会议议程Journal of Financial Economics73(2004)201–228
Firm size and the gains from acquisitions$
Sara B.Moeller a,Frederik P.Schlingemann b,Ren!e M.Stulz c,*
a Cox School of Business,Southern Methodist University,Dallas,TX75275,USA
b Katz Graduate School of Business,University of Pittsburgh,Pittsburgh,PA15260,USA
c Department of Finance,Fisher College of Business,The Ohio State University,806Fisher Hall,
郑州文庙2100Neil Avenue,Columbus,OH43210,USA
Received12February2003;accepted22July2003
Abstract
We examine a sample of12,023acquisitions by publicfirms from1980to2001.The equally weighted abnormal announcement return is1.1%,but acquiring-firm shareholders lo$25.2 million on average upon announcement.This disparity suggests the existence of a size effect in acquisition announcement returns.The announcement return for acquiring-firm shareholders is roughly two percentage points higher for small acquirers irrespective of the form of financing and whether the acquiredfirm is public or private.The size effect is robust tofirm and deal characteristics,and it is not reverd over time.
r2004Elvier B.V.All rights rerved.
votes
JEL classification:G31;G32;G34
Keywords:Acquisitions;Bidder;Size effect;Organizational form
英语心得1.Introduction
In this paper,we examine the gains to shareholders offirms that announce acquisitions of publicfirms,privatefirms,or subsidiaries of otherfirms.We consider the different types of acquisitions together since corporations making such $We thank Evrim Akdogu,Harry DeAngelo,Hemang Desai,Eugene Fama,David Hirshleifer,Cliff Holderness,Bengt Holmstrom,Jin-Lung(Jim)Hsieh,Paul Malatesta,Jeffry Netter,Bill Schwert,Mike Stegemoller,Vish Viswanathan,Ralph Walkling,minar participants at Boston College and the Federal Rerve Bank of New York,and an anonymous referee for uful comments.
*Corresponding author.Tel.:+1-614-292-1970;fax:+1-614-292-2359.
E-mail address:stulz@cob.osu.edu(R.M.Stulz).
0304-405X/$-e front matter r2004Elvier B.V.All rights rerved.
doi:10.1016/j.jfineco.2003.07.002
acquisitions could be acquiring similar asts.1Typically,acquisitions are sizable investments for the firms that undertake them.We form a sample of all such purchas over $1million by public firms from 1980to 2001recorded by the Securities Data Corporation.After imposing some additional sampli
60甲子年表ng criteria,we obtain a sample of 12,023acquisitions.Such a comprehensive sample has not been studied before.The equally-weighted average announcement return for acquiring-firm shareholders in our sample is 1.1%,reprenting a gain of $5.61per $100spent on acquisitions.If the capital markets’asssment is unbiad,this gain reprents the economic benefit of the acquisition for the shareholders of the acquiring firm together with the stock-price impact of other information relead or inferred by investors when firms make acquisition announcements.悬铃花
The equally-weighted average announcement return implies that the wealth of acquiring-firm shareholders increas when acquisitions are announced.Strikingly,however,the average dollar change in the wealth of acquiring-firm shareholders when acquisition announcements are made is negative.From 1980to 2001,the sample firms spent roughly $3.4trillion on acquisitions and the wealth of the shareholders of the firms fell by $303billion dollars (in 2001dollars),for a dollar abnormal return,defined in Malatesta (1983)as the abnormal return times the firm’s equity capitalization cumulated over the event window,of À$25.2million per acquisition.The dollar abnormal return can differ in sign from the percentage abnormal return if the percentage abnormal return differs in sign for large and small firms.This is the ca here.We define small firms in a given year to be firms who capitalization falls below the 25th percentile of NYSE firms that year.Acquisitions by
small firms are profitable for their shareholders,but the firms make small acquisitions with small dollar gains.Large firms make large acquisitions that result in large dollar loss.Acquisitions thus result in loss for shareholders in the aggregate becau the loss incurred by large firms are much larger than the gains realized by small firms.Roughly,shareholders from small firms earn $9billion from the acquisitions made during the period 1980-2001,whereas the shareholders from large firms lo $312billion.Though it is common to focus on equally-weighted returns in event studies,it follows from the numbers that value-weighted returns lead to a different asssment of the profitability of acquisitions.The value-weighted return is À1.18%.
