations for club deals based on capital
K21
constraints, diversification motives, or the ability of clubs to obtain favorable debt
amounts or prices, but it is possible that the lower pricing of club deals is an inadvertent
Keywords:
Club deals byproduct of an unobserved benign motivation for club formation.
Leveraged buyouts & 2010 Elsevier B.V. All rights reserved.
Private equity firms
1. Introduction rose from approximately $30 billion in 2001 to over $450

billion in 2007 (Thomson Financial M&A Review, 2007).
Following the leveraged buyout (LBO) boom of the Kaplan and Strömberg (2009) report that total equity
1980s and the relatively quiet 1990s, LBO activity again capital commitments to U.S. private equity funds reached
boomed in the 2001–2007 period before collapsing in $228.0 billion in 2007, or 1.57% of total U.S. stock market
2008. Total LBO deal volume in the United States alone capitalization.1
This sharp increase in private equity activity has
reignited the debate among economists and policymakers
$
We thank Ola Bengtsson, Harry DeAngelo, Stu Gillan, Mike Hertzel, regarding the efficiency and welfare implications of LBOs.
Yael Hochberg, Alexander Ljungqvist, Rich Mathews, Clara Raposo, Mike Proponents of the private equity/LBO model stress the
Stegemoller, Per Strömberg, Mark Westerfield, two anonymous referees,
seminar participants at Chapman University, Cornerstone Research, CRA
potential for creating value in target firms through
International, Duke University, Fordham University, ISCTE Business operating improvements, reduced agency costs of free
School (Lisbon), Loyola Marymount University, Ohio State University,
the Universities of California Riverside, North Carolina, Rochester, and
1
Southern California, Vanderbilt University, and Washington University We use the terms LBO and private equity sponsored deal
in St Louis, and conference participants at the 2009 EFA meetings and interchangeably, as do Kaplan and Strömberg (2009). Some authors also
the 2008 Frontiers of Finance conference for helpful comments, and refer to these as ‘‘going private transactions’’ or management buyouts
Derek Horstmeyer for able research assistance. This research began (MBOs). In this paper we focus on buyouts of publicly traded targets in
while Officer and Sensoy were on the faculty at the Marshall School of deals sponsored by prominent private equity firms. These deals almost
Business (USC). universally involve large amounts of leverage and also frequently
Corresponding author. involve the target’s pre-deal executive management team. Therefore,
E-mail addresses: micah.officer@lmu.edu (M.S. Officer), we consider all these terms to be practically identical in describing the
ozbas@usc.edu (O. Ozbas), bsensoy@fisher.osu.edu (B.A. Sensoy). types of transactions that we are interested in.
0304-405X/$ - see front matter & 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfineco.2010.05.007
M.S. Officer et al. / Journal of Financial Economics 98 (2010) 214–240 215
cash flow, and stronger managerial incentives (Jensen, induce funds to syndicate sufficiently large or risky deals.
1986). Skeptics suggest that the large profits earned by Club deals may also be motivated by a desire to certify
some private equity partnerships may be partly due to deal quality to debt financiers. LBOs are highly levered,
expropriation of target shareholders and stakeholders, or and it may be easier to acquire debt financing in sufficient
private equity investors, rather than bona fide efficiency quantity and on favorable terms if multiple private equity
improvements (Perry and Williams, 1994; Phalippou, firms attach their names and reputations to a deal.
2009). Other criticisms of LBOs in the 1980s and the In this paper we provide evidence on these explanations
2001–2007 period are that private equity firm profits stem for, and concerns about, club deals in leveraged buyouts by
in part from taking advantage of the tax deductibility of examining the pricing and characteristics of club deals
interest payments and/or inefficient bond markets in relative to sole-sponsored LBOs and to other merger and
which credit spreads are occasionally excessively low. acquisition (M&A) transactions. In doing so, we add to the
Recently, particular criticism has been directed at the evidence on the differential pricing and characteristics of
so-called club deals, in which two or more private equity acquisitions conducted by private equity firms compared
firms jointly sponsor an LBO. Clearly, one concern about to those conducted by other types of acquirers (Bargeron,
club deals is that private equity partnerships may be Schlingemann, Stulz, and Zutter, 2008). Concerns about
colluding to depress prices by limiting the number of club deals in the press, and elsewhere, are naturally
competing bidders in an auction for a takeover target, and focused on prominent (or large) private equity firms,
thereby may be shortchanging passive, dispersed share- because minor private equity firms are less likely to have
holders of target publicly traded corporations. This the market power to meaningfully reduce competition and
concern has strong grounding in the auction literature, therefore prices (inadvertently or otherwise) by forming
in which it is well-recognized that bidder collusion may clubs. Accordingly, we conduct our analysis using a
depress sale prices (e.g., Graham and Marshall, 1989; comprehensive sample of completed LBOs of U.S. publicly
Marquez and Singh, 2009), and in the regulatory econom- traded targets conducted by prominent private equity
ics literature (e.g., Cramton and Schwartz, 2000; Hen- firms between January 1984 and September 2007.2
dricks and Porter, 1992). These literatures stress that Our main finding is that target shareholders in club
collusion can reduce prices even in the absence of repeat deals receive significantly lower premiums than in sole-
play and even if collusion does not involve all potential sponsored LBOs and other merger and acquisition trans-
bidders for a target. Another, more innocuous, concern is actions. The differences are economically large: target
that the limited number of private equity firms interested shareholders receive approximately 10% less of pre-bid
in a given target may mean that clubbing, even when firm equity value, or roughly 40% lower premiums, in club
done in the absence of collusion and for benign reasons, deals compared to sole-sponsored LBOs.3 These results are
may lead to a reduction in bid competition and hence to robust to controls for target and deal characteristics,
lower premiums for target shareholders. including size, Q, measures of risk, and time and industry
In the absence of an instrumental variable for club fixed effects. The results are also robust to private equity
formation, it is, of course, not possible to conclusively firm fixed effects, indicating that, even though the
prove any causal effect of club deals on premiums paid to tendency to club together is concentrated amongst a
target shareholders (Bailey, 2007). Even if such an instru- few prominent private equity firms, our results are not
ment were available, it is also impossible to explicitly driven by the ability of a few private equity partnerships
distinguish between deliberate collusion to reduce prices to pay low prices regardless of whether they act alone or
and inadvertent reduction in bid competition resulting as part of a club. Our results also hold in regression and
from the limited number of private equity firms interested matching analyses using an LBO-only sample.
in any given target, because intent is unobservable. Nor is it This club deal discount is concentrated in club deals
possible to conclusively prove whether club deals result in announced prior to 2006; the financial media began
increased or decreased social welfare. expressing concerns about club deals at the end of 2005
That said, it is possible to explore the association and the U.S. Department of Justice started an informal
between club deals and premiums paid to target share- inquiry into the practice in 2006.4 In fact, the club deal
holders, how this association varies in the time-series and
cross-section, whether there exist differences in observable
2
characteristics between target firms in club deals and those We define prominent private equity firms as the largest 50 firms
by worldwide fundraising over 2002–2007, as well as the private equity
in other types of deals, and also to consider possible arms of major investment banks, and private equity firms that were
interpretations to advance our understanding of club deals. prominent in the 1980s but are now defunct. We describe the sample in
It is also possible to investigate whether any differences in detail in Section 3.
3
premiums paid in club deals relative to other types of deals Depending on the premium measure, time period, and estimation
technique, our estimates of the club deal discount range from 5% to 24%
are likely to be due to characteristics suggested by many of
of pre-bid target equity or firm value, with most estimates in the range
the classic theories of deal syndication. of 8–12%.
The classic theories of deal syndication suggest a 4
See Andrew Sorkin, New York Times, October 16, 2005, and ‘‘Private
number of benign reasons, unrelated to competition, for equity firms face anticompetitive probe,’’ The Wall Street Journal, October
why private equity firms may syndicate deals by forming 10, 2006. A competing explanation for this structural break is the
increased inflow of funds to the private equity industry around the same
clubs. Capital constraints may induce the formation of time, but we do not find supporting evidence for this. Even so, it is, of
clubs for sufficiently large transactions. Even if capital course, not possible to conclusively attribute any structural break in
constraints do not bind, diversification motives may time to a specific event with time-series evidence alone.
216 M.S. Officer et al. / Journal of Financial Economics 98 (2010) 214–240
discount virtually disappears in 2006 and 2007. In we find that the size difference is not statistically
addition, the intensity of club deals relative to overall significant in the pre-2006 period (in which we find that
LBO activity declines in 2006–2007 compared to 2004– club targets receive significantly lower premiums). More-
2005. over, only 23% of club deals are larger than the largest
In the cross-section, the discount associated with club sole-sponsored LBO conducted by any of the club
deals is less pronounced for target firms with higher members in the four-year window centered on the club
institutional ownership. This result is incremental to the deal announcement date.5 These facts cast doubt on the
positive relation between institutional ownership and notion that capital constraints are a primary motivation
premiums in the overall sample. Notably, there is no for the majority of club deals, especially before 2006,
similar incremental effect for sole-sponsored LBOs. These although it is possible that the degree of capital
results suggest that institutional investors are able to constraints faced by certain private equity firms may be
bargain effectively with clubs, and can at least partially dependent on deal characteristics that we cannot observe.
counteract the pricing consequences of any reduction in Regarding diversification motives, we find no evidence
competition (inadvertent or otherwise) associated with that club deal targets are systematically riskier or harder
clubs. The lack of a relation in sole-sponsored deals may to value as measured by beta, historical stock-return
reflect ineffective bargaining or that competitive pres- volatility, historical cash flow volatility, number of
sures are strong enough that the bargaining abilities of business segments (a measure of complexity), or analyst
institutional investors are redundant. forecast errors (a measure of information asymmetry)
We also examine the robustness of our conclusions to than targets of sole-sponsored LBOs are.
alternative measures of prices paid in private equity Moreover, we find that debt financing terms are
sponsored deals. Consistent with the premiums-based insignificantly better in club deals than they are in sole-
evidence, we find that deal values as multiples of sales or sponsored LBOs. If club deal target firms are riskier, we
earnings before interest, taxes, depreciation, and amorti- would expect to observe worse financing terms. In fact,
zation (EBITDA) (measured relative to multiples paid by we do not find statistically significant differences in
acquirers other than prominent private equity firms in financing terms, measured by loan spreads and maximum
time, size, and industry-matched acquisitions) are sig- debt-to-EBITDA covenants, between club and sole-spon-
nificantly lower in club deals than in sole-sponsored LBOs sored LBOs.6,7 These results also partially address the
prior to 2006. Our results using returns-based measures of concern that club targets receive lower prices because
premiums are also robust to defining premiums as the they are riskier in ways that are unobservable to us as
final bid price scaled by the pre-announcement target econometricians but observable to market participants.
share price. Overall, we interpret these results as providing only
To further investigate the notion that our findings of limited support for the notion that capital constraints,
lower target premiums in club deals are due to club diversification motives, or financing terms are first-order
formation reducing competition for target firms (inad- motivations for club deals.
vertently or otherwise), we examine the extent to which Given the potential legal and regulatory implications
the target firm receives competing bids after announce- of the possibility of collusion by prominent private equity
ment of the private equity takeover. Consistent with clubs firms, it is important to emphasize that the patterns we
reducing competition for target firms, we find that even show in the data are not necessarily due to collusion.
controlling for target size, there are significantly fewer Indeed, all of our evidence is potentially consistent with
post-announcement competing bids in successful club clubs forming for (possibly unobservable) benign reasons,
deals than in successful sole-sponsored LBOs. In addition, with inadvertent pricing consequences.
club targets – but not sole-sponsor targets – with high In addition, because there does not exist a credible
institutional ownership receive more post-announcement instrument in this setting, it is also possible that there are
competing bids. These findings suggest that soliciting unobservable factors that both drive club formation and
competing bids ex post – thereby forcing the club to raise can explain the level pricing result, the 2006 break, and
its price – is one mechanism by which institutions the cross-sectional relation between club premiums and
mitigate the discount otherwise associated with club target institutional ownership, even in the absence of any
deals. reduction in competition. However, it seems difficult to
We next investigate directly whether the benign explain our results with simple unobservable aspects of
rationales for club formation outlined above appear to deal risk or quality.
be important motivations for club formation. Of course,
we are only able to investigate benign motivations for
5
club formation that are related to observable target Our main results are qualitatively unaffected if we drop these club
deals (16 in total), which may have been too large to have been sole-
characteristics, and cannot rule in or out motivations
sponsored.
based on target characteristics that are unobservable to 6
Interestingly, we find that club deals involve significantly more
the econometrician. lenders than sole-sponsored LBOs, which may constrain the supply of
Regarding capital constraints, the evidence is mixed. debt financing for competing bids to the extent that lenders are asked to
We find that club deals are significantly larger on average be exclusive and commit to not lend to other potential bidders for the
same target.
than sole-sponsored LBOs, which is prima facie evidence 7
These results are consistent with Axelson, Jenkinson, Strömberg,
that capital constraints may be a motivation for club and Weisbach (2007), who find that debt-to-EBITDA ratios are insignif-
deals. Examining the distribution more closely, however, icantly higher in club deals.
Our results on debt terms, described above, suggest meaningful pricing consequences, and because they
that club deals are likely not riskier in unobservable ways account for the vast majority of dollar deal volume and
that would be reflected in debt terms. Moreover, Guo, fundraising in the private equity industry.
Hotchkiss, and Song’s (2009) results suggest that it is Our paper proceeds as follows. Section 2 presents
unlikely that our finding of a pre-2006 club deal discount relevant related literature. Section 3 describes our sample
can be explained by target firms in club deals being of selection procedure and provides summary statistics on
lower quality. Specifically, they find that the ex post the resulting sample. Section 4 analyzes the gains to
performance of leveraged buyouts in the 1990–2006 target shareholders using stock-return measures. Section
period is higher in club deals than in sole-sponsored 5 considers multiples-based pricing measures. Section 6
deals, both in terms of return on the price paid, and in provides further evidence on post-announcement compe-
terms of accounting performance, suggesting that club tition and on benign motivations for club deals. Section 7
deal target firms are of higher quality (i.e., offer higher concludes.
potential returns) than sole-sponsor targets are.8
Another possibility is that coordination costs amongst
club members lower a club’s willingness to pay. This 2. Related literature
effect could explain a club deal discount if capital
constraints bind and so make a club the only viable Our work adds to the literature on the pricing and
acquirer. However, as noted above, the data offer little characteristics of LBOs and going private transactions, and
support for the notion that capital constraints are a the possible sources of returns to private equity partner-
primary motivation for the majority of club deals, ships. DeAngelo, DeAngelo, and Rice (1984) examine a
especially before 2006, and our pricing results are robust sample of 81 going private transactions announced
to excluding the club deals that are the most likely to be between 1973 and 1980, and conclude that the average
driven by capital constraints. going private transaction does not result in expropriation
Our analysis does not speak to the social welfare or of wealth from dispersed target shareholders. Kaplan
efficiency implications of club deals. Club deals may be (1989a) investigates pre- to post-buyout changes in
economically efficient and socially beneficial if they operating performance and market value for management
redeploy assets to more productive uses that would buyouts conducted in the 1980s that subsequently went
remain undiscovered without the pooling of multiple public and finds evidence consistent with substantial
expert opinions, even if there is room for target share- efficiency improvements. Kaplan (1989b) finds that the
holders to share more of the surplus. Several studies of the tax deductibility of interest payments is an important
economic impact of private equity show substantial long- source of the wealth gain in buyouts. Kaplan and Stein
term benefits to society (Gurung and Lerner, 2008). (1993) examine the LBO wave of the 1980s, and find
Whether club deals are different from sole-sponsored evidence consistent with ‘‘overheating’’ of the LBO market
LBOs along this dimension is an important question for towards the end of the wave, when credit spreads may
future research. have been inefficiently low. Perry and Williams (1994)
Finally, we note that our results on the pricing of club find that target managers appear to manipulate earnings
deals differ from those of Boone and Mulherin (2009) and downward in the year preceding a management buyout
Bargeron, Schlingemann, Stulz, and Zutter (2008). Boone (using discretionary accruals), consistent with complicity
and Mulherin (2009) examine the pricing of club deals in in lowering the acquisition price, although DeAngelo
the 2003–2007 period, but, unlike us, find no evidence of a (1986) finds contradictory evidence. Despite these possi-
club deal discount in long-horizon acquisition premiums. ble sources of returns, Phalippou (2009) argues that
Similarly, Bargeron, Schlingemann, Stulz, and Zutter investors in private equity funds earn low risk-adjusted
(2008, footnote 4) report but do not tabulate that they returns.
find an insignificant club deal discount using long bid- Club deals in leveraged buyouts have interesting
period windows in a sample of 43 club deals conducted parallels to the syndication of venture capital (VC, or
over 1980–2005. Neither of these studies distinguishes early-stage private equity) investments. While a number
between prominent and other private equity firms, nor of papers examine syndication motives in VC, to our
accounts for the important structural break in 2006 that knowledge we are the first to provide direct evidence on
we find (the Bargeron et al. sample ends in 2005). Our the empirical relevance of possible syndication motives in
results combine with theirs to suggest that the pre-2006 the LBO market. In the VC setting, Lerner (1994) and Du
club deal discount we show is likely specific to large, (2008) provide evidence that syndication more commonly
prominent private equity firms. Given the nature of the occurs between VC firms of similar prominence, at least in
concerns about club deals, we believe our focus on first-round investments, which mirrors our findings.
prominent private equity firms is a distinguishing Brander, Amit, and Antweiler (2002), among others, find
strength of our work, both because these private equity support for diversification motives in VC syndication,
firms are the most likely to have the market power to whereas we find little support for those motives in our
constrain competition (inadvertently or otherwise) with setting.
With respect to deal pricing, Hochberg, Ljungqvist, and
Lu (2009) find that syndication networks among venture
8
Similarly, Brander, Amit, and Antweiler (2002) find that syndicated capital firms are associated with low valuations for
venture capital (VC) investments outperform non-syndicated ones. entrepreneurs seeking financing, i.e., they find that VC
syndication is associated with lower deal prices, but cannot there is no data source that we know of that provides a
distinguish between benign and other explanations. Their comprehensive list of prominent private equity firms that
findings and conclusions are consistent with ours. have ever been active in the U.S. public-to-private market.
Finally, our work is also part of a nascent literature in We create our own list in the following way.
‘‘forensic finance.’’ Ritter (2008) provides an excellent We begin with the names of the 50 largest private
discussion of this literature, emphasizing the important equity firms in the world as reported in the May 2007
role of statistical analysis in assessing whether patterns in issue of Private Equity International (PEI) magazine. PEI’s
financial market data are likely to have innocuous ranking is based on capital raised over the five-year period
explanations. between 2002 and 2007, and this list of prominent private
equity firms contains all the well-known financial spon-
sors of LBOs.10 To this list of prominent LBO sponsors we
3. Sample add the names of the in-house private equity units of the
investment banks Merrill Lynch, Morgan Stanley, and JP
Our sample of mergers and acquisitions comes from Morgan (including Chase Capital Partners). Despite their
Thomson Financial’s Securities Data Company (SDC) Plati- importance, these sponsors may not be on the PEI list
num database. We extract all transactions announced because they may use internal capital rather than relying
between January 1984 and September 2007 that are on external fundraising. We also add Forstmann Little and
completed by the end of November 2007, in which the HM Capital Partners (formerly Hicks, Muse, Tate, and
deal value is greater than $100 million, the deal is Furst) because they are historically prominent LBO
characterized by SDC as a merger, acquisition, or LBO, and sponsors that have been less active in recent fundraising.
the acquirer is seeking to own more than 50% of the target’s If a deal synopsis from SDC contains the name of one of
shares outstanding.9 To ensure that we obtain a sample of the prominent private equity firms described above, and
publicly traded target firms, we match this preliminary the role of that firm is an acquirer, then we code that deal
sample of 14,335 transactions to the Center for Research in as a private equity deal. Further, if that deal synopsis also
Security Prices (CRSP) database, and require that the target contains the name of another private equity firm (even a
firm have a stock price in the CRSP database at least once in minor one) as a joint acquirer, then the deal is coded as a
the seven days prior to the announcement date reported by ‘‘club’’ deal. While there is some discretion involved in
SDC. Furthermore, to ensure that all transactions in the data hand-coding transactions based on text synopses from
set are completed acquisitions, we require a delisting event SDC, we conduct the search in a conservative way that is
within one year (most are within one month) of the deal intended to ensure that all deals coded as sponsored by
effective date in SDC. This procedure leaves 4,031 com- private equity firms are, in fact, such transactions. The
pleted acquisitions of publicly traded target firms. resulting sample consists of 201 private equity deals of
Because the concerns about club deals primarily focus which 70 are club deals. The Appendix contains a list of
on large buyouts by prominent private equity firms, we the private equity deals in our sample, with year, target
determine which of these transactions are LBOs led by name, and acquirer name (truncated at two acquirer
prominent private equity firms. It is not obvious how to names for club deals) tabulated.11 Panel A contains the list
do this or how to determine whether any given LBO is a of club deals in our sample, and sole-sponsored private
club deal. While SDC contains an LBO flag and a ‘‘going equity deals are in Panel B. From this point onwards in
private’’ flag, it is unclear what criteria are used to set this paper, when we refer to private equity deals (sole-
these flags. We observe that the flags capture some sponsored or club) we are referring to deals involving
transactions that are not sponsored by private equity these prominent private equity partnerships.
firms, such as corporation-led buyouts and management Table 1 displays the distribution of our sample by year
buyouts that do not involve a private equity firm, and also and acquirer type. We consider four mutually exclusive
miss some transactions that are sponsored by private and exhaustive acquirer types: private equity clubs, sole
equity firms. In fact, the LBO flag provided by SDC misses private equity firms, private firms that are not prominent
about one in five of the deals we identify using the private equity firms (therefore, this category includes not
procedure outlined below. only private firms but also non-prominent private equity
In light of these difficulties, we code indicator variables firms), and public firms. The merger waves of the late
for whether a private equity firm is involved in the deal
and for whether multiple private equity firms are involved.
This coding is done by hand for a sample of acquisitions 10
Highlighting the importance of focusing on prominent private
that we identify using a text search for the names of
equity sponsors, PEI reports that the members of its top-50 list were
prominent private equity firms in the transaction synopses responsible for 75% of global LBO deal volume during 2002–2007.
for the sample of 4,031 deals from SDC. Unfortunately 11
We thank one of our anonymous referees and several colleagues
for pointing out several sample inaccuracies that resulted from the sole
use of SDC data for classification. We decided to classify both the 2007
9
Our main results are qualitatively unaffected if we instead impose CDW and 1999 Concentra Managed Care LBOs as sole-sponsored deals
a real deal value screen of $100 million in 2007 dollars. We use a because the mode of club formation in both deals was unusual (both
nominal cutoff of $100 million to be consistent with prior literature, were sole-sponsored deals at announcement). However, all our results
even though this results in more stringent criteria for inclusion in the are robust to reclassifying these deals as Club LBOs. Details of the sample
sample in 1984 than in 2000 or 2007 (because of the dramatic rise in the classifications that would not be replicable using SDC data (but which
level of stock market prices around 2000 and 2007). are reflected in the Appendix) are available from the authors by request.
Table 1
Sample distribution.This table contains the time-series distribution of a sample of acquisitions of publicly traded targets from SDC. The transactions are
classified into years based on announcement dates. We identify private equity (PE) deals by performing a text search for the names of prominent private
equity firms in the deal synopses provided by SDC and manually confirming the results. Prominent private equity firms are defined as the 50 largest
private equity firms in the world as reported in the May 2007 issue of Private Equity International (PEI) magazine (with five additions described in Section
3). If the synopsis from SDC contains the name of only one prominent private equity firm (and no other firms are identified as the acquirer), then the deal
is classified as a Sole PE deal. If the synopsis from SDC contains the name of at least one prominent private equity firm (and at least one other private
equity firm, prominent or not, is identified as the acquirer), then the deal is classified as a Club deal. In a few cases the SDC synopses are incorrect or
misleading, and we use merger proxy statements to classify these deals (see Section 3 for details). The remaining transactions in the sample are classified
by whether the acquirer is privately held (Private bidder deals) or publicly traded (Public bidder deals), with listing status determined by the bidder’s
inclusion in the CRSP database. Total deal value is the annual sum of deal value excluding assumed liabilities reported by SDC (in $m).
Year All deals Deals by prominent private equity firms Other deals
Club deals Sole PE deals Private bidder Public bidder
No. of Total deal No. of Total deal No. of Total deal No. of Total deal No. of Total deal
deals value deals value deals value deals value deals value
1984 33 17,359.1 0 0.0 2 882.2 13 5,809.7 18 10,667.2

