American Law & Economics Association Annual Meetings Year Paper Does Shareholder Voting on Acquisitions Matter? Ehud Kamar University of Southern California This working paper site is hosted by The Berkeley Electronic Press (bepress) and may not be commercially reproduced without the publisher’s permission. http://law.bepress.com/alea/16th/art64 Copyright c 2006 by the author.
62
Embed
American Law & Economics Association Annual Meetings€¦ · Association Annual Meetings Year 2006 Paper 64 Does Shareholder Voting on Acquisitions Matter? Ehud Kamar University of
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
American Law & EconomicsAssociation Annual MeetingsYear Paper
Does Shareholder Voting on Acquisitions
Matter?
Ehud KamarUniversity of Southern California
This working paper site is hosted by The Berkeley Electronic Press (bepress) and may not becommercially reproduced without the publisher’s permission.
Very preliminary results from data collected thus far. Please do not cite. Instead, ask me
for a current draft. Comments are welcome.
Hosted by The Berkeley Electronic Press
2
Abstract
I examine the effects of shareholder voting on acquisitions by comparing acquisitions that
require acquirer shareholder approval to acquisitions that do not in a sample of 1,969
acquisitions announced between 1995 and 2003. I find that shareholder approval is less
likely to be required in acquisitions that can be completed faster without it, and that these
acquisitions take longer to complete when approval is required. By contrast, I do not find
that acquisitions are less likely to require approval when opposition is likely, or that the
approval requirement is related to announcement returns or premiums. Finally, I find that
acquisitions requiring shareholder approval are less likely to be completed. This relation,
however, is not affected by announcement returns, the presence of guaranteed shareholder
support, or the type of majority required, suggesting that opposition is not the reason for
the lower completion rate.
http://law.bepress.com/alea/16th/art64
I. Introduction
Modern corporate law affords shareholders only a handful of opportunities to make
decisions for the firm. Apart from annual elections for the board of directors, which
seldom result in management changes, the matters requiring shareholder approval are
charter amendments, sales of all or substantially all assets, liquidations, and business
combinations. Even business combinations do not always require shareholder approval.
The approval requirement is universal only for target shareholders. Acquirer shareholder
approval, by contrast, can generally be avoided by issuing less than 20% of the acquirer
common stock as consideration.
The non-universal applicability of the approval requirement offers a unique
opportunity to evaluate the effectiveness of shareholder voting by comparing acquisitions
that require approval to acquisitions that do not. Any finding this comparison yields will
have important policy implications. If voting protects shareholders, it might be extended
to all large acquisitions regardless of the amount of stock issued as consideration, as some
states and foreign jurisdictions require.1 Conversely, if voting does not protect
shareholders, it should be reconsidered.
1 See California Corporations Code, Section 1201; Ohio General Corporation law, Section 1701.83. New Jersey law requires approval by a majority of the votes cast or, in the case of a firm organized prior to 1969, two-third of the votes cast, of acquisitions in which the number of voting shares or nonvoting shares that participate without limitation in distributions issued at closing or would be issued on conversion or exercise of securities issued at closing exceeds 40% of the number of outstanding shares. The Toronto Stock Exchange requires listed acquirers to obtain shareholder approval by a majority of the votes outstanding for acquisitions involving the issuance of 25% or more of the outstanding shares of any security. See Toronto Stock Exchange Company Manual, Section 611. The London Stock Exchange requires listed firms to obtain shareholder approval for any transaction worth at least 25% of the firm value by any one of four alternative calculations. See United Kingdom Listing Authority Listing Rules, Section 10.
Hosted by The Berkeley Electronic Press
4
I address this question by comparing acquisitions that require acquirer shareholder
approval to acquisitions that do not using 1,969 acquisitions announced between 1995 and
2003. The analysis focuses on four related questions. First, I examine whether acquirers
avoid shareholder approval and, if so, whether they do this in acquisitions that shareholders
are likely to oppose. Second, I examine whether acquisitions that need acquirer
shareholder approval are less likely to be completed. If voting is effective and managers
do not perfectly predict shareholder reaction, poorly received acquisitions should be
withdrawn or voted down. Third, I examine whether the acquirer announcement returns
are higher when approval is required. If voting is effective and managers at least
imperfectly predict shareholder reaction, they should announce acquisitions more
acceptable to acquirer shareholders when approval is required. Fourth, I examine whether
acquirers pay less in acquisitions that need shareholder approval. Here too, if voting is
effective and managers at least imperfectly predict shareholder reaction, they should
propose acquisitions at lower premiums when shareholder approval is required.
I find that shareholder approval is less likely to be required in acquisitions that can
be completed faster without it and that these acquisitions take longer to complete when
approval is required. By contrast, I do not find that acquisitions are less likely to require
approval when opposition is likely (acquisitions by actual or rumored targets and
acquisitions by acquirers with substantial public pension fund ownership), or that the
approval requirement is related to announcement returns or premiums. Finally, I find that
acquisitions requiring shareholder approval are less likely to be completed. This relation,
however, is not affected by announcement returns, the presence of guaranteed shareholder
http://law.bepress.com/alea/16th/art64
5
support, or the type of majority required, as would be expected if shareholder opposition
was the reason for the lower completion rate.
The article proceeds as follows. Part II reviews the literature. Part III explains the
mechanics of acquisitions. Part IV describes the data. Part V discusses the methodology
and the results. Part VI concludes.
II. Related Literature
The efficacy of shareholder voting on acquisitions relates to two strands in the
corporate governance literature. The first strand examines the agency costs that distort
acquisition decisions. Roll (1986) and Black (1989) argue that managers are prone to
overpaying or otherwise making ill-advised acquisitions because they derive psychic and
financial gains from expanding their firms. Avery, Chevalier, and Schaefer (1998), Bliss
and Rosen (2001), Grinstein and Hribar (2004), Bebchuk and Grinstein (2005), and
Harford and Li (2006) find that managers benefit from acquisitions. Harford (1999),
Datta, Iskandar-Datta, and Raman (2001), Malmendier and Tate (2005a, 2005b), and
Moeller, Schlingeman, and Stulz (2005) find that these distorted incentives cause losses to
acquirer shareholders.
The losses to acquirer shareholders contrast with the gains to target shareholders.
While studies typically find zero or negative announcement returns to the former, they find
large and positive returns to the latter. Dent (1986) attributes this difference to an
imbalance in the protection afforded to the two groups of shareholders. Target
shareholders are well protected. They can sue the board when it mishandles the sale, they
can veto the sale by not tendering their stock or voting against it, and federal law shelters
them from coercive tender offers. Acquirer shareholders are far less protected. Courts
Hosted by The Berkeley Electronic Press
6
rarely scrutinize acquisition decisions, most acquisitions do not require shareholder
approval, and federal law protects acquirer shareholders only when they vote.
Another related strand in the literature is the debate on the efficacy of voting as a
solution to agency costs in the firm. Recent work in corporate governance focuses on
whether stock ownership by institutional investors warrants greater reliance on voting as a
check on management. Bebchuk (2005, 2005a) argues for facilitating proxy fights to
replace directors and allowing shareholders to initiate governance changes. The American
Bar Association (2006) proposes to allow shareholders to increase the majority of their
votes needed for electing directors. In a similar vein, Etter (2006) report, some lawmakers
propose requiring shareholder approval for any executive compensation coinciding with
the sale or purchase of assets, and some institutional investors push to require shareholder
approval of executive compensation generally.