After documenting that small firms are good acquirers and large firms are not,we examine possible explanations for this size effect,defined as the difference between the abnormal returns of small acquirers and large acquirers.First,roughly one quarter of the firms acquiring public firms are small whereas half of the firms acquiring private firms are small.If acquiring private firms is more profitable than acquiring public firms,this could explain the size effect.Fuller et al.(2002)show for 1
酒入Kaplan and Weisbach (1992)have a sample of 282large acquisitions.They find that almost 44%of the acquisitions are subquently divested.216of their acquisitions are acquisitions of public companies.The acquired asts are then spun off in some cas and acquired by other companies i
n most cas.Hence,in their sample,the same asts most likely are first organized as a public firm and then as a division.In this paper,we u the term subsidiary acquisition to denote the acquisition of a subsidiary,division,or branch.S.B.Moeller et al./Journal of Financial Economics 73(2004)201–228
202
S.B.Moeller et al./Journal of Financial Economics73(2004)201–228203 a sample offirms that makefive or more acquisitions in the1990s that abnormal returns are higher forfirms acquiring privatefirms or subsidiaries than forfirms acquiring publicfirms.Second,smallfirms are more likely to pay for acquisitions with cash than with equity.Travlos(1987)and others show that acquisitions of publicfirms paid for with equity are accompanied by lower announcement returns. However,Chang(1998)and Fuller et al.(2002)show that acquisitions of private firms paid for with equity do not have lower announcement returns than private acquisitions paid for with cash.Third,small and large acquirers have different characteristics.The literature has shown that a number of acquiring-firm and deal characteristics are related to announcement returns for public-firm acquisitions.For instance,Lang et al.(1991)and Servaes(1991)show that high q bidders have higher announcement abnormal returns for tender offer acquisitions and public-firm acquisitions,respectivel
y,and Maloney et al.(1993)find that bidders with higher leverage have higher abnormal returns.Wefind that controlling for a wide variety of acquiring-firm and deal characteristics does not alter the size effect.In all of our regressions,the estimate of the size effect is positive and significantly different from zero at the1%probability level.
A number of explanations have been offered for why the stock price offirms announcing an acquisition can be negative.Roll(1986)hypothesizes that managers of biddingfirms may suffer from hubris,so they overpay.Travlos(1987)points out thatfirms with poor returns generally pay with equity,and Myers and Majluf(1984) show thatfirms that issue equity signal that the market overvalues their asts in place(the equity signaling hypothesis).A related hypothesis,formalized by McCardle and Viswanathan(1994)and Jovanovic and Braguinsky(2002),is that firms make acquisitions when they have exhausted their internal growth opportunities(the growth opportunities signaling hypothesis).Jenn(1986)argues that empire-building managements would rather make acquisitions than increa payouts to shareholders(the free cashflow hypothesis).Recently,Dong et al.(2002) show thatfirms with higher valuations have wor announcement returns.This could be becau highly valued acquirers communicate to the market that the high valuations are not warranted by fundamentals,perhaps becau they are under-taking efforts to acquire less overvalued asts with
more overvalued equity(the overvaluation hypothesis).2Finally,Mitchell et al.(2004)show that there is a price pressure effect on the stock price of the bidder for acquisitions paid for with equity becau of the activities of arbitrageurs(the arbitrageur hypothesis).
For the hypothes to explain the size effect for some or all types of acquisitions, they have to be more pertinent for largefirms than for smallfirms.This is not implausible.Generally,the incentives of managers in smallfirms are better aligned with tho of shareholders than is the ca in largefirms.In particular,Demtz and Lehn(1985)find that managers in smallfirms typically have morefirm ownership than managers in largefirms.Managers of largefirms might be more prone to hubris,perhaps becau they are more important socially,have succeeded in growing 2Shleifer and Vishny(2003)provide a model in which overvaluedfirmsfind it advantageous to acquire less overvaluedfirms to lock in real asts,but they make no predictions about abnormal returns.
the firm,or simply face fewer obstacles in making acquisitions becau their firm has more resources.A firm may be large becau its equity is highly valued,so a large firm is more likely to be overvalued.A firm that is further along in its lifecycle might be more likely to be large and to have exhausted its growth opportunities.Agency costs of free cash flow occur when a firm no longer has growth opportunities,which could be more likely for large firms than for small firms.Finally,arbitrageur
s are unlikely to u their resources for a merger when the acquirer is a small firm becau it will be too difficult and costly to establish large short positions.