1985 111 96,555.4 1 637.0 5 11,696.8 19 8,190.4 86 76,031.2
1986 131 79,560.9 2 1,381.0 8 12,387.4 41 25,642.2 80 40,150.3
1987 127 73,839.8 1 103.9 7 11,973.0 49 25,310.0 70 36,452.9
1988 135 127,241.3 1 538.9 10 34,013.0 74 49,630.0 50 43,059.4
1989 103 89,764.5 1 923.0 2 2,204.1 48 24,679.9 52 61,957.5
1990 49 41,629.3 0 0.0 3 2,500.3 17 7,644.1 29 31,484.9
1991 47 29,686.7 0 0.0 0 0.0 3 2,508.9 44 27,177.8
1992 54 24,431.9 0 0.0 0 0.0 3 998.8 51 23,433.1
1993 74 67,274.8 1 191.3 2 497.7 4 1,259.7 67 65,326.1
1994 130 88,392.1 0 0.0 1 2,243.4 25 17,825.6 104 68,323.1
1995 181 180,040.4 2 1,642.3 2 1,357.2 24 28,171.6 153 148,869.3
1996 220 247,262.4 1 112.8 5 2,334.2 31 33,570.2 183 211,245.2
1997 353 332,194.3 3 2,187.4 11 5,277.2 51 23,106.9 288 301,622.8
1998 351 960,167.2 5 2,691.2 3 1,352.8 64 43,783.0 279 912,340.2
1999 402 829,847.4 5 2,437.0 7 3,478.2 95 73,128.2 295 750,804.0
2000 330 857,332.8 2 613.7 3 1,783.2 83 85,100.9 242 769,835.0
2001 187 276,961.3 0 0.0 1 995.0 43 52,786.5 143 223,179.8
2002 112 141,999.3 3 2,302.5 2 741.7 28 8,521.7 79 130,433.4
2003 141 201,788.7 0 0.0 2 490.7 21 10,501.6 118 190,796.4
2004 163 356,586.9 5 12,850.6 6 8,000.6 22 51,857.5 130 283,878.2
2005 199 493,836.1 9 28,517.7 10 10,331.3 55 62,955.3 125 392,031.8
2006 240 633,778.8 19 124,309.1 19 48,388.2 61 57,105.7 141 403,975.8
2007 158 385,159.0 9 83,868.2 20 103,054.5 48 60,546.1 81 137,690.2
Total 4,031 6,632,690.4 70 265,307.6 131 265,982.7 922 760,634.5 2,908 5,340,765.6
1980s and late 1990s are evident in the table. Table 1 also and especially 2007 is actually a reversal of an increasing
shows that club deals are a relatively new phenomenon: trend towards clubbing in the years before that.
37 of the 70 take place in 2005–2007, and 63 of the 70 Panel A of Table 2 gives a frequency distribution of deals
take place in 1995–2007. by private equity firm. Kohlberg Kravis Roberts (KKR) leads
Interestingly, the intensity of club deals relative to the table with 26 deals (11 of which are club deals),
overall LBO activity declines slightly in 2006–2007 followed by Blackstone with 24 deals (12 of which are club
(especially 2007) compared to 2004–2005 (following the deals), and TPG with 21 deals (18 of which are club deals,
onset of media and government scrutiny of the practice). the vast majority of TPG’s LBO activity). Participation in
Specifically, there are 30 LBOs in our sample in 2004– club deals roughly mimics overall participation in LBO
2005, of which 14 (47%) are club deals. For 2006–2007, activity, although there are a few exceptions such as TPG.
there are 67 LBOs in our sample, of which 28 (42%) involve Panel B of Table 2 displays a participation matrix,
clubs of private equity firms. While one could argue that documenting the frequency with which private equity
this trend is mechanical, because the number of promi- firms partner with each other in club deals. TPG (the most
nent private equity firms is fixed while the number of common club participant in Panel A) partners four times
non-prominent firms likely increases over time, the trend each with the other three most prominent private equity
prior to 2004–2005 suggests otherwise. In 2002–2003, groups in our sample (Blackstone, KKR, and Goldman
there are seven LBOs in our sample, of which three (43%) Sachs). Panel B shows that the most prominent private
are club deals. Furthermore, in 2000–2001, there are six equity firms (who are also the subject of the litigation
LBOs in our sample, of which two (33%) are club deals. referred to in the introduction) are more likely to partner
This suggests that the drop in club deal intensity in 2006 with each other than with less prominent private equity
Table 2
LBOs completed by private equity firms: counts and cross-participation.Panel A contains the number of deals sponsored by the most active private equity
firms in our sample. Sole PE and Club deals are defined in the legend to Table 1. Panel B presents the number of times prominent private equity firms have
undertaken club deals with one another (as opposed to other, non-prominent private equity firms). Our sample contains club LBOs by 29 prominent
private equity firms of which Panel B lists 24. The remaining five prominent private equity firms, Berkshire Partners, Cerberus Capital Management,
Clayton, Dubilier & Rice, Fortress Group, and Sun Capital Partners, are not listed because although Cerberus has done two club deals and the others have
done one club deal each, none of them are with other prominent private equity firms. In Panel B, columns that do not contain additional information are
omitted to save space.
Panel A: Number of deals
Rank PEI 50 Name of private equity firm All deals Sole PE deals Club deals
1 2 Kohlberg Kravis Roberts 26 15 11

2 4 The Blackstone Group 24 12 12
3 5 TPG 21 3 18
4 3 Goldman Sachs Principal Investment Area 18 5 13
5 42 Welsh, Carson, Anderson & Stowe 14 9 5
6 12 Apollo Management 13 9 4
7 y Morgan Stanley 12 10 2
8 y Merrill Lynch 10 6 4
9 30 Thomas H. Lee Partners 10 4 6
10 31 Leonard Green & Partners 9 6 3
11 1 The Carlyle Group 9 5 4
12 y Forstmann Little 8 8 0
13 8 Bain Capital 8 2 6
14 y HM Capital Partners (formerly Hicks, Muse, Tate, and Furst) 7 6 1
15 32 Madison Dearborn Partners 7 3 4
16 14 Warburg Pincus 6 4 2
17 9 Providence Equity Partners 6 0 6
18 47 Clayton, Dubilier & Rice 5 4 1
19 25 Lehman Brothers Private Equity 5 4 1
20 19 Silver Lake Partners 5 2 3
21 16 Hellman & Friedman 5 1 4
22 y JP Morgan (including Chase Capital Partners) 5 1 4
23 27 Fortress Investment Group 4 3 1
24 34 Cerberus Capital Management 3 1 2
25 44 GTCR Golder Rauner 3 1 2
26 40 Berkshire Partners 2 1 1
27 33 Onex 2 1 1
28 28 Sun Capital Partners 2 1 1
29 39 TA Associates 2 1 1
30 29 BC Partners 1 1 0
31 21 EQT Partners 1 1 0
32 41 Pacific Equity Partners 1 1 0
33 6 Permira 1 0 1
Panel B: Cross-participation matrix
TPG The Kohlberg Goldman Providence Bain Thomas H. The JP Morgan Merrill GTCR
Blackstone Kravis Sachs Equity Capital Lee Carlyle Morgan Stanley Lynch Golder
Group Roberts Partners Partners Group Rauner
The Blackstone 4 y y y y y y y y y y y
Group
Kohlberg Kravis 4 3 y y y y y y y y y y
Roberts
Goldman Sachs 4 4 4 y y y y y y y y y
Providence 3 1 2 2 y y y y y y y y
Equity
Partners
Bain Capital 1 2 3 1 1 y y y y y y y
Thomas H. Lee 1 0 0 1 1 0 y y y y y y
Partners
The Carlyle 1 1 1 1 2 0 0 y y y y y
Group
JP Morgan 0 0 0 1 0 0 1 0 y y y y
Morgan Stanley 1 0 1 1 0 0 0 0 0 y y y
Merrill Lynch 0 0 1 0 0 1 1 0 0 0 y y
GTCR Golder 0 0 0 0 0 0 0 0 0 0 0 y
Rauner
Silver Lake 3 1 1 1 1 1 0 0 0 0 0 0
Partners
Table 2 (continued )
Panel B: Cross-participation matrix
TPG The Kohlberg Goldman Providence Bain Thomas H. The JP Morgan Merrill GTCR
Blackstone Kravis Sachs Equity Capital Lee Carlyle Morgan Stanley Lynch Golder
Group Roberts Partners Partners Group Rauner
Lehman 1 0 1 1 0 0 0 0 0 1 0 0
Brothers
Private Equity
Madison 1 0 0 0 1 0 1 0 0 0 1 0
Dearborn
Partners
Hellman & 2 1 1 0 0 0 0 0 0 0 0 0
Friedman
Warburg Pincus 1 0 0 1 0 0 1 0 1 0 0 0
Welsh, Carson, 0 1 1 0 0 0 0 0 0 0 0 1
Anderson &
Stowe
Permira 1 1 0 0 0 0 0 1 0 0 0 0
Leonard Green 2 0 0 0 0 0 0 0 0 0 0 0
& Partners
Apollo 0 0 0 0 0 0 0 0 1 0 0 0
Management
HM Capital 0 0 1 0 0 0 0 0 0 0 0 0
Partners
TA Associates 0 0 0 0 0 0 0 0 0 0 0 1
Onex 0 0 0 0 0 0 0 0 0 1 0 0
firms. This finding is consistent with the prima facie dividend distributions) from the compound return to the
concern about collusion lowering deal prices: the most target shares over the given period.12 We calculate all CARs
prominent private equity firms, with the market power to by summing market-model residuals (e~ it ), and estimate
effectively collude and, in all likelihood, the greatest market-model parameters (a~ i and b~ i ) for each target using
ability to compete against one another absent collusion, daily returns from trading day 379 to trading day 127
are the most likely to partner in LBO club deals. relative to the deal announcement date:
4. Returns to target shareholders e~ it ¼ rit ða~ i þ b~ i rmt Þ,