Others take a dimmer view of shareholder voting. Anabtawi (2006) claims that the
largest modern shareholders have substantial private interests that conflict with
maximizing overall shareholder value. Bainbridge (2006) argues that the limited
shareholder voting regime has withstood the test of time and ought to be preserved. Strine
(2006) proposes expanding shareholder influence on director election while not expanding
shareholder voting on governance matters and business decisions.
The merit of these proposals depends on whether shareholders in general and
institutional investors in particular vote effectively. A number of empirical studies suggest
that they do. Brickley, Lease, and Smith (1994) and Bethel and Gillan (2002) find that
institutional investors are more likely than other shareholders to vote against management.
Gordon and Pound (1993) find that nonbinding shareholder proposals to improve
http://law.bepress.com/alea/16th/art64
7
governance receive more support when they are sponsored by large institutional investors
or active dissidents, when they restore shareholder voting rights, when they target poorly
performing or poorly governed firms, when ownership is concentrated among institutional
investors or unaffiliated blockholders, and when ownership by directors and officers is
small. Morgan and Poulsen (2001) and Bethel and Gillan (2002) find that shareholders are
less likely to approve dilutive equity-based compensation plans. Balachandran, Joos, and
Weber (2004) find that poorly performing firms and poorly governed firms tend to adopt
equity-based compensation plans without seeking shareholder approval, and subsequently
perform worse than similar firms that adopt plans with approval.
On the other hand, Christofferson, Géczy, Musto, and Reed (2005), Crawford
(2005), Hu and Black (2006), and Kahan and Rock (2006) report that equity lending by
institutional investors transfers votes to transient shareholders free of charge. This
suggests that transient shareholders, rather than long-term shareholders, determine voting
outcomes in firms with high institutional ownership.
The writing on voting by acquirer shareholders is part of this literature. Legal
commentators disagree on the desirability of shareholder approval of acquisitions. Coffee
(1984) proposes that all acquisitions receive shareholder approval. Dent (1986) argues that
requiring approval would deter bidders by delaying acquisitions, and that collective action
problems would render the approval meaningless. Black (1989) suggests requiring
acquirer shareholder approval only when the need for approval by target shareholders
delays the acquisition anyway.
In light of the importance of acquisitions to firm value, it is surprising how little
evidence exists on acquirer shareholder voting. One study is Hamermesh (2003). He
Hosted by The Berkeley Electronic Press
8
reports a negative relation between acquisition premiums and the presence of acquirer
shareholder vote in a sample of 97 stock-for-stock mergers announced in the years 1999
through 2002 and reported on the Securities and Exchange Commission’s Electronic Data
Gathering, Analysis, and Retrieval website (EDGAR). He cautions against viewing his
finding as evidence that acquirer shareholder voting affects premiums due to the small
sample size and its limited chronological coverage. Moreover, his study does not describe
the criteria for including acquisitions in the sample, and controls only for two factors other
than the presence of shareholder voting that can affect premiums — the percentage of
directors of the combined firm who were directors or officers of the target, and the ratio of
target revenue to acquirer revenue.
Another study is Burch, Morgan, and Wolf (2004). They examine 209 acquisitions
by public firms excluding utilities that were brought to acquirer shareholder approval in the
years 1990 through 2000 and are included in the voting results database maintained by
Investor Responsibility Research Center. Approximately 92% of the acquisitions in their
sample involved only stock consideration, and approximately 43% required approval by a
majority of the votes outstanding. They find that the percentage of favorable votes cast is
positively related to acquirer announcement returns and transaction characteristics. They
also find, however, that acquisitions are always approved with minimal opposition. This
suggests that acquirers are concerned mainly with collecting enough shareholder proxies to
satisfy quorum requirements (and, in some cases, supermajority requirements), rather than
with overcoming opposition, and stop soliciting shareholder proxies once they have
enough votes.
http://law.bepress.com/alea/16th/art64
9
Finally, Slovin, Sushka, and Polonchek (2005) study the relation between the
method of payment and returns in acquisitions of corporate divisions. They do not find
significantly different acquirer returns in the 32 acquisitions (out of 311 in their sample)
that required acquirer shareholder approval.
III. The Mechanics of Corporate Acquisitions
Corporate acquisitions can be structured as mergers, asset purchases, or stock
purchases. In each case, acquirers can readily avoid the requirement under the laws of
most states to obtain shareholder approval. By contrast, to avoid the need for shareholder
approval under the rules of the three national stock exchanges, acquirers must not issue in
the transaction 20% or more of their outstanding common stock.
State laws provide the basic merger structure, known as a statutory merger. A
statutory merger transfers the assets and the liabilities of the target to the acquirer. When
the target is public, the merger sometimes follows as a second step after a tender offer that
concentrates a majority of the target stock in the hands of the acquirer. Statutory mergers
require target shareholder approval when the acquirer owns less than 90% of the target
stock before the transaction, and require acquirer shareholder approval when the acquirer
issues a substantial amount of stock (between 20% and 25%, depending on the state).2 The
approval typically requires a majority or a supermajority of the votes outstanding.
In most states, however, acquirers can avoid the approval requirement by
structuring the acquisition as a triangular merger, in which the target merges with a
2 See, for example, Delaware General Corporation Law, Sections 251(f), 253(a). New York requires approval of statutory mergers by acquirer shareholders no matter how small the stock issuance. See New York Business Corporation Law, Section 903.
Hosted by The Berkeley Electronic Press
10
wholly-owned subsidiary of the acquirer, rather than with the acquirer itself. Triangular
mergers are different from statutory mergers in three additional respects. First, triangular
mergers keep the assets of the acquirer away from the reach of target creditors by
maintaining the separate corporate existence of the target or the subsidiary with which it
merged. Second, triangular mergers obviate the need to obtain the consent of target
creditors when these creditors hold a contractual right to block mergers in which the target
disappears. Third, triangular mergers in which part of the consideration is not stock must
meet more onerous requirements to qualify as tax-free reorganizations.
An asset purchase transfers specific assets and specific liabilities of the target to the
acquirer. State laws do not require acquirer shareholder approval of asset purchases, and
require targets to obtain shareholder approval only when they sell all or substantially all of
their assets. A stock purchase transfers some or all of the target stock to the acquirer.
State laws do not require acquirers to obtain shareholder approval for stock purchases, but
require approval by every target shareholder whose equity stake is transferred.