We investigate whether the hypothes are helpful in understanding the size effect.We provide evidence that managers of large firms pay more for acquisitions.The premium paid increas with firm size after controlling for firm and deal characteristics.Large firms are also more likely to complete an offer.This is consistent with hubris being more of a problem for large firms.We find that the combined dollar return of the acquired and acquiring firms for acquisitions of public firms is positive and significant for small firms but significantly negative for large firms.In other words,there are no dollar synergy gains for acquisitions by large firms given how synergy gains are typically computed (following the method propod by Bradley et al.1988),but there are dollar synergy gains for acquisitions by small firms.Percentage synergy returns are positive for acquisitions by large firms as well as by small firms,but they are significantly higher for acquisitions by small firms.Of cour,the synergy gain estimate for acquisitions by large firms could be made negative by the adver information revealed about the acquirer through the acquisition announcement rather than by the adver impact on shareholder wealth of the acquisition itlf,although it is not clear why large acquirers reveal relatively more adver information than do small acquirers.
We also provide evidence that is inconsistent with the overvaluation hypothesis.In contrast to the market value of a firm’s equity,the book value of a firm’s asts is unlikely to be correlated with the overvaluation of the firm’s stock price.Conquently,if the size effect is due to the fact that large firms tend to be overvalued,it should disappear when we u the book value of the firm’s asts as a size measure.Nonetheless,we find that the size effect holds when we u the book value of a firm’s asts instead of the firm’s market value of equity.Though the outcome of the acquisitions by large firms is consistent with the existence of agency costs of managerial discretion,there is little support for the free cash flow hypothesis.Finally,we investigate the hypothesis that the market makes systematic mistakes in evaluating acquisitions that it rectifies over time.In this ca,acquisitions by small firms would be followed by negative abnormal returns and acquisitions by large firms would be followed by positive abnormal returns.This explanation cannot account for the size effect.The market ems fairly efficient in incorporating the information conveyed by acquisition announcements in the stock price.
The paper is organized as follows.In Section 2we describe our sample and document that abnormal returns for acquisition announcements are significantly positive and negatively correlated with firm size.In Section 3,we demonstrate that the size effect is robust to firm and deal characteristics.In Section 4,we investigate possible explanations for the size effect.We conclude in Section 5.
S.B.Moeller et al./Journal of Financial Economics 73(2004)201–228
204
手背痣
S.B.Moeller et al./Journal of Financial Economics73(2004)201–228205 2.Announcement returns for successful acquisitions
To estimate the shareholder gains from acquisitions,we consider acquisition announcements that are successful and result in a completed transaction.In Sections 3and4,we include unsuccessful acquisition announcements to investigate whether this focus introduces a bias in our analysis andfind that it does not.Wefirst describe our sample and then estimate the gains to shareholders.
2.1.The sample
The sample of acquisitions comes from the Securities Data Company’s(SDC) U.S.Mergers and Acquisitions Databa.We lect domestic mergers and acquisitions with announcement dates between1980and2001.We consider only acquisitions in which acquiringfirms end up with all the shares of the acquiredfirm or subsidiary,and we require the acquiringfirm to control less than50%of the shares of the targetfirm before the announcement.We further require that(1)the transaction is co
mpleted,(2)the deal value is greater than$1million,(3)a public or private U.S.firm or a non-public subsidiary of a public or privatefirm are acquired, and(4)the acquirer is a publicfirm listed on the Center for Rearch in Security Prices(CRSP)and Compustat during the event window.Deal value is defined by SDC as the total value of consideration paid by the acquirer,excluding fees and expens.After collecting the acquisitions,we eliminate tho in which the deal value relative to the market value of the acquirer is less than1%.The market value of the acquirer is defined as the sum of the market value of equity,long-term debt, debt in current liabilities,and the liquidating value of preferred stock.We also require that the number of days between the announcement and completion dates is between zero and one thousand.
Our requirements yield a sample of12,023successful offers.Slightly more than half of the acquisitions are by largefirms,which we define as tho with a market capitalization above the25th percentile of NYSEfirms in the year in which the acquisition is announced.Table1shows the number of acquisitions by year.The number of acquisitions does not increa monotonically through time:it falls in1990 and in recent years.The number of acquisitions in the1990s is dramatically larger than in the1980s.In our tests,we will often u time dummies to take into account the changes.Interestingly,though there are normally fewer acquisitions by small firms than by largefirms,this is not the ca for most of the1990s.
2.2.The gains to acquiring-firm shareholders
The most traditional way to evaluate bidder returns is to estimate abnormal percentage returns with standard event study methods(following Brown and Warner,1985).We estimate the abnormal returns over the three-day event window (À1,+1)using market model benchmark returns with the CRSP equally-weighted index returns.The parameters for the market model are estimated over the(À205,À6)interval,and the p-values are estimated using the time-ries and cross-ctional