where rit is target i’s return on day t and rmt is the return to
In this section, we use several measures of the gains to
the CRSP value-weighted market index (including dividend
target shareholders in our sample to provide evidence on
distributions) on day t. In addition to the above-mentioned
the pricing of club deals relative to sole-sponsored LBOs
variables, we also calculate the cumulative abnormal
and relative to non-private equity transactions.
return to the target’s shares from trading day 1 to
trading day +1 (labeled CAR3). These return/premium
4.1. Variable construction
measures are all used in a variety of papers in the M&A
literature (see, for example, Schwert, 1996; Bargeron,
We collect daily return data for each of our targets Schlingemann, Stulz, and Zutter, 2008).
from CRSP. We use these data to construct the raw
returns, buy-and-hold abnormal returns (BHARs), and
cumulative abnormal returns (CARs) over the following
intervals for each target (all dates are trading days relative 4.2. Target return comparisons by bidder type
to the deal announcement date (0)): day 42 to day 1
(labeled runup); day 0 to day +126 or the delisting date, Table 3 displays means and medians of these return
whichever occurs first (labeled markup); and day 42 to measures classified by acquirer type. Our results are easily
day + 126 or the delisting date, whichever occurs first summarized. For all return measures except runup,
(labeled premium). Our choice of intervals follows shareholders of club deal targets earn significantly lower
Schwert (1996). We consider long-run CARs for consis-
tency with some of the existing literature, even though 12
Our results are robust to defining the buy-and-hold abnormal
there is a greater potential for pronounced differences return as the target’s compound return over the relevant period minus
between compounded and cumulated returns over longer the compound return to the matching Fama-French (5 5) size and
horizons, and even though BHARs are a better measure of book-to-market portfolio (style-adjusted buy-and-hold abnormal re-
turn) as well as defining bid premiums using SDC deal prices relative to
the wealth effects to target shareholders. target share prices 43 days prior to announcement (as in Officer, 2003,
All BHARs are calculated by subtracting the compound Table 3). Results using these alternative measures of abnormal returns
return to the CRSP value-weighted market index (including and premiums are omitted for brevity.
Table 3
Target percentage returns, by bidder type.This table contains averages and medians (in brackets) for various measures of LBO target returns (in percent)
in the sample of deals from SDC described in the legend to Table 1. Runup is the compound return measured over the ( 42, 1) period (defined as
trading days relative to the announcement date (0)), CR3 is the compound return measured over the ( 1, + 1) period, Markup is the compound return
measured over the (0, + 126) period (unless the target is delisted prior to trading day + 126), and Premium is the compound return measured over the
( 42, + 126) period. All buy-and-hold abnormal returns (BHAR) are the compound return to the target shares minus the compound return to the CRSP
value-weighted market index (including dividend distributions) over the same period. Cumulative abnormal returns (CAR) are the sum of the target’s
daily raw market-model residuals over the corresponding periods (with market-model parameters estimated over the ( 379, 127) period as long as
there are 100 returns on CRSP in that period). , , and indicate that the difference in target returns is significantly different from zero at the 1%, 5%,
and 10% levels, respectively.
Differences
Club Sole PE Private Public Club Sole PE Club Private Club Public Sole PE Sole PE Private Public
Private Public
Raw returns
Runup 8.90 12.60 15.23 14.25 3.70 6.33 5.35 2.64 1.65 0.99
[7.88] [10.48] [10.23] [10.48] [ 2.60] [ 2.35] [ 2.61] [0.25] [ 0.01] [ 0.25]
CR3 11.76 19.12 22.62 20.19 7.37 10.86 8.44 3.50 1.07 2.43
[9.70] [15.55] [18.13] [16.50] [ 5.85] [ 8.43] [ 6.80] [ 2.59] [ 0.95] [1.64]
Markup 14.21 21.91 26.48 26.61 7.70 12.27 12.40 4.57 4.70 0.13
[13.63] [20.65] [22.18] [22.58] [ 7.02] [ 8.55] [ 8.96] [ 1.53] [ 1.93] [ 0.40]
Premium 24.04 36.11 44.10 43.81 12.07 20.06 19.76 7.99 7.70 0.29
[22.97] [32.34] [36.72] [37.44] [ 9.37] [ 13.75] [ 14.47] [ 4.38] [ 5.10] [ 0.72]
Number observations 70 131 922 2,908
Buy-and-hold abnormal returns

Runup BHAR 6.40 9.24 12.76 11.61 2.84 6.35 5.21 3.51 2.37 1.14
[4.35] [6.25] [7.63] [7.83] [ 1.91] [ 3.28] [ 3.48] [ 1.37] [ 1.58] [ 0.21]
Markup BHAR 8.08 16.09 22.91 21.97 8.01 14.83 13.89 6.82 5.88 0.94
[7.44] [14.41] [18.04] [18.17] [ 6.97] [ 10.60] [ 10.74] [ 3.63] [ 3.76] [ 0.13]
Premium BHAR 15.32 26.82 37.99 36.46 11.49 22.67 21.14 11.17 9.64 1.53
[12.92] [23.13] [31.02] [30.05] [ 10.21] [ 18.10] [ 17.13] [ 7.89] [ 6.92] [0.97]
Cumulative abnormal returns

Runup CAR 7.18 8.45 12.22 9.63 1.27 5.04 2.45 3.78 1.18 2.59
[6.31] [6.14] [8.02] [7.41] [0.17] [ 1.72] [ 1.11] [ 1.89] [ 1.27] [0.61]
CAR3 11.45 18.26 22.28 19.54 6.81 10.83 8.09 4.02 1.28 2.74
[10.10] [15.76] [17.75] [16.23] [ 5.66] [ 7.66] [ 6.13] [ 2.0] [ 0.47] [1.53]
Markup CAR 9.64 14.76 21.40 18.01 5.12 11.76 8.38 6.64 3.25 3.39
[7.02] [14.94] [18.14] [15.55] [ 7.92] [ 11.12] [ 8.53] [ 3.20] [ 0.61] [2.59]
Premium CAR 16.81 23.21 33.62 27.64 6.39 16.81 10.83 10.42 4.43 5.98
[19.18] [23.89] [28.82] [24.89] [ 4.71] [ 9.64] [ 5.71] [ 4.93] [ 1.01] [3.92]
returns than shareholders of targets that are acquired by leakage or deal anticipation is not different in club deals
any other category of acquirer. This is true in both means than in other types of deals.13
and medians. The only exception is that for one measure, Of course, our main interest is a comparison between
cumulative abnormal returns (CARs) measured over the club deals and sole-sponsored private equity deals. Even if
premium period, the point estimate of the difference private equity deals generally are priced lower than deals
between sole-sponsored and club deals ( 6.4%) is not by publicly traded acquirers, skepticism of club deals may
statistically significant at conventional levels, with a p- be misplaced if pricing does not differ between the two
value of 0.13. The economic magnitudes of the differences types of private equity deals.
in target returns between club and sole-sponsor deals are Table 3 shows that, on average, shareholders of club
generally large: for example, the differences in the buy- deal targets earn a buy-and-hold abnormal return over
and-hold abnormal returns over the whole bid-period the premium period ( 42 to +126 trading days relative to
(buy-and-hold abnormal premium) range from 11 to 23 the deal announcement) that is a statistically and
percentage points, on average. Thus, target shareholders
gain considerably less from acquisitions by clubs of 13
The sample sizes differ slightly between cumulative and buy-and-
private equity firms than from acquisitions by other hold abnormal returns in Table 3 because the former relies on the
types of acquirers. The insignificant results for the runup estimation of market-model parameters (described in Section 4.1) while
measures suggest that pre-announcement information the latter does not.
economically significant 11.5 percentage points, or 43% 4.3. Differences in target firm characteristics
less than shareholders of targets involved in sole-
sponsored deals. Another, more striking, way to view It is possible that the reason clubs of private equity
the differences between club deals and sole-sponsored firms pay less to target shareholders than do other types
private equity deals is that target shareholders gain 75% of acquirers is that they seek to acquire different types of
greater abnormal premiums, on average, if the acquisition firms. We investigate this possibility by focusing on
is by a sole private equity firm rather than a club (26.8% observable target characteristics that the literature has
vs. 15.3%). Table 3 also shows that the result that target found to be important in explaining gains to target
shareholders gain less in club deals than in sole- shareholders, while acknowledging that we are unable
sponsored private equity deals is robust to different to observe differences in dimensions that are unobser-
return measurement periods and return definitions. vable to the econometrician. Table 4 compares target firm
Table 4
Target characteristics, by bidder type.This table contains averages and medians (in brackets) for various target characteristics. Numbers beneath medians
are the number of observations. Size is the target’s market capitalization 43 days prior to bid announcement (immediately before the beginning of the
runup period), measured in billions of dollars. Accounting data for the target firm are from Compustat for the fiscal year ending immediately prior to deal
announcement. Industry-adjusted Q is the target’s Q ratio (defined in Kaplan and Zingales, 1997) minus the median Q for all firms in the same two-digit
SIC code industry in the same year. Industry-adjusted EBITDA/Assets is the target’s EBITDA/Assets minus the average EBITDA/Assets for all firms in the
same two-digit SIC code industry in the same year. Debt/(Debt +Equity) is defined as the book value of the target’s total debt (Compustat annual data item
#9 plus Compustat annual data item #34) divided by the sum of the book value of the target’s total debt and the target’s market capitalization 43 days
prior to bid announcement. Institutional ownership is measured in the quarter prior to deal announcement, and is the fraction of the target’s shares
owned by institutions required to file a 13F statement and included in Thomson Financial’s 13F Holdings database. Prior 12-month return is the
compound return to the target’s stock over the one year immediately preceding the beginning of the runup period (i.e., ending on trading day 43
relative to the announcement date), in percent. Prior 12-month BHAR (buy-and-hold abnormal return) is Prior 12-month return minus the compound
return to the CRSP value-weighted market index (including dividend distributions) over the same period, in percent. Prior 12-month return volatility is
the standard deviation of the target’s daily percent returns for the one year immediately preceding the beginning of the runup period. Beta is the target’s
market-model beta estimated over the ( 379, 127) trading day relative to the announcement date period (using daily data) as long as there are 100
returns on CRSP in that period (otherwise Beta is set to missing). , , and indicate that the difference in averages or medians is significantly different
from zero at the 1%, 5%, and 10% levels, respectively.
Differences
Club Sole PE Private Public Club Sole PE Club Private Club Public Sole PE Sole PE Private Public
Private Public
Size ($ billions) 2.81 1.36 0.54 1.28 1.45 2.27 1.54 0.82 0.08 -0.74
[0.93] [0.53] [0.20] [0.26] [0.40] [0.73] [0.67] [0.33] [0.27] [ 0.06]
70 131 922 2,903
Industry-adjusted Q 0.06 0.01 0.00 0.44 0.05 0.06 0.38 0.01 0.43 0.44
[ 0.01] [ 0.12] [ 0.09] [0.01] [0.11] [0.08] [ 0.02] [ 0.03] [ 0.14] [ 0.10]
62 117 821 2,620
Industry-adjusted 0.04 0.05 0.03 0.02 0.01 0.02 0.03 0.03 0.04 0.01
EBITDA/assets [0.02] [0.03] [0.02] [0.01] [ 0.01] [ 0.00] [0.01] [0.01] [0.02] [0.01]
64 114 813 2,639
Debt/(debt +equity) 0.28 0.30 0.30 0.26 0.02 0.02 0.02 0.00 0.04 0.04
[0.28] [0.29] [0.27] [0.21] [ 0.02] [0.01] [0.07] [0.02] [0.08] [0.06]
68 124 847 2,712
Institutional ownership 0.65 0.57 0.43 0.43 0.08 0.22 0.22 0.14 0.14 0.00
[0.69] [0.56] [0.42] [0.41] [0.14] [0.28] [0.28] [0.14] [0.14] [0.00]
62 110 814 2,669
Prior 12-month return 12.73 14.10 13.41 21.94 1.37 0.69 9.21 0.69 7.84 8.53
[4.86] [11.80] [7.10] [11.68] [ 6.94] [ 2.24] [ 6.82] [4.70] [0.12] [ 4.58]
70 131 922 2,903
Prior 12-month BHAR 2.84 0.49 0.57 6.15 2.35 2.26 8.99 0.09 6.64 6.73
[ 8.94] [ 2.85] [ 6.87] [ 4.71] [ 6.09] [ 2.07] [ 4.23] [4.02] [1.86] [ 2.17]
70 131 922 2,903
Prior 12-month return 2.41 2.46 3.04 3.11 0.05 0.62 0.69 0.58 0.65 0.07
volatility
[2.03] [2.26] [2.62] [2.72] [ 0.23] [ 0.59] [ 0.70] [ 0.36] [ 0.47] [ 0.11]
70 129 917 2,855
Beta 0.91 0.91 0.75 0.82 0.00 0.16 0.09 0.16 0.09 0.07
[0.89] [0.94] [0.69] [0.72] [ 0.05] [0.19] [0.17] [0.25] [0.23] [ 0.02]
70 128 902 2,786
characteristics across acquirer types. Given our interest in announcement. Our measure of institutional ownership is
club deals, we focus mostly on differences between club from Thomson Financial’s 13F Holdings database, which
and sole-sponsored private equity deals. contains the stock-level holdings of institutional money
We begin with target size, measured as market value managers (but not hedge funds) collected from manda-
of equity (from CRSP) 43 trading days prior to the tory quarterly filings with the Securities and Exchange
announcement (i.e., the day before the beginning of the Commission (SEC). We compute the fraction of a firm’s
runup period). Consistent with the notion that club deals outstanding shares owned by all institutions at the end of
are a way for private equity firms to share the risk the quarter immediately prior to the acquisition an-
inherent in large transactions, club deal targets are nouncement date. Targets in club private equity transac-
significantly larger than targets of other types of tions have significantly higher institutional ownership
acquirers, including targets of sole-sponsored private than targets of other acquirers, including sole private
equity deals. This is true in both means and medians. equity firms, both on average and at the median. Since
This fact is potentially important for the interpretation of these targets are larger firms (on average), this is not
the results in Table 3 that club deals produce lower surprising.
premiums for target shareholders in light of existing Lastly, we consider stock-return-based measures of
evidence that target shareholder gains are negatively pre-deal performance and risk. We measure pre-deal
related to target size (e.g., Comment and Schwert, 1995; stock-return performance using buy-and-hold stock re-
Officer, 2003). turns for the 12 months prior to the beginning of the
We also consider the Tobin’s Q ratio of the target runup period (trading day 294 to day 43 relative to
(defined as in Kaplan and Zingales (1997) as the market the announcement date), both raw and market-adjusted.
value of assets divided by the book value of assets, where Targets of club acquisitions have generally insignificantly
the market value of assets equals the book value of assets different pre-deal stock-return performance than targets
plus the market value of common equity less the sum of of other acquirers. There is some evidence that publicly
the book value of common equity and balance sheet traded bidders acquire targets with better pre-deal
deferred taxes). Lang, Stulz, and Walking (1989) argue performance, suggesting that publicly traded acquirers
that acquirers have greater scope to create value if low are more likely to target well-performing and highly
target Q (or market-to-book) is caused by rectifiable valued firms than private equity groups (or private firms)
agency problems at the target firm. Our results confirm are.
the well-known finding (e.g., Bargeron, Schlingemann, We measure pre-deal target risk in two ways: using
Stulz, and Zutter, 2008) that private equity firms in the standard deviation of the target’s daily returns over
general acquire significantly ‘‘cheaper’’ targets (lower the 12 months prior to the beginning of the runup period
industry-adjusted Q ratios; industries are defined using and using the beta estimated in the market model
two-digit Standard Industrial Classification (SIC) codes) described in Section 4.1. Neither metric indicates sig-
than publicly traded acquirers do. However, there is no nificant differences in target-firm pre-deal risk between
significant difference in average target industry-adjusted the categories of private equity deals (sole-sponsored and
Q between club deals and sole-sponsored private equity club) on average,14 though there is weak evidence that
deals. To the extent that the scope to create value in club targets have lower return volatility at the median.
private equity transactions stems from existing opera- The targets of both sole-sponsored and club private equity
tional inefficiencies that would be reflected in low Q, deals have significantly lower total volatility and higher
these results suggest that lack of scope to create value is beta than the targets of acquisitions by publicly traded
unlikely to explain the lower premiums associated with acquirers do.
club deals. As an additional measure of risk, we also examine the
We also examine the target’s performance in terms of target’s operating risk (historical cash flow volatility, as
operating cash flow as measured by the ratio of EBITDA to measured by the volatility of EBITDA to assets based on up
total assets (from Compustat) for the fiscal year prior to to ten years of history, with a minimum of five). We also
the deal announcement—EBITDA is the most often used examine target complexity (as measured by the number
measure of cash flow in the private equity literature. of reported segments in the Compustat segment files) and
There is no significant difference in pre-deal industry- information asymmetry surrounding the target (as mea-
adjusted operating performance for targets of club versus sured by analyst forecast errors in the Institutional
sole private equity firm deals. Brokers’ Estimate System (IBES)). We find no significant
We next consider a measure of pre-transaction target differences between club and sole-sponsored private
leverage, the ratio of the book value of total debt (short- equity target firms along these dimensions. To conserve
and long-term, from Compustat) to the sum of the book space, we do not tabulate these results.
value of debt and the market value of equity. There are no Taken together, our univariate comparisons of target
significant differences in pre-deal target leverage between characteristics suggest that it is unlikely that the lower
club deal and non-club deal transactions. gains to club deal target shareholders are due to
Institutional ownership of the target firm may reduce
the ability of private equity firms to acquire such firms 14
As an alternative, we estimated summed betas, where we add up
cheaply. In Table 4 we show the average and median the betas with respect to today’s and yesterday’s excess market returns.
fraction of target shares that are reported as owned by We found no significant difference in these betas between club and sole-
institutional investors in the calendar quarter prior to deal sponsor targets. We thank an anonymous referee for this suggestion.
differences in pre-deal target performance, to the pre- geneity as possible, cluster standard errors by year, and
mium being incorporated in the stock price before deal correct the standard errors for heteroskedasticity.16
announcement, or to differences in risk. The results are displayed in Panel A of Table 5. The
In unreported analysis, we also estimate probit, linear results in the first three columns represent our base
probability, and logit models for club formation using the specifications. In these regressions the coefficient on the
combined sample of sole-sponsored and club private equity club indicator variable is significantly negative in the
deals. In these models the dependent variable is an indicator announcement return and buy-and-hold abnormal
variable for club deals and the independent variables are the markup regressions. These specifications suggest that
target characteristics in Table 4 plus year fixed effects. The target shareholders receive 6.8% and 8.6%, respectively,
results indicate that clubs do involve larger targets on less as a percentage of pre-bid equity value in club deals
average, but the coefficients on the other variables in Table 4 compared to sole-sponsored LBOs. The club deal
are not significant. This result is consistent with the coefficient in the buy-and-hold abnormal premium
conclusions from Table 4 with the exception that the regression is also negative but, because runups are noisy
difference in institutional ownership is no longer significant. but do not differ between sole-sponsored and club LBOs,
insignificantly different from zero.17
We hypothesize that target firms with high institu-
4.4. Target return regressions tional ownership will be less likely to accept a low offer
price, due to the strong incentives and sophistication of
4.4.1. Full-sample analysis their institutional owners to negotiate a fair price at the
To further investigate the possibility that the differ- bargaining table. To the extent that clubs inadvertently or
ences in gains to target shareholders between club deals intentionally constrain competition, we expect institu-
and other transactions may be due to differences in tional owners to mitigate the pricing consequences. On
observable target or deal characteristics, we estimate the other hand, since the assumption maintained
regressions in which we use three of the return measures throughout our analysis is that sole-sponsored LBOs are
in Table 3 as the dependent variables: announcement competitive even if club deals are not, we do not expect to
returns (CAR3), buy-and-hold abnormal markups, and find a similar effect in sole-sponsored deals.
buy-and-hold abnormal premiums. Because the differ- We investigate this idea by adding the target firm’s
ences in runups between sole-sponsored and club LBOs institutional ownership (as in Table 4) to the regressions
are insignificant in Table 3, and because runups are in columns 4, 5, and 6 in Panel A of Table 5. We interact
inherently noisy, the total deal premium (which is the target institutional ownership with our club and sole-
sum of the runup and markup) may be too noisy to sponsored indicator variables to examine whether the
capture salient differences between sole-sponsored and effect of institutional ownership differs across different
club private equity deals. This is the reason for reporting types of acquirers. The first row of these columns shows
regressions with buy-and-hold abnormal markups.15 that the estimated club discount for a target with zero
Our main indicator variables of interest are whether the institutional ownership is large and significant—the point
acquirer is a club, a private firm, or a public firm. Thus, the estimates range from 14.3 percentage points lower
omitted category is sole-sponsored private equity firms. The announcement returns to 27.1 percentage points lower
other explanatory/control variables are observable target abnormal premiums in club deals. The interactions of the
and deal characteristics, and are all relatively standard in the club indicator with institutional ownership are all
M&A literature. One of the control variables is the natural significantly positive, suggesting an incremental effect
logarithm of target size, for consistency with the prior by which the lower premiums for targets of club deals
literature, but our results are robust to additional linear and dissipate as institutional ownership increases, and are
quadratic size controls. Most of our control variables are completely eliminated for targets with high institutional
defined in the prior section. Those that are not previously ownership. Importantly, there is no incremental effect of
defined are: indicator variables for the method of payment institutional ownership on target deal-period returns
(cash or a mix); the bidder’s toehold at announcement; an when the acquirer is a sole private equity (PE) firm.
indicator variable for whether the offer is a tender offer; an The coefficients on the institutional ownership vari-
indicator variable for whether target managers have a able by itself suggest that institutional ownership is
hostile reaction to the offer; and an indicator variable for associated with higher premiums in all deals, regardless
whether there are other bidders competing to acquire the of acquirer type, perhaps because higher quality targets
target firm (Pre-competition: equal to one if there is another have higher institutional ownership. We emphasize that
offer in the SDC database for the same target in the six the incremental club deal effect exists even when
months prior to the bid announcement date). These
variables are all defined using SDC data. We also include
year and industry (based on two-digit SIC codes) fixed 16
To address the possibility that significance levels may be
effects to capture as much remaining unmodeled hetero- overstated due to repeat buyers, we also consider clustering by buyer
or by both buyer and year and obtain similar significance levels for key
coefficient estimates. We thank an anonymous referee for bringing this
15
As mentioned earlier, all our results are qualitatively similar if we issue to our attention.
17
use the premium defined as SDC-reported final deal price divided by the Despite the noisiness of the premium BHAR measure, as discussed
target’s stock price 43 days prior to deal announcement instead of this below we do find a significant club discount in the pre-2006 period even
returns-based measure of premium. using premium BHAR.
226
Table 5
Multivariate regressions explaining target returns.This table contains the results of multivariate regressions of CAR3, Markup BHAR, and Premium BHAR described in Table 3, and expressed in percent. Club,
Private, and Public are indicator variables for the deal categories described in Table 1. Pre-2006 is an indicator variable equal to one for deals announced prior to 2006, and zero otherwise. All cash (Mix) is an
indicator variable equal to one if the takeover offer is 100% cash (mix of cash and stock), and zero otherwise. Toehold at announcement is the fraction of the target’s shares owned by the acquirer at
announcement. Tender (Hostile) is an indicator variable equal to one if the takeover offer is a tender offer (receives a hostile reaction from target managers, as defined by SDC), and zero otherwise. Pre-
M.S. Officer et al. / Journal of Financial Economics 98 (2010) 214–240