While acquirers can readily avoid shareholder approval under the laws of most
states, they have less freedom under the rules of the New York Stock Exchange, Nasdaq,
and the American Stock Exchange. These rules require listed acquirers to obtain
shareholder approval by a majority of the votes cast for any acquisition involving the
issuance of common stock, common stock options, and securities convertible into common
stock totaling 20% or more of the outstanding common stock before the issuance.3
3 See New York Stock Exchange Listing Rules, Section 312.03; National Association of Securities Dealers Manual, Section 4350; American Stock Exchange Manual, Section 712(b). In 1999, the American
http://law.bepress.com/alea/16th/art64
11
Acquirers can avoid the voting requirement under stock exchange listing rules by
issuing less than 20% of their outstanding common stock as consideration and paying the
balance in other securities or in cash.4 These substitutes are costly. First, using any of the
standard substitutes — cash, debt, or preferred stock — strains the borrowing power of the
acquirer. Second, using these substitutes in acquisitions initiated before July 2001
precludes the acquisition from a favorable accounting treatment as pooling of interests.5
Third, cash and debt are taxable and, when they exceed 40% of the consideration, cause the
remainder of the consideration also to be taxable. Fourth, preferred stock is subject to a
discount because, except in the rare case when it is publicly traded before the acquisition,
the price at which it will trade is uncertain.6
Bar Association added to its widely-followed Model Business Corporation Act a similar requirement in Section 6.21(f) for acquisitions involving the issuance of any voting stock carrying 20% of the voting power outstanding before the issuance.
4 Acquirers sometime choose a middle way by issuing preferred stock that automatically converts, or becomes convertible into, common stock upon acquirer shareholder approval after the deal closes. Examples in my sample include Jersey Integrated HealthPractice’s attempted acquisition of privately-owned Healthcare Integrated Services in 2000 (which was terminated by the target), Claimsnet.com’s acquisition of VHx’s assets in 2000 (which was completed 29 days after signing), and Healthwatch’s acquisition of Paul Harrison Enterprises in 1998 (which was completed a day after signing). See Healthcare Integrated Systems, Inc. Annual Report on Form 10-K for the fiscal year ending December 31, 2000; Claimsnet.com Inc. Current Report on Form 8-K dated March 23, 2000; Healthwatch, Inc. Current Report on Form 8-K dated October 1, 1998. None of the acquisitions involved the filing of a registration statement. Postponing shareholder approval until after deal completion could have both expedited the acquisition and sidestepped opposition.
5 In the past, acquirers could account for most all-stock acquisitions as pooling of interests, and thereby report higher earnings after the acquisition. In June 2001, however, the Financial Accounting Standards Board abolished pooling-of-interest accounting for acquisitions initiated after June 30, 2001. See Statement of Financial Accounting No. 141: Business Combinations, June 2001.
6 The registration statement filed by El Paso in connection with the Sonat acquisition explains: “In the event that the merger agreement is not approved by El Paso stockholders and the merger is completed under the alternative merger structure, we will attempt to list the depositary shares [of exchangeable voting preferred stock] on the New York Stock Exchange. The New York Stock Exchange does not currently have a market for El Paso depositary shares and there can be no assurance that such a trading market will develop … Until the depositary shares are fully distributed and an orderly market develops, the prices at which trading in any such shares occurs may fluctuate significantly.” See El Paso Energy Corporation Registration Statement on Form S-4 dated April 7, 1999.
Hosted by The Berkeley Electronic Press
12
Dealmakers might choose to bear these costs for three reasons. First, there may be
business reasons unrelated to voting for limiting the stock component of the consideration
and incidentally bringing it below the threshold that would trigger the voting requirement.
Second, dealmakers may wish to avoid voting for fear that shareholders would block the
acquisition. Third, dealmakers may wish to avoid voting because it is time consuming. It
requires calling a special shareholder meeting, filing a proxy statement and amending it as
needed, waiting until the proxy statement is approved, distributing the proxy statement to
shareholders, and soliciting proxies to ensure that half of the outstanding votes are cast to
meet quorum requirements.7 All of these actions delay the acquisition and expose it to
competing bids, employee departure, uncertainty among customers and trade partners,
financing costs, and adverse changes in the market.
From a policy perspective, it is important to ascertain whether acquirers issue less
stock than they would otherwise to avoid shareholder approval and, if so, whether their
motivation is to save time or to sidestep opposition. Only a finding that the latter
motivation drives acquirers to avoid shareholder approval would be evidence that this
approval is meaningful. If, conversely, acquirers avoid shareholder approval only to
expedite the acquisition, the voting requirement will need to be reconsidered. Such a
finding would suggest that voting creates a cost without serving its purpose.
7 See, for example, Delaware General Corporation Law, Section 216. Delaware law allows firms to lower the quorum to a third of the shares entitled to vote in their certificate of incorporation or bylaws. No firm in my sample has done so.
http://law.bepress.com/alea/16th/art64
13
IV. Data
I collect from Thomson Financial’s Securities Data Company Platinum database
(SDC) information on successful and unsuccessful non-hostile, uncontested acquisitions of
U.S. public and private firms by U.S. public firms containing common stock in the
consideration and announced between January 1, 1995 and December 31, 2003.
Acquisitions are included in my sample when the acquirer seeks to increase its stake in the
target from 20% or less to 50% or more because these acquisitions transfer control in the
target to the acquirer and therefore involve payment of a control premium. I exclude
hostile acquisitions and acquisitions challenged by competing bids for the same target
because they likely differ in structure, premiums, returns, and completion rates. I exclude
acquisitions without common stock consideration, which are typically cash acquisitions,
for similar reasons and because they rarely require shareholder approval. The sample
period begins in 1995 because many corporate filings are missing from EDGAR before
that year.
For each transaction, I obtain from SDC the transaction value, the method of
payment, the number of shares of common stock issued, the percentage of common stock
in the consideration, whether the transaction included a tender offer, the transaction status,
the announcement date and, when available, the completion date or the withdrawal date. I
also obtain from SDC the primary Standard Industry Classification (SIC) 4-digit code for
each acquirer and each target, and the Committee on Uniform Securities Identification
Procedures (CUSIP) 6-character code for each acquirer and each public target.
I add acquirer stock data from University of Chicago’s Center for Research in
Securities Prices (CRSP), accounting data from Standard & Poor’s Compustat database,
Hosted by The Berkeley Electronic Press
14
institutional investor ownership from Thomson’s CDA/Spectrum database, and executive
compensation data from Standard & Poor’s ExecuComp database. To be included in the
sample, the transaction value must be between 15% and 300% of the acquirer’s market
capitalization one day before the announcement to focus on acquisitions large enough to
require shareholder approval, but not so large as to effectively constitute a sale of the
acquirer to the target.8
For each acquisition, I obtain from acquirer registration statements, proxy
statements, current reports, and periodic reports information on whether the acquisition
needed acquirer shareholder approval, whether the acquisition was tax free, whether the
acquirer registered the stock to be issued, and whether and when acquisitions classified by
SDC as pending were ultimately completed or withdrawn. For acquisitions that need
acquirer shareholder approval, I also obtain the required majority for approval and, if more
than a majority of the votes cast in the shareholder meeting is required, the reason for the
higher requirement.
Finally, I search newspapers for mentions of actual or potential takeover bids for
each acquirer within two years before the announcement to study the effect of acquirer
shareholder voting on using acquisitions to stave off takeover bids. I record the number of
mentions, the date of the most recent mention, and whether any mention refers to an actual
bid, rather than a potential one.9
8 The results of my analysis remain qualitatively the same, though some coefficients lose significance, when I exclude transactions exceeding in value the acquirer’s market capitalization.
9 Specifically, I search the newspapers Wall Street Journal, Washington Post, Los Angeles Times, Boston Globe, Chicago Tribune, Houston Chronicle, San Francisco Chronicle, USA Today, Financial Times in the “Major Newspapers” file in Thomson’s Westlaw database for “[Acquirer name] /P acqui! merg! buy!
http://law.bepress.com/alea/16th/art64
15
Table 1 presents the distribution of acquisitions by industry. The industries most
represented in the sample are services, manufacturing, banking, and trade. Public target
acquisitions tend to require shareholder approval in every industry. In total, 666 public
target acquisitions required approval, and 217 did not. By contrast, private target
acquisitions tend not to require approval in most industries. Other than in banking and oil
and gas exploration, 648 private target acquisitions did not require approval, and 328 did.
Of private bank acquisitions, 45 acquisitions did not require approval, and 61 did. All 4
private target acquisitions in the oil and gas exploration industry required approval.
Table 2 provides summary statistics of deal characteristics. As can be expected,
deal value and the ratio of deal value to acquirer capitalization are on average higher in
acquisitions requiring shareholder approval. In addition, these acquisitions more often
involve only stock consideration, and otherwise involve on average a higher percent of
stock in the consideration. Consistently, part-stock acquisitions requiring shareholder
approval are more likely to be tax free. Acquirer capitalization is on average higher in
acquisitions requiring shareholder approval, suggesting that large acquirers make larger
acquisitions relative to their size and use stock as consideration more than small acquirers
despite the shareholder approval requirement this entails. Acquisitions requiring
shareholder approval are more likely to involve public targets, more likely to involve
payment in registered stock, less likely to be tender offers, and on average take longer to
complete, and are associated with lower announcement returns. All of these differences
bid! rumor! target play candidate takeover! & DA (AFT [Announcement date – 2 years] & BEF [Announcement date]).”
Hosted by The Berkeley Electronic Press
16
are significant at the 1% level in a Wilcoxon test (for the interval variables) and a Chi-
square test (for the indicator variables). Finally, acquisitions requiring shareholder
approval are associated with a lower average premium, though this difference is significant
only at the 10% level.
Of 1,969 transactions in the sample, 485 required shareholder approval by a
majority or a supermajority of the votes cast (“simple majority”), 439 required approval by
a majority of the votes outstanding, and 76 required approval by a supermajority of the
votes outstanding. The reasons for requiring a majority or a supermajority of the votes
outstanding (“absolute majority”) in 515 acquisitions are as follows (some acquisitions
required absolute majority approval for more than one reason).10
201 of the acquisitions that required absolute majority approval were conditional on
a charter amendment to increase the authorized stock. The need to authorize additional
stock stemmed either from the lack of sufficient stock to complete the acquisition, or from
the need to replenish stock reserves for future acquisitions, stock splits, public offerings,
private placements, or option grants. I classify an acquisition as conditional on
authorization of additional stock if the proxy statement filed by the acquirer so provides,
10 The proxy statements filed in connection with many acquisitions in the sample also solicited shareholder approval of option plans or stock plans. In most cases, the necessary approval was a simple a majority and was not a condition to the acquisition. There are four reasons to obtain shareholder approval of these plans. Section 162(m) of the Internal Revenue Code requires shareholder approval of executive incentive compensation packages in excess of $1 million as a condition for deducting the compensation from the firm’s taxable income. Section 422 of the Internal Revenue Code defers the tax liability on employee stock options to the time of exercising the options as long as the options meet certain criteria, including approval by shareholders. The listing rules of the New York Stock Exchange, Nasdaq, and American Stock Exchange require shareholder approval for the issuance of stock, options, or convertibles to executives or officers. In 2004, this requirement was extended to all employee stock or option plans. In addition, some states, notably New York (see New York Business Corporation Law, Section 505(d)), require shareholder approval of option grants to directors or officers. Until 1998, the required approval was by absolute majority. In 1998, New York law was changed to require approval by a simple majority.
http://law.bepress.com/alea/16th/art64
17
regardless of whether the stock is needed for the pending acquisition or for future use.
Most states require charter amendments to be approved by a majority of the votes
outstanding.11
37 of the acquisitions that required absolute majority approval were by acquirers
incorporated in California. 21 other acquisitions subject to this requirement were by
acquirers incorporated in Ohio. California law requires approval by a majority of the
acquirer votes outstanding of acquisitions involving the issuance of acquirer stock
representing 20% or more of the outstanding voting power before the issuance. Ohio law
requires approval of such acquisitions by two-thirds of the acquirer votes outstanding or a
smaller majority of the acquirer votes outstanding set in the charter. Unlike the stock
exchange listing rules, California law and Ohio law do not count stock options or
convertible securities, but do count voting stock other than common stock, towards the
20% threshold.12
193 of the acquisitions that required absolute majority approval were statutory
mergers of the target into the acquirer. Some states require approval by a majority or a
supermajority of the outstanding acquirer votes of statutory mergers in which 20% or a
similar portion of the acquirer common stock is issued.13 Since this threshold is similar to
11 See, for example, Delaware General Corporation Law, Section 242. Louisiana law requires approval by two-thirds of the votes cast in the meeting. See Louisiana Business Corporation Law, Section 312.
12 See California Corporations Code, Section 1201; Ohio General Corporation law, Section 1701.83. 13 See, e.g., Delaware General Corporation Law, Section 251(f). Some states adopted variations of
this rule. Louisiana, for example, requires that acquisitions in which 15% of the common stock is issued be approved by two-thirds of the acquirer votes cast. See Louisiana Business Corporation Law, Section 112. In North Carolina, any acquirer voting stock is counted. See North Carolina Business Corporation Act, Section 55-11-03. New Jersey requires approval by a majority of the votes cast or, in the case of a firm organized
Hosted by The Berkeley Electronic Press
18
the one triggering the voting requirement under the stock exchange listing rules, most
statutory mergers that need acquirer shareholder approval under state law would need
approval under the listing rules anyway. While structuring these acquisitions as statutory
mergers does not subject them to an otherwise avoidable approval requirement, it increases
the required majority from a majority of the votes cast to a majority of the votes
outstanding.14
123 of these statutory mergers were among banks, defined as firms whose primary
2-digit SIC code is 60 (depository institutions) or whose primary 4-digit SIC code is 6712
(offices of bank holding companies). Statutory mergers constitute 52% of bank
acquisitions in the sample that needed shareholder approval, compared to 10% of non-bank
acquisitions that needed shareholder approval. The ubiquity of statutory mergers in the
banking sector is noteworthy because neither banking regulation nor tax law discourages
the structuring of bank acquisitions as triangular mergers and many bank acquisitions are
in fact structured this way. The only historical disincentive to structuring bank
acquisitions as triangular mergers — an exemption of one-bank holding companies from
federal regulation, which one-bank holding companies forfeit once they acquire another
bank in a triangular merger — disappeared when this exemption was repealed in the Bank
prior to 1969, two-thirds of the votes cast. See New Jersey Business Corporation Act, Section 14A:10-12. New York requires approval by two-thirds of the acquirer votes outstanding of any statutory merger unless the acquirer owns at least 90% of the target stock before the merger. See New York Business Corporation Law, Section 903.