competition is an indicator variable equal to one if another potential acquirer has bid for the target within the six months prior to the bid announcement date, and zero otherwise. PE firm(s) AUM is the total
assets under management of all prominent PE firms involved in the deal, measured as total fundraising in the prior ten years, in billions of dollars; the variable is zero for non-PE deals. PE inflow is the total
fundraising in the year prior to the deal by all PE funds with a U.S. geographical focus, in billions of dollars. All other explanatory variables are defined in prior tables. All regressions include industry (two-digit
SIC code) fixed effects. All regressions in Panel A also include year fixed effects. Standard errors are clustered by year and heteroskedasticity-consistent. t-Statistics are in brackets. , , and indicate that the
coefficient estimate is significantly different from zero at the 1%, 5%, and 10% levels, respectively.
CAR3 Markup BHAR Premium BHAR CAR3 Markup BHAR Premium BHAR CAR3 Markup BHAR Premium BHAR
Panel A: Regressions including year fixed effects
Club 6.77 8.58 5.08 14.27 23.04 27.14 1.30 0.36 6.03
[ 2.61] [ 2.52] [ 1.00] [ 2.42] [ 2.70] [ 2.06] [0.86] [ 0.13] [1.00]
Club Inst. ownership 12.02 28.91 50.14
[2.20] [2.62] [2.56]
Sole PE Inst. ownership 0.12 4.84 11.58
[0.02] [0.83] [1.41]
Inst. ownership 4.10 7.10 8.65
[2.48] [2.04] [2.39]
Club Pre-2006 8.49 12.13 23.91
[ 4.09] [ 2.77] [ 3.16]
Sole PE Pre-2006 4.00 1.13 4.84
[1.52] [0.31] [ 0.84]
Private 0.93 0.46 0.84 0.92 5.02 10.75 1.99 1.22 2.90
[ 0.47] [0.19] [0.24] [ 0.19] [1.02] [1.70] [1.87] [0.70] [ 0.88]
Public 2.22 4.06 7.83 2.03 8.06 17.29 5.15 4.85 4.14
[1.06] [1.56] [2.23] [0.43] [1.69] [2.71] [4.76] [3.09] [1.16]
Ln (Size) 2.49 3.05 5.48 2.90 3.68 6.51 2.53 3.11 5.60
[ 6.70] [ 6.69] [ 6.63] [ 6.41] [ 6.78] [ 6.49] [ 6.87] [ 6.84] [ 6.74]
Prior 12-month BHAR 0.02 0.02 0.03 0.02 0.02 0.03 0.02 0.02 0.03
[ 3.67] [ 0.90] [ 0.91] [ 3.08] [ 0.84] [ 0.84] [ 3.66] [ 0.89] [ 0.89]
Runup BHAR 0.21 0.21 0.21
[ 5.01] [ 5.04] [ 5.06]
Ind.-adjusted EBITDA/assets 0.35 0.00 6.68 0.38 0.48 5.41 0.35 0.01 6.63
[0.10] [0.00] [ 0.71] [ 0.11] [ 0.06] [ 0.56] [0.11] [0.00] [ 0.70]
Industry-adjusted Q 0.48 0.82 0.59 0.43 0.76 0.41 0.48 0.83 0.61
[ 2.41] [ 3.10] [ 1.10] [ 2.14] [ 2.89] [ 0.75] [ 2.44] [ 3.11] [ 1.12]
Beta 2.06 0.34 0.49 2.05 0.11 0.75 2.12 0.44 0.69
[2.39] [0.27] [0.21] [2.42] [0.08] [0.31] [2.48] [0.35] [0.29]
Prior 12-month ret. volatility 0.29 0.68 0.00 0.30 0.64 0.17 0.29 0.69 0.02
[ 0.60] [ 0.60] [ 0.00] [ 0.59] [ 0.51] [ 0.10] [ 0.61] [ 0.61] [ 0.01]
All cash 4.61 0.74 2.04 4.38 1.14 1.19 4.66 0.66 2.21
[3.50] [ 0.28] [0.59] [2.95] [ 0.40] [0.33] [3.53] [ 0.25] [0.65]
Mix 0.82 2.90 2.68 0.99 2.52 1.79 0.83 2.85 2.55
[0.81] [ 1.81] [ 1.01] [0.88] [ 1.54] [ 0.67] [0.83] [ 1.78] [ 0.97]
Toehold at announcement 6.77 3.21 20.48 6.17 6.64 15.97 6.60 3.43 20.03
[ 1.28] [0.34] [ 2.51] [ 1.00] [0.60] [ 1.69] [ 1.26] [0.36] [ 2.48]
Tender 8.47 6.12 7.36 8.66 6.60 8.02 8.40 6.05 7.25
[6.49] [5.62] [5.88] [6.75] [5.94] [6.18] [6.43] [5.55] [5.71]
Hostile 5.20 10.47 7.00 6.05 11.16 5.88 5.29 10.56 7.13
[3.10] [4.39] [2.81] [3.59] [4.25] [1.87] [3.15] [4.41] [2.87]
Pre competition 11.71 9.05 2.99 12.44 9.86 3.55 11.64 8.97 3.10
[ 8.25] [ 3.78] [0.95] [ 8.49] [ 3.73] [0.99] [ 8.22] [ 3.76] [0.98]

Constant 22.37 7.59 20.41 18.00 6.96 8.76 19.34 6.78 24.19
[1.67] [0.88] [3.44] [1.31] [0.63] [0.81] [1.43] [0.81] [4.27]
Observations 3382 3382 3382 3171 3171 3171 3382 3382 3382
R squared 0.15 0.12 0.14 0.15 0.12 0.14 0.16 0.12 0.14
Panel B: Regressions including controls for PE firm(s) assets under management (AUM) and PE industry inflow
Club 10.52 13.53 17.00 15.02 25.95 33.13 4.77 5.85 7.41
[ 4.80] [ 4.28] [ 3.02] [ 2.53] [ 3.37] [ 3.04] [1.12] [0.76] [0.67]
Club PE inflow 0.06 0.07 0.16 0.04 0.01 0.05 0.03 0.05 0.01
[1.64] [1.50] [2.37] [1.05] [0.11] [0.60] [ 0.81] [ 0.66] [ 0.06]
PE inflow 0.04 0.07 0.05 0.03 0.05 0.04 0.08 0.16 0.20
[1.94] [1.65] [1.05] [1.51] [1.30] [0.72] [2.62] [2.78] [2.79]
Club Inst. ownership 6.45 24.82 39.57
[1.08] [2.50] [2.55]
Sole PE Inst. ownership 1.45 0.01 3.75
[ 0.22] [0.00] [0.57]
[2.41] [2.16] [2.36]
Club Pre 2006 10.29 15.49 23.89
[ 3.71] [ 2.38] [ 2.27]
Sole PE Pre-2006 5.35 3.97 0.85
[1.80] [1.03] [ 0.13]
Pre-2006 5.24 13.30 22.00
[1.47] [1.96] [2.57]
Assets under management 0.05 0.06 0.14 0.07 0.13 0.27 0.08 0.09 0.14
[0.45] [0.49] [1.05] [0.59] [1.01] [2.07] [0.68] [0.65] [0.93]
Private 0.39 2.67 3.93 0.44 4.65 10.03 4.35 5.26 2.42
[0.18] [0.98] [0.96] [ 0.09] [0.89] [1.60] [2.22] [2.17] [0.50]
Public 3.29 5.40 9.28 2.37 7.01 15.03 7.26 8.03 7.83
[1.62] [1.95] [2.21] [0.50] [1.42] [2.48] [4.30] [3.68] [1.50]
Ln (size) 2.13 2.16 4.35 2.52 2.82 5.39 2.28 2.52 4.92
[ 4.60] [ 3.38] [ 4.39] [ 4.48] [ 3.78] [ 4.84] [ 4.87] [ 4.04] [ 4.81]
Prior 12-month BHAR 0.02 0.02 0.03 0.02 0.02 0.03 0.02 0.02 0.03
[ 4.04] [ 0.99] [ 1.32] [ 3.48] [ 0.93] [ 1.29] [ 3.73] [ 0.93] [ 1.20]
Runup BHAR 0.19 0.19 0.19
227
[ 4.78] [ 4.73] [ 4.93]
228
Table 5 (continued )
Panel B: Regressions including controls for PE firm(s) assets under management (AUM) and PE industry inflow
Ind.-adjusted EBITDA/Assets 2.65 6.09 2.39 1.79 5.37 3.65 2.15 4.87 0.44
[0.73] [0.70] [0.25] [0.47] [0.58] [0.37] [0.61] [0.59] [0.05]
Industry-adjusted Q 0.52 0.88 0.59 0.45 0.77 0.34 0.51 0.86 0.55
[ 2.55] [ 3.49] [ 1.16] [ 2.11] [ 3.00] [ 0.66] [ 2.50] [ 3.54] [ 1.09]
Beta 0.85 2.33 3.56 0.77 2.56 3.32 1.37 1.04 1.46
[0.75] [ 1.19] [ 1.20] [0.72] [ 1.39] [ 1.16] [1.27] [ 0.63] [ 0.57]
Prior 12-month ret. volatility 0.63 1.58 3.36 0.55 1.47 3.07 0.34 0.86 2.17
[1.19] [1.53] [2.39] [0.99] [1.28] [2.00] [0.64] [0.86] [1.53]
All cash 4.46 1.02 1.03 4.23 1.33 0.14 4.67 0.45 2.00