14 There is no evidence that acquisitions are structured as statutory mergers when absolute majority approval is required for other reasons. 67 of the 206 (33%) statutory mergers that require approval require absolute majority approval for other reasons, the most common of which is the need to increase the authorized stock in the charter. Similarly, 280 of the 784 (36%) non-statutory mergers that require approval require absolute majority approval for similar reasons.
http://law.bepress.com/alea/16th/art64
19
Holding Company Act Amendments of 1970. Many bank acquisitions are still structured
as statutory mergers probably due to inertia, reinforced by the fact that, in the banking
industry, a triangular merger does not shelter the assets of the acquirer from creditors of the
target.15 That dealmakers routinely structure bank acquisitions as statutory mergers
suggests that the higher majority needed for approving statutory mergers is not a deterrent,
and that the motivation for structuring acquisitions in other industries as triangular mergers
is the desire to shelter the assets of the acquirer from the liabilities of the target and avoid a
vote altogether.
238 acquisitions required absolute majority approval for other reasons. Some
acquisitions, for example, involved a merger of the acquirer and the target with
subsidiaries of a new holding company. This so-called double-dummy structure is
typically used to qualify the acquisition as a tax-free reorganization when a triangular
merger structure would not achieve this goal.16 State laws require double-dummy
acquisitions to be approved by a majority of the acquirer votes outstanding because they
involve a merger of the acquirer with another firm. [Describe other typical reasons]
15 This deviation from the doctrine of limited liability in corporate law emanates from the “source of strength” policy adopted by the Federal Reserve Bank in 1987 and incorporated into its regulations in 1997. See Policy Statement on the Responsibility of Bank Holding Companies to Act as a Source of Strength to Their Subsidiary Banks, 52 Fed. Reg. 15,707 (Apr. 30, 1987); Bank Holding Companies and Change in Bank Control (Regulation Y), 12 CFR, Section 225.4(a). Another reason to structure an acquisition as a statutory merger could be the reluctance of the merging banks to appear as being acquired. I find no evidence, however, that either the portion of the acquirer stock issued in the acquisition or the ratio of the deal value to the market capitalization of the acquirer one trading day before the announcement explains the prevalence of statutory mergers in bank acquisitions.
16 Triangular mergers qualify as tax-free reorganizations if they satisfy Section 368 of the Internal Revenue Code, which practitioners have traditionally construed as requiring that at least 40% of the consideration be any class of stock. Double-dummy mergers qualify as tax-free reorganizations if they satisfy Section 351 of the Code, which does not contain a similar requirement.
Hosted by The Berkeley Electronic Press
20
V. Analysis
I use four methods to examine whether shareholder voting on acquisitions
disciplines management and whether it delays acquisitions. First, I test whether and when
acquirers avoid shareholder approval, and whether doing so affects deal duration. Second,
I test whether acquisitions requiring shareholder approval are viewed more favorably by
shareholders. Third, I test whether these acquisitions involve lower premiums. Finally, I
test whether these acquisitions are less likely to be completed. The analysis below
explains each method and presents my findings.
1. Do Acquirers Avoid Shareholder Approval?
The intent to sidestep opposition by avoiding shareholder approval is rarely
evident. Obvious cases include acquisitions that are publicly criticized for avoiding
shareholder approval,17 and acquisitions that are restructured to obviate the need for
shareholder approval after having being voted down.18 More ambiguous cases include
contested acquisitions that are restructured to avoid shareholder approval before having
17 One example is the 2001 failed attempt by M & F Worldwide to buy a controlling block in Panavision from a shareholder who controlled both firms. Another example, not in my sample, is Sovereign Bancorp’s pending acquisition of Independence Community Bank using the proceeds from issuing 19.8% of Sovereign’s common stock to Banco Santander. See Aaron Lucchetti, Carrick Mollenkamp, and Jesse Eisinger, NYSE won’t require vote for Sovereign deal, Wall Street Journal, November 23, 2005, C1; Sovereign Bancorp , Inc. Current Report on Form 8-K dated November 22, 2005.
18 Examples of this scenario in my sample (all involving acquirers represented by the same law firm) are Rite Aid’s 1995 failed acquisition of Revco, El Paso’s 1996 acquisition of Tenneco, and El Paso’s 1999 acquisition of Sonat. Each of these transactions contemplated seeking acquirer shareholder approval first and, if this fails, reducing the common stock component of the consideration below the 20% threshold and completing the acquisition without approval. See Rite Aid Corporation Current Report on Form 8-K dated November 29, 1995; El Paso Natural Gas Company Registration Statement on Form S-4 dated August 27, 1996; El Paso Energy Corporation Registration Statement on Form S-4 dated April 7, 1999. In each of these cases, shareholders overwhelmingly approved the acquisition.
http://law.bepress.com/alea/16th/art64
21
been voted down. Although the fear of opposition is part of the motivation for these
restructured deals, the desire to complete the acquisition quickly is a factor as well.19
In the more common case, where the original deal does not require shareholder
approval and is not criticized for this, it is unclear whether the absence of a vote is
deliberate and, if so, whether the motivation is to sidestep opposition or to save time.
Answering these questions requires an understanding of the circumstances in which each
of the relevant drivers of deal structure — time saving, fear of opposition, or reluctance to
issue common stock — comes to bear. I examine these circumstances next.
A. Saving Time
Avoiding shareholder approval should expedite deal completion when there are no
other delaying factors. Some of these factors are exogenous. For example, obtaining
target shareholder approval delays acquisitions of public targets. Obtaining regulatory
approval delays acquisitions in regulated industries (banking, insurance, transportation,
communications, and utilities) even longer. Accordingly, in these acquisitions, acquirer
shareholder approval should be more common, and the delay associated with it should be
shorter.
19 Examples include Time’s 1989 acquisition of Warner, which is not in my sample, and PeopleSoft’s 2003 acquisition of J.D. Edwards, which is. Both deals were modified from an all-stock triangular merger (which needed acquirer shareholder approval) to a cash-and-stock tender offer (which did not) in response to an unsolicited bid for the acquirer. See Robin Sidel, PeopleSoft’s J.D. Edwards offer adds $850 million cash element, Wall Street Journal, June 17, 2003, A3; Paramount Communications, Inc. v. Time Incorporated, 571 A.2d 1140 (Del.1989). In both, the fear of opposition did not fully explain the restructuring. Had the parties only intended to avoid a negative vote, they could have reduced the stock component of the consideration while retaining the merger structure. By also switching to a tender offer, the parties obviated the need for a target shareholder meeting as well, thereby expediting the acquisition.
Hosted by The Berkeley Electronic Press
22
Other delaying factors are endogenous. In private target acquisitions, paying in
registered stock (rather than unregistered stock) will significantly delay the deal. In public
target acquisitions, a one-step merger (rather than a tender offer followed by a merger) will
result in a similar delay. These structures should be more common in acquisitions of
regulated targets, shareholder approval should be more common when they are employed,
and the delay associated with it should be shorter.
I shall now elaborate on the above predictions. Private target acquisitions that are
not subject to acquirer shareholder approval can be completed immediately and without
any public filing. A proxy statement soliciting the target shareholders’ votes is not needed
because the target is closely held, and a registration statement for the stock issued in the
acquisition is not needed because the issuance is not a public offering. A registration
statement is necessary only if target shareholders wish to dispose of the stock they receive
as soon as the deal closes. In other cases, the stock will be registered later.
Public target acquisitions can never be completed immediately because they always
require public filings. But they can be completed faster if structured as tender offers.
Mergers require target shareholder approval. When the target is public, obtaining this
approval requires the target to call a special shareholder meeting, file a proxy statement,
amend it as needed, wait until it is approved, distribute it to shareholders, solicit proxies,
and hold the meeting. By contrast, tender offers do not require target shareholder approval
and can close in 20 business days.20 The acquirer must still file a disclosure schedule but,
20 The only target shareholder vote needed is for approving a backend merger after the tender offer closes and the acquirer controls the target, and even this approval is not needed when the tender offer leaves
http://law.bepress.com/alea/16th/art64
23
unlike a proxy statement, this filing needs no regulatory approval and imposes no delay. In
addition, the acquirer must file a registration statement when issuing stock. This
requirement can nevertheless be avoided by making a cash tender offer and issuing stock
only in a backend merger.21 Moreover, since the year 2000, tender offers can close quickly
even when they include stock under a rule allowing them to commence before the
registration statement is approved.22
B. Sidestepping Opposition
While the benefits of avoiding the delay associated with a shareholders meeting can
suffice to motivate acquirers to avoid shareholder approval, the motivation should be
stronger if shareholders are expected to vote the acquisition down. A number of factors
can make this outcome more likely. The first factor is the majority needed for approval. It
is easier to win the support of a majority of the votes cast in the meeting than to win the
support of a majority of the votes outstanding. The second factor is the composition of
acquirer shareholders. It is easier to win shareholder approval when a larger portion of the
stock is held by insiders, and it can be either easier or harder to win approval when a larger
portion of the stock is held by blockholders or institutional investors. The third factor is
past performance of the acquirer generally, and the success of past acquisitions
the acquirer with 90% of the target stock. 21 The 1989 tender offer by Time for the stock of Warner was structured this way and closed in 38
days. The registration statement filed in connection with the backend merger was approved 135 days later. See Time Warner Incorporated Letter to the Securities and Exchange Commission dated December 21, 1989, available at Wastlaw’s SEC No-Action Letter library.
22 See Securities Act Release No. 33-7760, Exchange Act Release No. 34-42,055, Investment Company Act Release No. 24,107, Final Rule: Regulation of Takeovers and Security Holder Communications, 64 Fed. Reg. 61,408 (November 10, 1999) (effective January 24, 2000). The 2003 cash-and-stock tender offer by PeopleSoft for the stock of J.D. Edwards, for example, closed in 29 days. See J.D. Edwards & Company Amendment No. 3 of Schedule 13D dated August 29, 2003.
Hosted by The Berkeley Electronic Press
24
particularly. Shareholders can be expected to be more critical of acquisitions announced
by poor performers. The fourth factor is the availability of an opportunity to sell the
acquirer rather than buy the target. Shareholders should be less supportive of an
acquisition when the acquirer has been targeted by other bidders or when its industry is
consolidating.
C. Limiting Stock Issuance without Relation to Voting
Acquirers can limit the stock component of the consideration, incidentally avoiding
the need for shareholder approval, for reasons unrelated to this approval. Martin (1996)
and Faccio and Masulis (2005) find that acquirers are more likely to pay in cash when they
have ample retained earnings or borrowing power. Both of these studies, as well as
Rhodes–Kropf, Robinson, Viswanathan (2005), Ang and Cheng (2006), and Dong,
Hirshleifer, Richardson, and Teoh (2006), find that lower valued acquirers tend to pay in
cash. Acquirers are more likely to pay in cash for private targets because issuing stock
would create a significant blockholder and because private target shareholders prefer cash
(Faccio and Masulis (2005)). Last, in public target acquisitions, acquirers are more likely
pay in cash when institutional investors hold a large portion of the target stock because
they are less sensitive to tax than individuals.
D. Testing the Hypotheses
I begin the analysis with the time-saving hypothesis, which predicts that
acquisitions of public targets and regulated targets will take longer to complete, will be
more likely to require shareholder approval and employ time-consuming acquisition
http://law.bepress.com/alea/16th/art64
25
techniques (acquisition for registered stock in the case of private targets, and merger in the
case of public targets), and suffer a shorter delay when doing so.
To test whether regulated targets take longer to acquire regardless of shareholder
approval, I regress the duration of completed acquisitions from announcement to closing
on regulated industry indicators, an indicator for acquisitions that need shareholder
approval (VOTE), and interactions between the regulated industry indicators and the vote
indicator.23 Initially, I combine all regulated industries (ALL_REGULATED). Next, I
separate regulated industries into banks (BANK), communications
(COMMUNICATIONS), and other regulated industries (OTHER_REGULATED). The
latter category pools transportation, utilities, and insurance because there are not enough
observations to separate them in the regressions.
Table 3 presents the results. The first two columns report results for private target
acquisitions. As predicted, the presence of shareholder approval is associated with a large
increase (92 days) in deal duration beyond the baseline duration (52 days). Regulated
target acquisitions taken as a whole are somewhat less sensitive to shareholder approval.
While these acquisitions take 56 more days to complete than other acquisitions,
shareholder approval prolongs them only by 68 days (VOTE + ALL_REGULATED×
VOTE). When regulated industries are separated into different categories, it becomes clear
that banks are driving the effect. Bank acquisitions take 110 more days to complete than
other acquisitions regardless of whether shareholder approval is needed: VOTE +
23 The shortest deal duration is zero (172 observations, all involving private targets), corresponding to acquisitions announced at deal closing. In unreported regressions, I obtain similar results when using the log of deal duration as the dependent variable.
Hosted by The Berkeley Electronic Press
26
BANK×VOTE is not significantly different from zero. The remaining regulated industries
are not significantly different from unregulated industries.
The second two columns report results for public target acquisitions. As in private
target acquisitions, regulated target acquisitions and bank acquisitions take significantly
longer to complete (60 days and 56 days, respectively). Acquisitions of non-bank
regulated targets take even longer (119 days). As predicted, in difference from private
target acquisitions, the presence of shareholder approval is not associated with an increase
in the duration even when the target is unregulated. Shareholder approval is associated
with a longer duration only in acquisitions of communications firms. Given the finding for
unregulated target acquisitions, this probably does not reflect a delay caused by voting.
To further test whether bank acquisitions take longer to complete independent of
shareholder approval, I collect voting dates from 450 acquirer proxy statements and
calculate the lag between this date and deal completion. The mean lag is 28 days in bank
acquisitions and 15 days in other acquisitions, and the difference between the two is
significant at the 1% level in a Wilcoxon test. These findings are noteworthy given that
dealmakers seem to anticipate that bank acquisitions will take longer to complete and set a
late voting date accordingly: The mean lag between deal announcement and voting is 143
days in bank acquisitions and 130 days in other acquisitions, and the difference between
the two is significant at the 1% level in a Wilcoxon test.
That bank acquisitions cannot be expedited is also suggested by the fact that private
banks are less likely to be acquired for unregistered stock than other private targets. Only
21% of banks in the sample are acquired for unregistered stock, compared to 78% of
private communications firms, 80% of other regulated private firms, and 83% of
http://law.bepress.com/alea/16th/art64
27
unregulated private firms. A likely interpretation of this difference is that issuing
unregistered stock is less beneficial in bank acquisitions because their duration is fixed.
Similarly, none of the tender offers in the sample is a bank acquisition (although none is an
acquisition of any other regulated target either).