[3.61] [ 0.38] [0.30] [3.06] [ 0.46] [0.04] [3.92] [ 0.17] [0.60]
Mix 1.11 2.32 2.58 1.13 2.17 2.01 1.08 2.32 2.52
[1.44] [ 1.37] [ 1.00] [1.34] [ 1.28] [ 0.79] [1.38] [ 1.35] [ 0.99]
Toehold at announcement 6.81 3.33 19.78 5.16 9.39 12.18 6.21 4.58 17.64
[ 1.27] [0.36] [ 2.48] [ 0.84] [0.91] [ 1.43] [ 1.18] [0.50] [ 2.25]
Tender 8.62nnn 6.64nnn 9.43nnn 9.04nnn 7.35nnn 10.41nnn 8.46nnn 6.28nnn 8.85nnn
[6.36] [5.12] [6.26] [6.93] [5.88] [6.74] [6.36] [5.05] [5.63]
Hostile 5.16nnn 10.85nnn 7.73nnn 6.02nnn 11.69nnn 6.87n 5.20nnn 10.85nnn 7.70nn
[2.94] [4.25] [2.88] [3.36] [4.16] [2.04] [2.95] [4.16] [2.71]
Pre-competition 12.03 9.68 3.36 12.43 9.99 4.43 11.84 9.20 3.99
[ 8.20] [ 3.69] [0.92] [ 7.88] [ 3.46] [1.13] [ 8.22] [ 3.67] [1.12]
Constant 18.47 1.29 14.27 17.79 4.05 3.93 9.36 14.06 5.28
[1.44] [0.22] [2.35] [1.30] [ 0.51] [0.54] [0.84] [ 1.99] [ 0.46]
Observations 3382 3382 3382 3171 3171 3171 3382 3382 3382
R-squared 0.14 0.09 0.11 0.14 0.09 0.10 0.14 0.10 0.11
controlling for any aspect of target quality that may be effects in our main specifications because including them
proxied by institutional ownership.18 may not produce a well-specified analysis of the potential
In the last three columns of Panel A of Table 5 we club discount. To illustrate the point, suppose that,
interact our club and sole-PE indicator variables with an holding all else constant, two private equity firms are
indicator variable for deals announced prior to 2006, the both able to pay 10% less when acting alone, but 15% less
year in which the financial media turned a spotlight onto when they form a club together. This is a true club
the practice of club deals by private equity firms and the discount, but including private equity firm fixed effects
U.S. Department of Justice started its informal inquiry into would result in an estimated 5% ( = 15%–( 10%–10%))
the practice.19 It is possible that the attention placed on club premium.20
club deals by the media and the U.S. government caused The coefficients on the control variables in Panel A of
private equity firms to be more timid in exploiting the Table 5 generally have signs and significance levels that are
lack of competition created by private equity clubs and, consistent with the prior literature. The coefficients on the
therefore, less aggressive with deal pricing in 2006 and pre-bid competition indicator variables appear unusual,
2007. It is also possible that increased funding to private because these coefficients are significantly negative (in-
equity firms in 2006 and 2007 increased competition for dicating discounts of about 10 percentage points) in most
target firms such that the reduction in competition of the specifications in the table, even though intuition
associated with clubs had negligible pricing consequences suggests that pre-bid competition would be associated
in those years. with higher premiums. However, it is likely that this
Consistent with these possibilities, the coefficients on coefficient reflects the fact that prior bids reduce the
the interactions of the pre-2006 and club indicator surprise associated with the current bid. The same result is
variables are significantly negative for all measures of reported in Officer (2003). There is also evidence of markup
returns (announcement, markup, and premium), while pricing (Schwert, 1996) in our regressions. Specifically, the
the coefficients on the club indicator variable by itself are coefficients on the runup variables in the markup regres-
insignificantly different from zero. The net pre-2006 club sions are all significantly negative and equal to about 0.2 in
discount, represented by the linear expression (Club + magnitude, suggesting that acquirers only shade their bids
Club Pre-2006 Sole PE Pre-2006), is significantly ne- downwards by about 20% of the pre-bid runup.
gative for all three return measures, and ranges from 11.2 In Panel B of Table 5 we repeat the analysis in Panel A
to 13.6 percentage points of pre-bid target equity value in including controls for private equity firm assets under
magnitude. Therefore, essentially all the club deal dis- management (AUM) and private equity fund inflows
count that we observe is driven by club deals prior to calculated using data from Preqin. This analysis is
2006. The same pattern is not evident in the coefficients intended to provide insight into whether capital con-
on the interactions of the sole-PE indicator variable. straints are likely to drive the pricing of club deals, and
All of the results discussed above are also robust to a whether increased inflows to PE funds around 2006,
variety of additional or alternative controls. Specifically, corresponding to looser capital constraints and more
the results are robust to scaling EBITDA by market value competition for deals, can explain the structural break in
43 days before announcement rather than assets; to a club deal pricing around 2006.21 We measure the AUM of
control for return on invested capital, measured by the prominent PE firms involved in a deal as their total
EBITDA scaled by total deal value; to a control for firm fundraising in the ten years prior to the deal (i.e., for club
size based on assets; to controls for governance variables deals, we add the AUMs of the individual prominent PE
(where data are available), specifically the fraction of firms; the variable is zero for non-PE deals). We measure
independent directors on the target’s board and the the annual inflow to the PE industry as the total
Gompers, Ishii, and Metrick (2003) index; to a control fundraising in the year prior to the deal by all PE funds
for the number of segments in the firm (where data are with a U.S. geographical focus. In Panel B of Table 5, we
available); and to a control for the age (a measure of drop year fixed effects because they would subsume this
experience) of the private equity firm(s). inflow variable.
The results are also robust to including private equity The results in Panel B of Table 5 continue to support our
firm fixed effects in the regressions. This suggests that our conclusions from Panel A. The estimated magnitude of the
results are not driven by the ability of some private equity club deal level effect is stronger, the cross-sectional
partnerships to pay low prices whether acting alone or as institutional ownership effect is slightly weaker, and the
part of a club. That said, we do not include these fixed time-series structural break around 2006 is just about the
18 20
These results are qualitatively unaffected if we instead use a A potential solution is to include ‘‘club’’ fixed effects—in the case
measure of long-term institutional ownership, defined as the percentage of a sole-sponsored LBO, these would be private equity firm fixed effects.
ownership of institutions who also held an ownership stake four However, a subset of these fixed effects would absorb the club indicator
quarters prior (similar to the rationale in Gaspar, Massa, and Matos, variable, a key variable of interest in our regressions. Moreover, the
2005). One caveat to our analysis of institutional ownership is that hedge structural break around 2006 would be identified off a few clubs that
funds and other blockholders may not report holdings as institutions. transact both before and after 2006. Likewise, the cross-sectional
Because of this, our institutional ownership measure is an imperfect institutional ownership effect would be identified off clubs that deal
measure of the holdings of shareholders with strong incentives and with different levels of institutional ownership in separate transactions.
sophistication to maximize value. None of our conclusions are affected by including such fixed effects. We
19
The direct effect of the pre-2006 indicator variable is subsumed thank an anonymous referee for bringing this solution to our attention.
21
by the year fixed effects. We thank an anonymous referee for suggesting this analysis.
same. The coefficient estimate on AUM is insignificant in of including these transactions is that they serve as a
eight of the nine specifications. The positive and sometimes control sample that allows us to estimate the effects of a
significant coefficient estimates on the PE inflow level battery of control variables. However, it may be that firms
variable are consistent with the ‘‘money chasing deals’’ targeted by prominent private equity bidders differ in
phenomenon described in Gompers and Lerner (2000). important unobservable ways from those targeted by
However, such money chasing deals do not explain the other types of bidders.
2006 structural break in the club deal discount. The Therefore, in Table 6 we confirm that the conclusions
interaction of the inflow measure with the club indicator from Table 5 continue to hold in analysis using only the
variable is insignificant in all nine specifications except sample of 201 private equity transactions. Panel A repeats
one. Crucially, our finding of a structural break in 2006 (the the regression analysis in Panel A of Table 5, but we drop
last three specifications) continues to be significant. some of the control variables given the limited number of
observations. We retain controls for target size and risk, as
4.4.2. Private equity sample analysis well as year fixed effects.
Our analysis in Table 5 includes M&A transactions The first three columns of Panel A of Table 6 confirm
completed by non-private equity bidders. The advantage that club deals do involve significantly smaller gains to
Table 6
Prominent private equity bidder sample analysis This table contains the results of regression and matching analyses for the sample of 201 private equity
transactions conducted by prominent private equity firms. The dependent variables, CAR3, Markup BHAR, and Premium BHAR, are described in Table 3,
and expressed in percent. All explanatory variables are defined in prior tables. All regressions in Panel A contain year fixed effects, and standard errors are
clustered by year and heteroskedasticity-consistent. The first column of Panel B reports club treatment effects from kernel propensity-score matching
using propensity scores from a probit model for club formation with the control variables in Panel A and year fixed effects. The second column of Panel B
reports treatment effects based on Abadie-Imbens (2007) matching estimator with matching conducted based on the control variables in Panel A. t-
Statistics are in brackets. , , and indicate that the coefficient estimate is significantly different from zero at the 1%, 5%, and 10% levels, respectively.
In the last row of Panel A, the p-values are two-tailed values.
Panel A: Multivariate regressions explaining target returns
CAR3 Markup Premium CAR3 Markup Premium CAR3 Markup Premium

BHAR BHAR BHAR BHAR BHAR BHAR
Club 7.37 9.37 11.65 16.49 28.56 39.85 2.97 0.52 2.31
[ 2.07] [ 2.45] [ 2.11] [ 2.31] [ 2.29] [ 2.44] [1.95] [0.24] [ 0.36]
ClubInst. ownership 14.91 29.04 43.19
[1.66] [1.94] [2.02]
[0.72] [0.71] [0.93]
Club Pre-2006 16.03 15.33 14.47
[ 3.67] [ 3.35] [ 1.54]
Ln (size) 1.34 2.17 1.62 1.68 2.35 1.67 2.12 2.92 2.33
[ 1.37] [ 1.39] [ 1.15] [ 1.42] [ 1.12] [ 0.84] [ 1.75] [ 1.73] [ 1.63]
Prior 12-month BHAR 0.05 0.00 0.06 0.05 0.00 0.05 0.04 0.01 0.05
[ 1.83] [ 0.04] [ 0.75] [ 1.31] [ 0.03] [ 0.79] [ 1.44] [0.16] [ 0.63]
Runup BHAR 0.20 0.23 0.20
[ 1.43] [ 1.82] [ 1.52]
Beta 0.11 2.24 2.93 0.55 1.08 5.56 1.92 4.18 1.10
[ 0.03] [0.41] [ 0.45] [ 0.11] [0.18] [ 0.83] [0.46] [0.80] [ 0.17]
Prior 12-month ret. 2.57 0.92 0.11 4.32 2.79 1.94 2.51 0.86 0.06
volatility
[1.03] [0.22] [0.02] [1.20] [0.57] [0.32] [0.98] [0.20] [0.01]
Constant 11.68 13.07 28.72 4.30 5.42 20.50 9.42 10.93 26.67
[3.17] [1.95] [2.73] [0.61] [0.57] [1.37] [2.33] [1.57] [2.49]
Observations 198 198 198 169 169 169 198 198 198
R-squared 0.26 0.29 0.27 0.32 0.37 0.34 0.31 0.32 0.28
Club+ Club Pre-2006 13.05 14.82 16.78
p-Value 0.004 0.006 0.032
Panel B: Estimated treatment effect from matching analysis for club effects in target returns
Propensity-score matching Abadie-Imbens matching
CAR 3 9.75 6.03

[ 3.05] [ 2.40]
Markup BHAR 12.11 7.28

[ 2.87] [ 2.10]
Premium BHAR 14.74 11.86

[ 2.74] [ 2.38]
target shareholders. The estimated club discounts are typically focuses on pricing measures based on multiples
somewhat larger in magnitude than those in Table 5, of an accounting variable, most commonly sales or
ranging from 7.4 to 11.7 percentage points. EBITDA. Unfortunately, multiples are a noisy valuation
The next three columns confirm the result that the metric and it is not clear what the appropriate benchmark
club discount is mitigated by high institutional owner- for comparison should be. Comparisons of multiples also
ship. In all three columns the coefficients on the club suffer from a potential omitted variables bias. In general,
indicator are significantly negative. The coefficients on the the main advantage of studying valuation multiples is that
interaction of institutional ownership with the club they facilitate comparison of acquisitions of public and
indicator are all positive, and significantly so for the private targets.
buy-and-hold abnormal returns. The magnitudes of these In this paper, we do not have to rely on noisy valuation
interactions are similar to those in Table 5. The insignif- multiples because we study acquisitions of public targets
icant coefficients on institutional ownership by itself for which stock-return data are available. Indeed, to the
again indicates that, consistent with Table 5, there is no extent that the market does not anticipate the deal, post-
similar effect for sole-sponsored private equity deals. deal announcement stock returns (i.e., markups) should
The final three columns confirm that the club discount is represent the true wealth gain experienced by target
concentrated before 2006 and virtually disappears in 2006 shareholders, since stock prices reflect all available value-
and 2007. In the announcement return and abnormal relevant information in an efficient market. Nevertheless,
markup regressions, the coefficients on the interaction of valuation multiples can serve as a robustness check for
the club and pre-2006 indicator variables are significantly returns-based measures of premiums because the analy-
negative, with magnitudes of 16.0 and 15.3 percentage sis does not require a choice of measurement window and
points. In the buy-and-hold abnormal premium regression, because deal valuation multiples should not be affected
this coefficient is negative but not significant at conven- even if the target shares are misvalued by the market
tional levels. However, the sum of the coefficients on the before the deal. For these reasons, and in keeping with the
club indicator and the interaction term ( 2.3+– private equity literature, we analyze transaction multiples
14.5= 16.8, or a 16.8 percentage point discount) is in the following way.
significantly negative (p-value=0.03). This sum of coeffi- We use the ‘‘comparable industry transaction method’’
cients represents the club deal discount before 2006 described in Kaplan and Ruback (1995) and employed in
(Club=1 and Pre-2006=1), and is also significantly negative Officer (2007). We measure deal multiples by dividing
in the prior two columns with p-values of 0.00 and 0.01, each of four different measures of enterprise value by
respectively. annual sales or EBITDA (from Compustat for the fiscal year
In Panel B of Table 6, we confirm the robustness of our ending immediately prior to the deal announcement
finding of a club discount when we employ matching, rather date). The reason for using four potential measures of
than regression, analysis. The first column of Panel B enterprise value is that each is potentially flawed in its
contains estimated club treatment effects on premiums own way.
from a propensity-score matching analysis in which The four measures of target enterprise value we use
propensity scores are determined using a probit model for are:
club formation that uses the control variables in Panel A of
Table 6 plus year fixed effects as independent variables. The
(1) Deal value excluding assumed liabilities (from
estimates are computed using a kernel estimator that
SDC) + total debt (from Compustat) cash (from Com-
weights matches by their closeness in propensity scores.
pustat);
The second column of Panel B contains estimated club
(2) Market capitalization at the delisting date (from
treatment effects on premiums using the bias-corrected
CRSP) +total debt (from Compustat) cash (from
matching estimator of Abadie and Imbens (2007), in which
Compustat);
the matching is conducted using the control variables in
(3) (Delisting stock price (split adjusted) multiplied by
Panel A of Table 6. The estimated club treatment effects,
the number of shares outstanding at the beginning of
which essentially represent average differences in pre-
the runup period (both from CRSP)) +total debt (from
miums between club and sole-sponsored deals with similar
Compustat) cash (from Compustat);
observable characteristics, are similar in magnitude to the
(4) (Price paid to acquire each target share (from SDC)
regression estimates in Table 5 and Panel A of Table 6.
multiplied by the number of shares outstanding at the
Overall, the evidence presented in this section suggests
beginning of the runup period (from CRSP)) +total
that target shareholders gain less in club deals than in sole-
debt (from Compustat) cash (from Compustat).
sponsored private equity deals, that this discount in
concentrated before 2006, and that the discount is more
pronounced in target firms with low institutional ownership. The principal difference between these four measures is
the measurement of the market capitalization of the
target firm (including the premium paid by the acquirer).
5. Multiples-based measures of acquisition prices Values (1) and (4) above employ SDC data to calculate this
value, while values (2) and (3) employ CRSP data only.
While the M&A literature typically focuses on returns- Value (2) differs from value (3) if the number of shares
based measures of premiums paid to target shareholders, outstanding changes between the beginning of the runup
as we have done so far, the private equity literature period and the delisting date: this could happen in, for
example, two-tier offers (which are likely part of our Table 7