Next, I examine the time-saving hypothesis together with the other two hypotheses.
To test whether acquisitions are structured to avoid acquirer shareholder voting and, if they
are, whether the motivation lies in a desire to save time or in a concern that shareholders
will vote against the acquisition, I estimate a probit model in which the dependent variable
is VOTE, and the independent variables are factors that can motivate dealmakers to avoid
shareholder voting or limit for other reasons the amount of stock that will be issued:
private target (PRIVATE), issuance of unregistered stock (PRIVATE×UNREGISTERED),
In a modified specification, I add the portion of acquirer stock held by institutions
(INSTITUTIONS), and an indicator for acquirers that had been actual or rumored targets
within two years before the acquisition announcement (DEFENSIVE).24
The time-saving hypothesis predicts that BANK will be positive or insignificantly
different from zero (because public bank acquisitions cannot be expedited by avoiding
shareholder approval, but so cannot most other public target acquisitions),
PRIVATE×UNREGISTERED will be negative (because private target acquisitions for
unregistered stock can be expedited), PRIVATE×UNREGISTERED×BANK will be
24 The results do not materially change when, following and Qiu (2004) and Chen, Harford, and Li (2005), I define INSTITUTIONS as the portion of the acquirers outstanding common stock held by the largest institutional investor, all public pension funds, the largest five public pension funds, or the largest public pension fund.
Hosted by The Berkeley Electronic Press
28
positive (because bank acquisitions cannot be expedited even when the target is private and
is acquired for unregistered stock), TENDER will be negative (because tender offers can
be expedited), and PRIVATE will be negative or insignificantly different from zero
(because private target acquisitions for registered stock cannot be much expedited).
The opposition avoidance hypothesis predicts that the INSTITUTIONS will be
negative (because institutional investors are more likely than retail investors to oppose
management), and DEFENSIVE will be negative (because shareholders favor a sale of the
acquirer over the proposed acquisition).
The hypothesis that acquirers limit the issuance of stock for reasons unrelated to the
approval requirement predicts that LEVERAGE will be positive (because paying in stock
is cheaper for leveraged acquirers), PRIVATE will be negative (because shareholders of
private target shareholders prefer cash, and acquirers of private targets prefer to avoid the
creation of a new blockholder), and SIZE will be positive (because it is difficult to finance
large acquisitions).
Table 4 presents the marginal effects of each independent variable on the
probability that shareholder approval will be required. The results are consistent with the
time-saving hypothesis. Shareholder approval is less likely to be required in tender offers
and in acquisitions of private targets for unregistered stock, and more likely to be required
in bank acquisitions. As expected, the probability that shareholder approval is required is
positively related to the relative deal size and to acquirer leverage, which make payment in
cash costly. Acquisitions of private targets for registered stock are not significantly
different from acquisitions of public targets, suggesting that when time is of the essence,
acquirers of private targets avoid both shareholder approval and stock registration. The
http://law.bepress.com/alea/16th/art64
29
probability that shareholder approval is required is not significantly related to the
acquirer’s being an actual or rumored takeover target, which can expose the acquisition to
stricter scrutiny, or to institutional ownership, which can both subject the acquisition to
stricter scrutiny and expedite the approval process.
To further explore the possibility that acquirers avoid shareholder approval to save
time, I regress the duration of completed acquisitions from announcement to closing
(DURATION) on VOTE, PRIVATE, PRIVATE×UNREGISTERED, stock, interactions
between these indicators and the vote indicator, and TENDER. By comparing the effect of
voting on acquisitions involving the issuance of registered stock and acquisitions involving
the issuance of unregistered stock, this specification should yield unbiased estimates even
if the sample contains deals in which a registration statement disclosing the approval
requirement was filed but is missing from EDGAR. I do not include in the regression an
interaction between the vote indicator and the tender offer indicator because, consistent
with the time saving hypothesis, only 5 of 20 tender offers in the sample required a vote.
Table 5 presents the results. The first column presents the results for the full
sample. As the time saving hypothesis predicts, both tender offers and private target
acquisitions for unregistered stock without shareholder approval are completed
significantly faster than other acquisitions. The delay associated with shareholder approval
is insignificant in mergers with public targets (VOTE) and acquisitions of private targets
for registered stock (VOTE + PRIVATE×VOTE), but is large and highly significant in
private target acquisitions for unregistered stock (VOTE + PRIVATE×VOTE +
PRIVATE×UNREGISTERED×VOTE). The latter acquisitions are completed 103 days
before other acquisitions if shareholder approval is not required, but no faster than other
Hosted by The Berkeley Electronic Press
30
acquisitions if approval is required. Tender offers are completed 61 days before other
acquisitions, and the difference is highly significant.
The second column presents results for a sample of non-bank acquisitions to ensure
that the results are not driven by the fact that, as noted earlier, bank acquisitions are more
likely than other acquisitions to require shareholder approval, involve payment in
registered stock, and be structured as mergers rather than tender offers. The results are
similar to those for the full sample. The third column reports results for a sample of bank
acquisitions. Here the delay associated with shareholder approval in private target
acquisitions for unregistered stock (VOTE + PRIVATE×VOTE +
PRIVATE×UNREGISTERED×VOTE) is only 46 days and is barely significant at the 10%
in a Wald test. R-squared is 0.03, compared to 0.25 for non-bank acquisitions, and 0.28 for
the full sample. Given that there are 370 bank acquisitions and only 6 independent
variables in the regression, the weak results for bank acquisitions are unlikely to be driven
by sample size.
Taken as a whole, the results in Table 5 suggest that voting significantly delays
tender offers and private target acquisitions for unregistered stock when the target is not a
bank. The effect of voting on the duration of bank acquisitions is minimal. In unreported
regressions using log of DURATION as the dependent variable, the difference between
bank acquisitions and non-bank acquisitions becomes more pronounced. While the
coefficients retain their signs and significance in non-bank acquisitions, the effect of voting
on the duration of private target acquisitions for unregistered stock disappears. The
difference in explanatory power increases as well: R-squared increases to 0.34 in non-bank
acquisitions, compared to 0.03 in bank acquisitions.
http://law.bepress.com/alea/16th/art64
31
2. The Effect of Voting on Announcement Returns
Another measure of the effectiveness of acquirer shareholder voting as a constraint
on acquirer management, in addition to evidence that acquirers avoid it when opposition is
likely, is the effect of voting on announcement returns.25 Acquisition announcements
almost always list the main conditions to consummation of the acquisition, including
acquirer shareholder approval, when this approval is needed.26 However, if managers
expect shareholders to vote down poorly received acquisitions, announcement returns
should be higher when shareholder approval is required even if shareholders do not know
at the time of announcement whether their approval is required because the approval
requirement will affect the terms of the transaction, which are at the center of acquisition
announcements.
To test whether acquirer announcement returns are affected by shareholder voting, I
regress the acquirer’s cumulative abnormal returns in the 5 trading days around the
announcement on an indicator for whether shareholder approval was required and a
number of controls: the portion of acquirer common stock issued in the deal (PORTION), a
private target, an interaction between these two variables, a large acquirer (LARGE), a
recently targeted acquirer, a diversifying acquisition, and year fixed effects. In an
25 Announcement returns measure the effectiveness of shareholder voting better than long-term returns because long-term returns reflect information that becomes public only after shareholders have voted. Qiu (2004) and Chen, Harford, and Li (2005) find that concentrated ownership by independent, long-term institutional investors is unrelated to announcement acquirer returns but is positively related to post-acquisition returns and performance. While this finding supports the hypothesis that these institutional investors monitor management after the acquisition, it does not support the hypothesis that they monitor acquisition decisions.