sample). Multivariate regressions explaining acquisition discounts based on deal
multiples.This table contains the results of multivariate regressions
Cash (from Compustat for the fiscal year ending
where the dependent variables are acquisition discounts for private
immediately prior to the deal announcement date) is equity sponsored deals based on enterprise value multiples, EV/Sales
subtracted to ensure that we are consistent in capturing and EV/EBITDA. The acquisition discount is defined as the percent
payment by the acquirer for the real assets and operations difference between a deal multiple for a private equity sponsored deal
of the target firm: private equity firms, for example, often and the average multiple for an equal-weighted portfolio of non-private
equity sponsored deals. A negative discount means that the private
use the target’s cash to directly offset any liabilities equity deal multiple is lower than that of the comparison portfolio.
assumed as part of the acquisition (Kaplan and Stein, There are four different measures of enterprise value (EV) employed in
1993). By subtracting cash we are also ensuring that our the numerators of the multiples, and the results reported in this table are
results are robust to systematic differences in excess cash based on the average discount across the four different EV/Sales or EV/
EBITDA multiples. In Panel A, each portfolio contains all non-private
holdings by targets of the various types of acquirers. We
equity sponsored deals announced in the three year window centered on
also discard observations from the sample if either the the announcement date of the private equity sponsored deal and for
numerator (enterprise value) or denominator (EBITDA or which the target is in the same two-digit SIC code industry as the target
sales) is negative, and then winsorize each of the multi- of the private equity sponsored deal. In Panel B, each portfolio contains
ples independently at the 5th and 95th percentiles. all non-private equity sponsored deals announced in the three year
window preceding the announcement date of the private equity
We then use these deal multiples to produce estimates sponsored deal, for which the target is in the same two-digit SIC code
of the ‘‘discount’’ in private equity sponsored acquisitions industry as the target of the private equity sponsored deal, and for which
(as in Officer, 2007). Specifically, for each private equity the target is within + /- 50% of the market value of the target of the
sponsored deal in our sample (club or sole-sponsored), we private equity sponsored deal, measured as market value 43 trading days
before takeover announcement. All explanatory variables are defined in
find a matching portfolio of non-private equity sponsored
prior tables. Standard errors are clustered by year and heteroskedasti-
deals and calculate the average multiple for that portfolio. city-consistent. t-Statistics are in brackets. , , and indicate that
Each portfolio is formed by finding all non-private equity the coefficient estimate is significantly different from zero at the 1%, 5%,
sponsored deals announced in a three-year window and 10% levels, respectively. In the last row of Panels A and B, the p-
centered on or trailing the announcement date of the values are two-tailed values.
private equity sponsored deal and for which the target is Panel A: Three-year match window centered on deal announcement
in the same two-digit SIC code industry as the target of
the private equity sponsored deal. We compute the EV/sales EV/EBITDA
percent difference between the sales or EBITDA multiple
Club 24.44 14.95
for each private equity sponsored deal and the compar-
[4.31] [1.62]
able multiple for the matching portfolio. These percent Club Pre-2006 51.91 26.95
differences are measures of the acquisition discount for [ 4.11] [ 2.06]
each private equity sponsored deal (more negative Pre-2006 13.94 13.13
numbers represent larger discounts). For brevity, we [1.33] [1.00]
Ln (size) 13.49 0.78
average the discounts across the different definitions of [3.86] [0.20]
enterprise value to obtain two discount measures for each Constant 4.82 23.51
deal (one for sales multiples and one for EBITDA multi- [ 1.40] [ 2.23]
ples; we obtain similar results with each of the four Observations 183 166
R-squared 0.08 0.02
different definitions of enterprise value), and compare the
Club+Club Pre-2006 27.47 12.01
resulting discount measures across club and sole-spon- p-Value 0.017 0.080
sored private equity deals. This analysis is therefore akin
to a differences-in-differences methodology. Panel B: Three-year match window preceding deal announcement and
Table 7 presents regressions with the acquisition size matched
discounts for private equity sponsored deals as the EV/sales EV/EBITDA

dependent variable. The sample includes only private
equity sponsored deals (sole-sponsored or club), because Club 62.61 31.05
the discount is calculated relative to portfolios including [9.41] [2.69]
Club Pre-2006 97.16 44.74
all other deals (non-private equity sponsored) in the main
[ 5.57] [ 3.25]
sample. Pre-2006 29.58 7.04
In Panel A of Table 7, the matching window is a three- [1.50] [0.36]
calendar-year window centered on the announcement Constant 7.67 14.33
date (as in Officer, 2007), and we control for the natural [ 0.58] [ 0.82]
Observations 128 108
log of target size (market value of equity from CRSP 43 R-squared 0.04 0.05
days prior to the announcement) in lieu of matching on Club+Club Pre-2006 34.55 13.68
size (which is addressed below). The independent vari- p-Value 0.045 0.085
ables of interest to us are the club indicator variable and
its interaction with the pre-2006 indicator variable. In
both specifications (based on sales or EBITDA multiples)
the interaction of the club indicator and the pre-2006 Furthermore, the net pre-2006 club discount (represented
indicator has a significantly negative coefficient, consis- by (Club +Club Pre-2006), reported at the bottom of the
tent with the returns-based results in Tables 5 and 6. table) is significantly negative for both sales- and EBITDA-
based discount measures. The estimated magnitude of the 6. Evidence on post-announcement competition and
net pre-2006 club discount using EBITDA-based discount benign rationales for club formation
measures, 12.0%, is similar to the estimates using target
stock returns in Section 4, while that using sales-based 6.1. Post-announcement competition
multiples, 27.5%, is considerably larger.
While matching comparable non-private equity trans- To further investigate the possibility that our findings of
actions only by industry (and controlling for size in the lower pricing in club deals are due to club formation
regression) has the benefit of producing more matches reducing competition for target firms, we examine the
than implementing an additional match on size would, extent to which the target firm receives binding competing
one might be concerned that a control for (log) size may bids after announcement of the private equity takeover,
not adequately capture the relation between target size which we term post-announcement bid competition. This
and acquisition multiples. Moreover, one might be measure of competition contrasts with the pre-announce-
concerned that the three-year centered window incorpo- ment bid competition (the extent to which the target firm
rates information that was not available to the private receives competing expressions of interest or binding bids
equity firms, or their targets, at the time of the deal before announcement of the private equity takeover), which
(i.e., information about future deals). To address these is the focus of Boone and Mulherin (2009). Given the
concerns, Panel B of Table 7 employs matches from the possibility of collusion, non-binding or informal expressions
three-calendar-year window immediately preceding the of interest could be shams, and it is even possible that
announcement date and we further match on size (and so purposefully low binding bids are orchestrated prior to the
omit a control for size in the regressions); we require that announcement of the deal to provide the appearance of
matching non-private equity targets have a market value competition.23 In contrast, post-announcement competing
within + /- 50% of the market value of the private equity bids truly do compete with the bid of the private equity
target, measured as the market value 43 trading days firm(s) involved in the deal.
before the takeover announcement. In Table 8, we present evidence on the incidence of
As in Panel A, the specifications in Panel B show a post-announcement competition, and test whether club
significantly negative coefficient on the interaction of the deals encounter less post-announcement competition
club and the pre-2006 indicator variables, and the net pre- than sole-sponsored private equity deals. Our approach
2006 club discount (sum of the coefficients on the club relies on publicly disclosed bids in SDC supplemented
indicator variable and the interaction term) is again with a Factiva search for news articles related to
significantly negative. The estimated magnitudes of the competing bids for any of the private equity targets in
net pre-2006 club discount are similar to those in Panel A. our sample. We construct an indicator variable for post-
Overall, the results in Table 7 are consistent with the announcement competition that is equal to one if either
evidence in Tables 5 and 6 that there is a club deal SDC or a Factiva search indicates at least one competing
discount in the pre-2006 period.22 bid by another potential acquirer in the six-month period
One common facet of multiples-based analyses is following the announcement date in SDC, and zero
extreme noise in the data (see, for example, Officer, otherwise.
2007). While we confirm our principal result from Table 5, Panel A of Table 8 shows post-announcement compe-
that club deals are completed at significantly lower prices tition in 20 out of 131 sole private equity sponsored deals
than sole-sponsored deals are before 2006, the noise in (15.3%) and four out of 70 club deals (5.7%), consistent
the multiples-based measures is obvious in the coeffi- with club deals depressing bid competition. Interestingly,
cients and low R-squared values reported in Table 7. We in only four of these 24 cases is the competing bidder
find no evidence of club premiums after 2006 using another prominent private equity firm. The difference is
returns-based measures of premiums in Table 5, and yet statistically significant in a univariate probit regression
there is some evidence of such in Table 7 (the coefficient (column 1 of Panel B). The average target return from the
on the club indicator by itself). Moreover, some of the initial bid to the final bid in the 20 sole private equity
statistical significance of the effects in Table 7 is weaker sponsored deals with post-announcement competition is
than in Table 5. Both of these conditions are likely the 11.5% whereas the corresponding average for the four
direct result of noise coming from acquisition multiples. club deals with post-announcement competition is just
5.2%.
We consider separately a subsample in which we
expect the potential anticompetitive effects of clubs to be
22
The results in Table 7 continue to hold if we adopt a different
form of matching which relies on Standard & Poor’s (S&P)-500-level-
23
adjusted deal values. Specifically, instead of including a specific match Even in the absence of repeat play among a handful of private
on time, we adjust the equity values of the comparable transactions by equity firms, bids in a one-shot auction may not reliably measure the
multiplying them by the ratio of the S&P-500 index level at the private extent of real competition because it is not uncommon for private equity
equity deal announcement date to the S&P-500 index level at the firms to submit separate bids in an auction, and then for the eventual
comparable deal announcement date. We then match using either winner to later invite losers to participate in the same deal. In one such
industry alone (comparable to the results in Panel A) or industry and case, KKR won the auction for PanAmSat in April 2004, beating out the
pre-deal target market value of equity (comparable to the results in Carlyle Group and Providence Equity Partners, only to include them in
Panel B), and find qualitatively similar results to those reported in the deal two months later. ‘‘Carlyle, Providence Will Each Buy 27%
Table 7. We thank an anonymous referee for this suggestion. PanAmSat Stakes,’’ Bloomberg, June 29, 2004.
Table 8
Post-announcement competition. This table reports evidence on the incidence of post-announcement competition in leveraged buyouts sponsored by
prominent private equity firms. Post-announcement competition is an indicator variable equal to one if another potential acquirer bids for the target in
the six months after the deal announcement date (gathered through SDC and Factiva news searches), and zero otherwise. All of the explanatory variables
are defined in prior tables. Top 5 refers to the PEI list and includes Carlyle, KKR, Goldman, Blackstone, and TPG. The coefficients reported in Panel B are
marginal responses. Standard errors are heteroskedasticity-consistent and t-statistics are reported in brackets. , , and indicate that the coefficient
estimate is significantly different from zero at the 1%, 5%, and 10% levels, respectively.
Panel A: Frequency distributions
Sole PE % Club % All %
Full sample
Competing bid 20 15.3 4 5.7 24 11.9
No. of competing bid 111 84.7 66 94.3 177 88.1
N 131 70 201
Top 5
Competing bid 9 22.5 1 2.5 10 12.5
No. of competing bid 31 77.5 39 97.5 70 87.5
N 40 40 80
Panel B: Probit regressions explaining post-announcement competition
Sample All PEI5 All

(1) (2) (3)
Club 0.10 0.20 0.33

[ 2.00] [ 2.51] [ 2.97]
Club Institutional ownership 0.32
[2.49]
Institutional ownership 0.04
[ 0.62]
Ln (size) 0.00
[ 0.34]
Prior 12-month return 0.00
[ 1.70]
CAR3 0.00
[ 2.21]
Toehold at announcement 0.40
[ 0.80]
Tender 0.07
[1.81]
Hostile 0.34
[2.62]
Observations 201 80 169
Pseudo R-squared 0.03 0.14 0.21
strongest, namely clubs involving the top-five private institutional ownership by itself indicates that, similar to
equity firms from the PEI50 rankings, which are Carlyle, our pricing results, there is no similar effect for sole-
KKR, Goldman Sachs, Blackstone, and TPG. Our sample has sponsored private equity deals.
80 deals involving these private equity firms, 40 of which Overall, the evidence on post-announcement competi-
are sole-sponsored and 40 of which are clubs. Panel A of tion is consistent with our pricing results and with the
Table 8 shows post-announcement competition in nine view that club deals are associated with reduced competi-
out of the 40 sole-sponsored deals (22.5%) but only one of tion for target firms.
the 40 club deals (2.5%). This difference is statistically
significant (column 2 of Panel B).
Our main evidence on prices paid in club deals 6.2. Financing commitments
suggests that sophisticated institutional investors play
an important role in mitigating the inadvertent or To further study potential differences between club and
intentional anticompetitive effect of clubs. Column 3 of sole-sponsored private equity deals in our sample, we read
Panel B shows that the negative relation between club definitive merger proxy statements (DEFM14s, or in the few
deals and post-announcement competition is mitigated cases of tender offers, SC TOs) and do not detect important
by high institutional ownership, which further suggests differences in deal characteristics that may explain our
that encouraging or soliciting post-announcement bid finding of low prices in club deals. In the process, we collect
competition is one of the ways in which institutional information on (i) debt and equity commitments made to
investors improve prices. The insignificant coefficient on target shareholders, and (ii) total funds needed to complete
Table 9
Financing commitments and debt financing terms. This table reports evidence on financing commitments and debt financing terms in leveraged buyouts
sponsored by prominent private equity firms. Information on debt and equity commitments made by banks and private equity firms, respectively, for
deals completed in 1994–2007, is gathered from definitive merger proxy or tender offer statements filed with the SEC. Data on debt financing terms are
gathered from DealScan. For deals with multiple credit facilities in DealScan, data are weighted using face values. No. of lenders is the number of lenders
providing debt financing for the deal. Spread is the interest rate spread over LIBOR, and Adj. spread is the spread minus the average spread on all LBO debt
taken out in the previous calendar quarter, both expressed in basis points per year. Max. debt/EBITDA is the maximum allowed debt-to-EBITDA ratio
under the terms of the credit facility, and Adj. max. debt/EBITDA is Max. debt/EBITDA relative to the average corresponding covenant on all LBO debt
taken out in the previous calendar quarter. Post-LBO leverage is the ratio of debt-to-debt plus equity under the capital structure immediately post-LBO.
All other variables are defined in previous tables. In Panel C, the reported numbers are means for the samples of Sole PE and Club deals. Standard errors
are heteroskedasticity-consistent and t-statistics are reported in brackets. , , and indicate that the coefficient estimate is significantly different
from zero at the 1%, 5%, and 10% levels, respectively.
Panel A: Total financing commitments relative to total funds needed
Sole PE Club All
Mean (%) 99.4 101.6 100.4

Median (%) 100.0 101.2 100.0
Under 19 15 34
Over 15 5 20
Equal 27 25 52
N 61 45 106
Panel B: Equity commitments relative to total financing commitments
Sole PE Club All
Mean (%) 33.7 38.2 35.6

Median (%) 30.8 34.6 31.9
N 69 52 121
Panel C: Debt financing terms
No. of lenders Spread Adj. spread Max. debt/EBITDA Adj. max debt/EBITDA
Sole PE (mean) 7.5 289 32 6.1 0.2

N 61 58 58 28 28
Club (mean) 10.5 286 43 7.0 0.4
N 46 46 44 24 23
Panel D: Regressions explaining debt financing terms
No. of lenders Spread Adj. spread Max. debt/EBITDA Adj. max debt/EBITDA
Club 2.76 9.15 12.61 0.29 0.52

[2.33] [0.40] [0.51] [0.62] [1.04]
Ln (size) 1.43 19.67 24.18 0.46 0.02
[3.05] [ 3.47] [ 3.48] [3.41] [ 0.10]
EBITDA/assets 3.62 23.75 33.38 5.42 2.39
[0.71] [0.20] [0.25] [ 1.48] [ 0.64]
Post-LBO leverage 10.07 46.24 41.83 2.98 2.90
[2.68] [0.55] [0.50] [1.77] [1.38]
Constant 2.76 208.82 130.53 6.18 1.88
[0.97] [3.29] [ 2.01] [4.75] [ 1.16]
Observations 85 83 83 43 43
R-squared 0.26 0.06 0.09 0.33 0.08
the deal. These data may provide an objective assessment of somewhat higher in club deals (101.6%) than it is in
the financial credibility of offers made to target shareholders sole-sponsored private equity deals (99.4%), but the
and explain differences in prices. We collect these data from difference is not statistically significant. Hence, there
Capital IQ and the SEC’s Edgar database when available. does not appear to be a large enough systematic
Neither of these databases contains proxy statements for difference in the financial credibility of offers as proxied
deals prior to 1994. by financing commitments made to target shareholders to
We present our findings from this analysis in Table 9. explain their willingness to accept low prices in club
Panel A shows that the total amount of debt and equity deals. Panel A also reports the number of deals where
commitments made to target shareholders is, on average, total commitments fall below, exceed, or equal total funds
roughly equal to the total amount of funds needed to (as presented in rows labeled under, over, and equal,
complete the deal—the average ratio of total respectively). Total commitments made are less than total
commitments to total funds is 100.4%. This ratio is funds needed in 15 out of 45 club deals (33%) compared to
19 out of 61 sole-sponsored private equity deals (31%). further investigate whether differences in target complex-
The minor difference in offer credibility captured by this ity (as measured by the number of reported segments in
measure is unlikely to explain low prices in club deals. Compustat segment files) or information asymmetry
In Panel B, we present evidence on the mix of debt and surrounding the target (as measured by analyst forecast
equity commitments made to target shareholders. On errors in IBES) explain our findings. On average, club
average, equity commitments represent 35.6% of all commit- targets report 2.0 segments for the fiscal year prior to the
ments received. This ratio is higher in club deals (38.2%) than deal announcement, which is statistically indistinguish-
in sole-sponsored private equity deals (33.7%), but the able from 2.2 segments reported by sole-sponsored
difference is not statistically significant at conventional targets (despite the larger size of targets in club deals
levels. These results are inconsistent with the hypothesis on average). We also find no significant differences in
that clubs are able to benefit from higher leverage ratios as a analyst forecast errors between the two types of targets.
result of attaching multiple private equity sponsors to a deal. To shed further light on these results on risk, and to
explore whether certification to debt financiers is likely to
be an important rationale for club formation, we next
6.3. Size, risk, and debt terms investigate whether debt financing terms are better or
worse in club deals than they are in sole-sponsored private
In this subsection, we provide more direct evidence on equity deals. Recent papers find that more reputable
the view that club deals are motivated by capital private equity firms (Demiroglu and James, 2010) and
constraints, diversification motives, or the ability to those with stronger bank relationships (Ivashina and
borrow at favorable terms. Kovner, 2007) are able to obtain better terms on the debt
Bearing in mind that our pricing results are robust to financing for the LBOs they conduct. While Table 2 shows
controls for target size, we investigate in more detail that a substantial fraction of our club deals is undertaken
whether capital constraints appear to motivate club deals. by the oldest and best-known private equity partnerships,
While Table 4 shows that club deals are significantly a large fraction of our sole-sponsored deals is as well, so it
larger on average than sole-sponsored LBOs, which is is not clear a priori whether we should expect clubs to
prima facie evidence that capital constraints may be a receive better debt financing terms.
motivation for club deals, the difference is not statistically If systematic differences in debt financing terms do
significant in the pre-2006 period (in which we find that exist, they could proxy for unobserved target character-
club targets receive significantly lower premiums). In istics. Specifically, if club deals have systematically worse
addition, we ask how many club deals are large enough debt financing terms than sole-sponsored private equity
that the private equity firms involved likely could not deals, then one may wonder whether unobserved target
have acquired the target without pooling resources. characteristics that drive up observed financing costs are
Specifically, we ask how many club deals are such that also driving down target prices. If, on the other hand, club
the deal size (as measured by enterprise value (1) defined deals systematically have better debt financing terms
in Section 5) exceeds the largest sole-sponsored private than sole-sponsored private equity deals, this would be
equity deal done by the club members during a [ 2,+ 2] consistent with ‘‘benign’’ theories of club deals in which
year window centered around the announcement date. lenders offer superior terms when multiple private equity
We find that only 16 of our 70 club deals (22.9%) meet firms attach their names and reputations to a deal—
this criterion, and the average premium in those deals though this would further deepen the puzzle about low
(26.1%) is not statistically different from that of the other prices paid to target shareholders in club deals.
54 deals (23.4%). Moreover, when we restrict the sample To investigate these ideas, we link our sample of deals
to club deals announced on or before December 31, 2005 in SDC to credit facilities in DealScan using target and
(because 2006 and 2007 club deals lack the relevant two- buyer names in SDC and all available name fields in
year ahead comparison group), we find that only six of our DealScan.24 Because we do this by electronic text match-
42 club deals (14.3%) meet the criterion, and the average ing, we further require that the facility start date in
premium in those deals (23.2%) is not statistically DealScan fall within a [-30, +30] day window centered
different from that of the other 36 deals (22.0%). around the deal effective date in SDC. This procedure,
These facts cast some doubt on the notion that capital along with a manual check to eliminate two false
constraints are a primary motivation for the majority of club positives, results in 107 of our deals in SDC being matched
deals, especially before 2006, although it is possible that the to at least one credit facility in DealScan.
degree of capital constraints faced by certain private equity Panel C of Table 9 reports evidence on five measures,
firms may be dependent on deal characteristics that we namely the number of lenders, spread over the London
cannot observe (but may be related to the debt financing Interbank Offered Rate (LIBOR), adjusted spread (the spread
terms that the sponsor(s) can obtain in loan markets). minus the average spread for all LBO debt raised in the
Our evidence on target risk in Table 4 does not point to previous calendar quarter), the maximum debt-to-EBITDA
significant risk differences between targets in club and allowed under the credit facility, and adjusted maximum
sole-sponsored private equity deals. Our two risk mea- debt-to-EBITDA (relative to LBO debt raised in the previous
sures, return volatility and beta, do not differ significantly
between the two categories of targets. As mentioned
in our discussion of Table 4, neither does a measure 24
A caveat to this analysis is that DealScan may not identify all
of operating risk, historical cash flow volatility. We important sources of debt financing, such as mezzanine loans.
calendar quarter). On average, the number of lenders is higher there are unobservable factors that both drive club
in club deals (10.5) than in sole-sponsored private equity formation and can explain the level pricing result, the
deals (7.5). The difference is statistically significant at the 5% 2006 break, and the cross-sectional relation between club
level, and robust to controlling for transaction characteristics premiums and target institutional ownership, even in the
such as size, operating cash flow, and leverage in the absence of any reduction in competition.
transaction (reported in Panel D). The fact that club deals We note that our sample consists exclusively of public-
involve more lenders than sole-sponsored private equity deals to-private transactions and that if one considered private-
do is consistent with the allegation that clubs constrain the to-private transactions as well, the conclusions could be
supply of debt financing for competing bids by aggressively different. In fact, given the role of institutions in
locking up debt financiers. An alternative interpretation is that mitigating the club deal discount in public targets, we
although we control for target size in the regression, the larger might not expect clubs to lead to low prices for private
size of club deals results in more lenders. targets, which tend to have concentrated owners.
The rest of our findings in Panels C and D suggest that Our results on the pricing of club deals differ from
club deals have insignificantly different spreads (raw and those of Boone and Mulherin (2009) and Bargeron,
adjusted, with the univariate differences in Panel C and the Schlingemann, Stulz, and Zutter (2008). Boone and
regressions in Panel D pointing to a small difference in Mulherin (2009) examine the pricing of club deals in the
opposite directions) and insignificantly better maximum 2003–2007 period, but, unlike us, find no evidence of a
debt-to-EBITDA covenants (raw and adjusted) than sole- club deal discount in long-horizon acquisition premiums.
sponsored private equity deals do. Overall, the evidence is Similarly, Bargeron, Schlingemann, Stulz, and Zutter
not strongly supportive of debt certification benefits in club (2008, footnote 4) report but do not tabulate that they
deals, and does not support the notion that club targets are find an insignificant club deal discount using long bid-
riskier in unobservable ways because we would expect period windows in a sample of 43 club deals conducted
greater risk to be reflected in worse financing terms. over 1980–2005. Neither of these studies distinguishes
between prominent and other private equity firms, nor
7. Conclusion accounts for the important structural break in 2006 that
we show (the Bargeron et al. sample ends in 2005).
In this paper, we show that acquisitions by clubs of Broadening our focus to include transactions that SDC
prominent private equity acquirers are priced significantly claims are LBOs but are not in our main sample of LBOs
lower than sole-sponsored private equity transactions and conducted by prominent private equity firms substan-
non-private equity merger and acquisition transactions. tially reduces the magnitudes of the estimated club deal
The differences are economically large: target share- discounts both in the full 1984–2007 period and in the
holders receive approximately 10% less of pre-bid firm pre-2006 period, and renders the long-horizon premium
equity value, or roughly 40% lower premiums, in club deals difference statistically insignificant. Our results combine
compared to sole-sponsored LBOs. These results are robust with those of these authors to suggest that the pre-2006
to controls for observable target characteristics, including club deal discount we report is likely specific to large,
size, Q, measures of risk, and time and industry fixed prominent private equity firms.
effects. While our main results are based on target returns, Finally, our results on pricing in club deals do not rule
we obtain similar results when we consider pricing out the possibility that the transactions are economically
measures based on acquisition multiplies. efficient even if there is room for target shareholders to
We also find that the pricing differences are concen- share more in the surplus. Club deals may, in fact, be
trated before 2006. Indeed, the club discount virtually socially beneficial if their pooling of multiple expert
disappears in 2006 and 2007. This result is potentially opinions helps redeploy target assets to more productive
explained by the fact that the financial media began uses. Several studies of the economic impact of private
expressing concerns about club deals at the end of 2005 equity show substantial long-term benefits to society
and the U.S. Department of Justice started an informal (Gurung and Lerner, 2008). Whether club deals are
inquiry into the practice in 2006. We also find that the different from sole-sponsored LBOs along this dimension
club deal discount is less pronounced in target firms with is an important question for future research.
higher institutional ownership, suggesting that sophisti-
cated institutional investors are able to bargain effectively
with clubs. We find little support for motivations for club
deals based on capital constraints, diversification motives,
or the ability of clubs to obtain favorable debt amounts or Appendix. List of private equity deals
prices. Of course, we are only able to investigate
motivations for club formation that are related to This table lists the 201 deals conducted by prominent
observable target characteristics, and cannot rule in or private equity firms in our sample. Panel A lists club deals
out motivations based on target characteristics that are and Panel B lists sole-sponsor deals. We list a maximum of
unobservable to the econometrician. In particular, the two private equity acquirers, with prominent private
lower pricing of club deals may be an inadvertent equity firms listed first (in order of appearance in Panel A
byproduct of an unobserved benign motivation for club of Table 2). If there are more than two private equity
formation. In addition, because there does not exist a acquirers, the others are denoted ‘‘et al.’’ In the Table A1,
credible instrument in this setting, it is also possible that WCAS abbreviates Welsh, Carson, Anderson, and Stowe.
Table A1
Year Target name Private equity acquirers Year Target name Private equity acquirers
Panel A: Club deals
1985 SCOA Industries Inc. Thomas H. Lee Partners, Drexel 2005 Neiman Marcus Group TPG, Warburg Pincus
Burnham Lambert Inc.
1986 PT Components Inc. Merrill Lynch, Madison Dearborn, et al. 2005 Party City Corp Berkshire Partners, Weston Presidio
Capital
1986 Revco DS Inc. JP Morgan, Salomon Brothers 2005 ShopKo Stores Inc. Sun Capital Partners, Lubert-Adler
1987 Decision Industries WCAS, JH Whitney 2005 SunGard Data Systems KKR, Blackstone, et al.
Corp Inc.
1988 Wickes Cos Inc. Blackstone, Wasserstein Perella 2005 Town & Country Trust Morgan Stanley, Onex
1989 CNW Corp Blackstone, DLJ 2005 Toys ’’R’’ Us Inc. KKR, Bain Capital
1993 Digital WCAS, Sprout Group 2005 UICI Blackstone, Goldman Sachs, et al.
Communications Assoc
1995 Riverwood Clayton, Dubilier & Rice, 1818 Fund, 2006 ADESA Inc. Goldman Sachs, Partheon Capital,
International Corp et al. et al.
1995 Rockefeller Center Goldman Sachs, Exor Group, et al. 2006 Albertsons Inc. Cerberus Capital Management,
Properties Lubert-Adler
1996 Leslie’s Poolmart Inc. Leonard Green & Partners, Hancock Park 2006 ARAMARK Corp Goldman Sachs, Thomas H. Lee
Associates Partners, et al.
1997 Fisher Scientific Intl Inc. Merrill Lynch, Thomas H. Lee Partners, 2006 Biomet Inc. KKR, Blackstone, et al.
et al.
1997 Multicare Cos Inc. TPG, Cypress Group 2006 Direct General Corp TPG, Fremont Partners
1997 Telemundo Group Inc. Apollo Management, Bastion Capital 2006 Education Management Goldman Sachs, Providence Equity
Corp Partners
1998 Centennial Cellular Blackstone, WCAS 2006 Freescale Blackstone, TPG, et al.
Corp Semiconductor Inc.
1998 CompDent Corp GTCR Golder Rauner, TA Associates, 2006 HCA Inc. KKR, Merrill Lynch, et al.
et al.
1998 MedCath Inc. KKR, WCAS 2006 Intergraph Corp TPG,Hellman & Friedman, et al.
1998 Regal Cinemas Inc. KKR, HM Capital Partners 2006 Kinder Morgan Inc. Goldman Sachs, The Carlyle Group
1998 Republic Engineered Blackstone, Veritas Capital Partners 2006 Michaels Stores Inc. Blackstone, Bain Capital
Steels
1999 Berkshire Realty Co Inc. Blackstone, Whitehall Street (Goldman 2006 Open Solutions Inc. The Carlyle Group, Providence
Sachs) Equity Partners
1999 Big Flower Holdings Thomas H. Lee Partners, Evercore 2006 OSI Restaurant Partners Bain Capital, Catterton Partners
Inc. Capital Partners Inc.
1999 Genesis Health TPG, Cypress Group 2006 Petco Animal Supplies TPG, Leonard Green & Partners
Ventures Inc. Inc.
1999 Integrated Circuit Bain Capital, Bear Stearns 2006 Readers Digest Merrill Lynch, GSO Capital Partners
Systems Inc. Association Inc.
1999 Wilmar Industries Inc. JP Morgan (Chase Capital), Sterling Inv. 2006 Sabre Holdings Corp TPG, Silver Lake Partners
Partners
2000 Brookdale Living Fortress Investment Group, Capital Z 2006 Trizec Properties Inc. Blackstone, Brookfield Asset
Partners Management
2000 Petco Animal Supplies TPG, Leonard Green & Partners 2006 Univision TPG, Thomas H. Lee Partners, et al.
Inc. Communications Inc.
2002 Exco Resources Cerberus, Greenhill Capital Partners 2006 West Corp Thomas H. Lee Partners,
Quadrangle Group
2002 National Golf Properties Goldman Sachs, Starwood Capital 2007 Alltel Corp TPG, Goldman Sachs
Inc. Partners
2002 Quintiles Transnational TPG, One Equity Partners 2007 Avaya Inc. TPG, Silver Lake Partners
Corp
2004 AMC Entertainment Inc. Apollo Management, JP Morgan 2007 Dollar General Corp KKR, Goldman Sachs
2004 Metro-Goldwyn-Mayer TPG, Providence Equity Partners, et al. 2007 Eagle Hosp Prop Trust Apollo Management, JF Capital
Inc. Inc. Advisors
2004 PanAmSat Corp KKR, The Carlyle Group, et al. 2007 Kronos Inc. Hellman & Friedman, JMI Equity
2004 Select Medical Corp WCAS, GTCR Golder Rauner 2007 Laureate Education Inc. KKR, Citigroup Private Equity, et al.
2004 Texas Genco Holdings KKR, Blackstone, et al. 2007 Nuveen Investments Madison Dearborn Partners,
Inc. Inc. Citigroup
2005 DoubleClick Inc. Hellman & Friedman, JMI Equity 2007 Topps Co Inc. Madison Dearborn Partners,
Tornante Co.
2005 Linens n Things Inc. Apollo Management, Silver Point 2007 TXU Corp KKR, TPG, et al.
Capital, et al.
Panel B: Sole-sponsor deals
1984 Denny’s Inc. Merrill Lynch 1994 Borden Inc. KKR

1984 Palm Beach Inc. Merrill Lynch 1995 Bruno’s Inc. KKR
1985 Beatrice Companies Inc. KKR 1995 Diagnostek Inc. Warburg Pincus
Table A1 (continued )
Panel B: Sole-sponsor deals
1985 Jack Eckerd Corp Merrill Lynch 1996 Belden & Blake Corp TPG
1985 National Gypsum Co Goldman Sachs 1996 Community Health Systems Inc. Forstmann Little
1985 Storer Communications Inc. KKR 1996 Kinder-Care Learning Centers KKR
1985 UNIROYAL Inc. Clayton, Dubilier & Rice 1996 Syratech Corp Thomas H. Lee Partners
1986 Fruehauf Corp Merrill Lynch 1996 Triad Systems Corp HM Capital Partners
1986 Lear Siegler Inc. Forstmann Little 1997 Amscan Holdings Inc. Goldman Sachs
1986 Midland-Ross Corp Forstmann Little 1997 CommNet Cellular Inc. Blackstone
1986 Owens-Illinois Inc. KKR 1997 Control Data Systems Inc. WCAS
1986 Pandick Inc. Morgan Stanley 1997 Dynatech Corp Clayton, Dubilier & Rice
1986 Safeway Stores Inc. KKR 1997 ERO Inc. HM Capital Partners
1986 Sheller Globe Corp Lehman Brothers Private Equity 1997 Hechinger Co Leonard Green & Partners
1986 Sybron Corp Forstmann Little 1997 LIN Television Corp HM Capital Partners
1987 Becor Western Inc. Goldman Sachs 1997 Melamine Chemicals Inc. KKR
1987 Borg-Warner Corp Merrill Lynch 1997 Smith’s Food & Drug Centers KKR
1987 Burlington Industries Inc. Morgan Stanley 1997 Tuesday Morning Corp Madison Dearborn Partners
1987 Comdata Network Inc. WCAS 1997 Zilog Inc. TPG
1987 Jim Walter Corp KKR 1998 MTL Inc. Apollo Management
1987 Seaman Furniture Co KKR 1998 Rival Co Berkshire Partners
1987 Supermarkets General Corp Merrill Lynch 1998 SFX Broadcasting Inc. HM Capital Partners
1988 American Health Cos Inc. Thomas H. Lee Partners 1999 BancTec Inc. WCAS
1988 Colt Industries Inc. Morgan Stanley 1999 Blount International Inc. Lehman Brothers Private Equity
1988 Cullum Cos Inc. Morgan Stanley 1999 Concentra Managed Care Inc. WCAS
1988 Fort Howard Corp Morgan Stanley 1999 Empi Inc. The Carlyle Group
1988 Hyponex Corp Clayton, Dubilier & Rice 1999 Physicians Specialty Corp TA Associates
1988 Koppers Co Inc. Lehman Brothers Private Equity 1999 Walden Residential Properties HM Capital Partners
1988 Pullman Co Forstmann Little 1999 White Cap Industries Inc. Leonard Green & Partners
1988 RJR Nabisco Inc. KKR 2000 Coinmach Laundry Corp GTCR Golder Rauner
1988 Stanadyne Inc. Forstmann Little 2000 Mark IV Industries Inc. BC Partners
1988 Stop & Shop Cos Inc. KKR 2000 Veterinary Centers of America Leonard Green & Partners
1989 General Nutrition Inc. Thomas H. Lee Partners 2001 Citadel Communications Corp Forstmann Little
1989 Jefferson Smurfit Corp Morgan Stanley 2002 AmeriPath Inc. WCAS
1990 Aristech Chemical Corp Blackstone 2002 Ebenx Inc. WCAS
1990 General Instrument Corp Forstmann Little 2003 FTD Inc. Leonard Green & Partners
1990 Healthco International Inc. HM Capital Partners 2003 Varsity Brands Inc. Leonard Green & Partners
1993 Alden Press Co KKR 2004 Boca Resorts Inc. Blackstone
1993 Preferred Health Care Ltd Warburg Pincus 2004 Extended Stay America Inc. Blackstone
2004 LNR Property Corp Cerberus Capital Management 2006 SITEL Corp Onex
2004 Prime Hospitality Corp Blackstone 2006 SourceCorp Inc. Apollo Management
2004 Sola International Inc. EQT Partners 2006 Sports Authority Inc. Leonard Green & Partners
2004 US Oncology Inc. WCAS 2006 TransMontaigne Inc. Morgan Stanley
2005 Amli Residential Ppty Trust Morgan Stanley 2006 Water Pik Technologies Inc. The Carlyle Group
2005 Insight Communications Co The Carlyle Group 2006 Yankee Candle Co Inc. Madison Dearborn Partners
2005 Instinet Group Inc. Silver Lake Partners 2007 Archstone-Smith Trust Lehman Brothers Private Equity
2005 La Quinta Corp Blackstone 2007 Bausch & Lomb Inc. Warburg Pincus
2005 Metals USA Inc. Apollo Management 2007 Catalina Marketing Corp Hellman & Friedman
2005 SERENA Software Inc. Silver Lake Partners 2007 CDW Corp Madison Dearborn Partners
2005 Specialty Laboratories Inc. WCAS 2007 Ceridian Corp Thomas H. Lee Partners
2005 SS&C Technologies Inc. The Carlyle Group 2007 Claire’s Stores Inc. Apollo Management
2005 Worldwide Rest Concepts Inc. Pacific Equity Partners 2007 Crescent Real Estate Equities Morgan Stanley
2005 Wyndham International Inc. Blackstone 2007 EGL Inc. Apollo Management
2006 Aleris International Inc. TPG 2007 Equity Inns Inc. Goldman Sachs
2006 Burlington Coat Factory Bain Capital 2007 First Data Corp KKR
2006 CarrAmerica Realty Corp Blackstone 2007 Florida East Coast Inds Inc. Fortress Investment Group
2006 Columbia Equity Trust Inc. JP Morgan 2007 Guitar Center Inc. Bain Capital
2006 Encore Medical Corp Blackstone 2007 Hilton Hotels Corp Blackstone
2006 Equity Office Properties Trust Blackstone 2007 Innkeepers USA Trust Apollo Management
2006 Glenborough Realty Trust Inc. Morgan Stanley 2007 Interpool Inc. Fortress Investment Group
2006 Jacuzzi Brands Inc. Apollo Management 2007 ServiceMaster Co Clayton, Dubilier & Rice
2006 Laserscope Inc. Warburg Pincus 2007 Smart & Final Inc. Apollo Management
2006 Marsh Supermarkets Sun Capital Partners 2007 Synagro Technologies Inc. The Carlyle Group
2006 MeriStar Hospitality Corp Blackstone 2007 United Surgical Partners Intl WCAS
2006 RailAmerica Inc. Fortress Investment Group 2007 USI Holdings Corp Goldman Sachs
2006 Realogy Corp Apollo Management
References Gurung, A., Lerner, J., 2008. The global economic impact of private
equity. World Economic Forum.
Hendricks, K., Porter, R., 1992. Joint bidding in federal OCS auctions.
Abadie, A., Imbens, G., 2007. Bias corrected matching estimators for average
American Economic Review 82, 506–511.
treatment effects. Unpublished Working Paper, Harvard University.
Hochberg, Y., Ljungqvist, A., Lu, Y., 2009. Networking as a barrier to entry
Axelson, U., Jenkinson, T., Strömberg, P., Weisbach, M., 2007. Leverage
and the competitive supply of venture capital. Journal of Finance 65,
and pricing in buyouts: an empirical analysis. Unpublished Working
829–859.
Paper, Swedish Institute for Financial Research.
Ivashina, V., Kovner, A., 2007. The private equity advantage: private
Bargeron, L., Schlingemann, F., Stulz, R., Zutter, C., 2008. Why do private
acquirers pay so little compared to public acquirers? Journal of equity firms and relationship banking. Unpublished Working Paper,
Financial Economics 89, 375–390. Harvard University.
Bailey, E., 2007. Are private equity consortia anticompetitive? The Jensen, M., 1986. Agency costs of free cash flow, corporate finance, and
economics of club bidding. Antitrust Source. takeovers. American Economic Review 76, 323–329.
Boone, A., Mulherin, H., 2009. Do private equity consortiums facilitate Kaplan, S., 1989a. The effects of management buyouts on operating
collusion in takeover bidding? Unpublished Working Paper, Uni- performance and value. Journal of Financial Economics 24,
versity of Kansas. 217–254.
Brander, J., Amit, R., Antweiler, W., 2002. Venture capital syndication: Kaplan, S., 1989b. Management buyouts: evidence on taxes as a source of
improved venture selection vs. the value-added hypothesis. Journal value. Journal of Finance 44, 611–632.
of Economics and Management Strategy 11, 423–452. Kaplan, S., Ruback, R., 1995. The valuation of cash flow forecasts: an
Comment, R., Schwert, G., 1995. Poison or placebo? evidence on the empirical analysis. Journal of Finance 50, 1059–1093.
deterrence and wealth effects of modern antitakeover measures. Kaplan, S., Stein, J., 1993. The evolution of buyout pricing and financial
Journal of Financial Economics 39, 3–43. structure in the 1980s. Quarterly Journal of Economics 108,
Cramton, P., Schwartz, J., 2000. Collusive bidding: lessons from the FCC 313–357.
spectrum auctions. Journal of Regulatory Economics 17, 229–252. Kaplan, S., Strömberg, P., 2009. Leveraged buyouts and private equity.
DeAngelo, L., 1986. Accounting numbers as market valuation substi- Journal of Economic Perspectives 23, 121–146.
tutes: a study of management buyouts of public stockholders. The Kaplan, S., Zingales, L., 1997. Do investment-cash flow sensitivities
Accounting Review 61, 400–420. provide useful measures of financing constraints? Quarterly Journal
DeAngelo, H., DeAngelo, L., Rice, E., 1984. Going private: minority of Economics 112, 169–215.
freezeouts and stockholder wealth. Journal of Law and Economics 27, Lang, L., Stulz, R., Walking, R., 1989. Managerial performance, Tobin’s Q,
367–401. and the gains from successful tender offers. Journal of Financial
Demiroglu, C., James, C., 2010. The role of private equity group Economics 24, 137–154.
reputation in LBO financing. Journal of Financial Economics 96,
Lerner, J., 1994. The syndication of venture capital investments. Financial
306–330.
Management 23, 16–27.
Du, Q., 2008. Birds of a feather or celebrating differences? the formation
Marquez, R., Singh, R., 2009. The economics of club bidding in private
and impact of venture capital syndication. Unpublished Working
equity. Unpublished Working Paper, Boston University.
Paper, University of British Columbia.
Officer, M., 2003. Termination fees in mergers and acquisitions. Journal
Gaspar, J., Massa, M., Matos, P., 2005. Shareholder investment horizons
of Financial Economics 69, 431–467.
and the market for corporate control. Journal of Financial Economics
76, 135–165. Officer, M., 2007. The price of corporate liquidity: acquisition discounts
Gompers, P., Ishii, J., Metrick, A., 2003. Corporate governance and equity for unlisted targets. Journal of Financial Economics 83, 571–598.
prices. Quarterly Journal of Economics 118, 107–155. Perry, S., Williams, T., 1994. Earnings management preceding
Gompers, P., Lerner, J., 2000. Money chasing deals? The impact of fund management buyout offers. Journal of Accounting and Economics
inflows on the valuation of private equity investments. Journal of 18, 157–179.
Financial Economics 55, 281–325. Phalippou, L., 2009. Beware of venturing into private equity. Journal of
Graham, D., Marshall, R., 1989. Collusive bidder behavior at single-object Economic Perspectives 23, 147–166.
second-price and english auctions. Journal of Political Economy 95, Ritter, J., 2008. Forensic finance. Journal of Economic Perspectives 22,
1217–1239. 127–147.
Guo, S., Hotchkiss, E., Song, W., 2009. Do buyouts (still) create value? Schwert, G., 1996. Markup pricing in mergers and acquisitions. Journal of
Journal of Finance, forthcoming. Financial Economics 41, 153–219.

Club Deals in Leveraged Buyouts (2010)

Uploaded by

Copyright:

Available Formats

Club Deals in Leveraged Buyouts (2010)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Club Deals in Leveraged Buyouts (2010)

Uploaded by

Copyright:

Available Formats

Journal of Financial Economics 98 (2010) 214–240

Contents lists available at ScienceDirect

Journal of Financial Economics

Club deals in leveraged buyouts$

1. Introduction rose from approximately $30 billion in 2001 to over $450

Club deals Sole PE deals Private bidder Public bidder

1984 33 17,359.1 0 0.0 2 882.2 13 5,809.7 18 10,667.2

Panel A: Number of deals

1 2 Kohlberg Kravis Roberts 26 15 11

Panel B: Cross-participation matrix

Panel B: Cross-participation matrix

4. Returns to target shareholders e~ it ¼ rit ða~ i þ b~ i rmt Þ,

Buy-and-hold abnormal returns

Cumulative abnormal returns

Number observations 70 128 902 2,786

M.S. Ofﬁcer et al. / Journal of Financial Economics 98 (2010) 214–240

Panel A: Regressions including year ﬁxed effects

M.S. Ofﬁcer et al. / Journal of Financial Economics 98 (2010) 214–240

M.S. Ofﬁcer et al. / Journal of Financial Economics 98 (2010) 214–240

Panel A: Multivariate regressions explaining target returns

CAR3 Markup Premium CAR3 Markup Premium CAR3 Markup Premium

Propensity-score matching Abadie-Imbens matching

CAR 3 9.75 6.03

Markup BHAR 12.11 7.28

Premium BHAR 14.74 11.86

example, two-tier offers (which are likely part of our Table 7

discounts for private equity sponsored deals as the EV/sales EV/EBITDA

Panel A: Frequency distributions

Sole PE % Club % All %

Panel B: Probit regressions explaining post-announcement competition

Sample All PEI5 All

Club 0.10 0.20 0.33

Panel A: Total ﬁnancing commitments relative to total funds needed

Sole PE Club All

Mean (%) 99.4 101.6 100.4

Panel B: Equity commitments relative to total ﬁnancing commitments

Sole PE Club All

Mean (%) 33.7 38.2 35.6

Panel C: Debt ﬁnancing terms

Sole PE (mean) 7.5 289 32 6.1 0.2

Panel D: Regressions explaining debt ﬁnancing terms

Club 2.76 9.15 12.61 0.29 0.52

Panel A: Club deals

Panel B: Sole-sponsor deals

1984 Denny’s Inc. Merrill Lynch 1994 Borden Inc. KKR

Panel B: Sole-sponsor deals

You might also like