26 For example, the press release announcing the acquisition of OfficeMax by Boise Cascade in 2003 states: “This transaction has been approved by the boards of directors of both companies and is subject to approval by regulatory authorities and shareholders of both companies.” See Boise Cascade Corporation Current Report on Form 8-K dated July 14, 2003.
Hosted by The Berkeley Electronic Press
32
alternative specification, I replace the portion of acquirer common stock issued in the deal
(and its interaction with the private target indicator) with the portion of consideration paid
in stock (PCT_STOCK). I predict that acquirer returns will be higher in private target
acquisitions, positively related to the portion of acquirer stock issued (or the stock portion
of the consideration) in private target acquisitions, negatively related to this portion in
public target acquisitions, and lower for large acquirers, diversifying acquisitions, and
recently targeted acquirers.
These predictions are motivated by prior literature. Chang (1998) finds higher
acquirer returns for private target acquisitions in a sample of acquisitions for stock. Fuller,
Netter, and Stegemoller (2002) and Officer (2006) find similar results in acquisitions for
cash and acquisitions for cash and stock. Asquith, Bruner, and Mullins (1987) and Fuller,
Netter, and Stegemoller (2002) find that acquirer returns decrease in the portion of acquirer
stock issued in public target acquisitions. By contrast, Chang (1998), Fuller, Netter, and
Stegemoller (2002) find that acquirer returns increase in the portion of acquirer stock
issued in private target acquisitions, and Slovin, Sushka, and Polonchek (2005) find similar
results in acquisitions of divisions.
The authors offer several explanations for the differences they find between public
targets and private targets.27 First, private targets are cheaper because their stock is
illiquid.28 Second, private target shareholders that receive acquirer stock become monitors
27 Another explanation for the higher returns to acquirers that use cash to pay for the acquisition is that doing so increases acquirer leverage and thereby forces management to perform better to keep the acquirer afloat. This explanation receives empirical support in Maloney, McCormick, and Mitchell (1993), and Yook (2003). It does not explain, however, the higher returns to acquirers that use stock to pay for private targets.
28 This interpretation is consistent with the finding by Koeplin, Sarin, and Shapiro (2000) and Officer
http://law.bepress.com/alea/16th/art64
33
of acquirer management after the acquisition. Third, private target shareholders are willing
to be paid less in stock to avoid taxation. Fourth, while paying in stock signals that the
acquirer is overvalued in public target acquisitions, it affirms the acquirer’s value in private
target acquisitions because private target stockholders would accept acquirer stock only
after verifying its value.29 A fifth explanation, not discussed by the authors, is that paying
in stock induces arbitrageurs to buy target stock while short-selling acquirer stock in public
target acquisitions. This strategy is not possible in private target acquisitions.30
Moeller, Schlingemann, and Stulz (2004) find that large acquirers experience lower
returns. They attribute this finding to the presence higher agency costs in large firms.
Morck, Shleifer, and Vishny (1990), Maquieira, Megginson, and Nail (1998), and DeLong
(2001) find that diversifying acquisitions are associated with lower acquirer returns. Louis
(2004) finds that bank acquirers experience lower returns if they were targets of actual or
rumored bids before the acquisition announcement, and concludes that these acquisitions
serve as takeover defense. Gorton, Kahl, and Rosen (2005) develop a theoretical model of
value-decreasing defensive acquisitions.
(2006) that private targets receive lower acquisition premiums based on accounting performance. 29 Reports in the financial press are consistent with this explanation. For example, Shabelman (2005)
quotes “a source close to [eBay’s 2005 acquisition of Rent.com] who asked not to be identified” confirming that the sharp decline in the eBay stock price following the announcement of the acquisition motivated the parties to eliminate the stock component of the consideration because “eBay very much wants to do the deal and to signal to the market that its stock is depressed.” Another possible source of higher acquirer returns in acquisitions of private targets regardless of the method of payment is that many private targets are subsidiaries divested by firms under pressure to divest non-core assets (Jain (1985), Lang, Poulsen, and Stulz (1995), Maksimovic and Phillips (2001)).
30 Mitchell, Pulvino, and Stafford (2004) find that arbitrage short sales account for nearly half of the negative acquirer returns in stock-financed acquisitions of public targets.
Hosted by The Berkeley Electronic Press
34
Following Schwert (1996), I calculate for each acquirer the market model
regression (1) for the 253 trading days (about one year) ending 127 trading days (about six
months) before the announcement of the acquisition:
Rit = αi + βiRmt + εit, t = –379, . . . , –127 (1)
where Rit is the continuously compounded return to the stock of acquirer i and Rmt is the
continuously compounded return to the CRSP value-weighted portfolio of stock listed on
NYSE, AMEX, and Nasdaq for day t. Acquirers are included if they have at least 100
daily returns available to estimate the parameters of (1). Estimates of (1) are used to
compute CAR.
After estimating an ordinary least squares model, I repeat the estimation using a
selection model to address potential endogeneity. The insignificant coefficient of VOTE
can be due to an omitted variable that is correlated with VOTE but affects CAR in the
opposite direction. It is not clear that such an omitted variable exists. Payment in stock,
which is correlated with voting and should be negatively related to returns, is included in
the regression as PORTION. Shareholder opposition, which can affect both the presence
of voting and returns, and is omitted if not captured by DEFENSIVE, affects voting and
returns in the same direction. Nevertheless, I use instrumental variables to ensure that the
insignificance of VOTE is not due to omitted variables. Specifically, I estimate a two
stage least squares model in which two variables that increase the likelihood of voting but
should not affect returns — LEVERAGE and SIZE — substitute for VOTE.
http://law.bepress.com/alea/16th/art64
35
Table 6 presents the results. Columns (1) and (2) present the results of estimating
an ordinary least squares model. Shareholder approval is not significantly related to
returns. Other variables, however, generally follow the predictions. Consistent with prior
studies, private target acquisitions are associated with higher returns, the portion of
acquirer stock issued (and the stock portion of the consideration) in private target
acquisitions is positively related to returns, and large acquirers experience lower returns.
The portion of acquirer stock issued in public acquisitions is not significantly related to
returns, and the stock portion of the consideration in public acquisitions is only slightly
negatively related to returns — perhaps because the sample does not contain all-cash
acquisitions, which prior studies include. Somewhat surprisingly, diversifying acquisitions
are associated with positive returns.
Columns (3) and (4) present the results of estimating a two stage least squares
model using LEVERAGE and SIZE as instruments for VOTE. These instruments are
strong predictors of voting both in Table 4 and in an unreported probit regression in which
they are the only independent variables. In the latter regression, LEVERAGE is significant
at the 10% level, SIZE is significant at the 1% level, and the pseudo R-square is 0.10.
Nevertheless, using them does not materially change the results, suggesting that the
ordinary least squares model does not suffer from endogeneity.
3. The Effect of Voting on Premiums
Another measure of the effectiveness of acquirer shareholder voting is the
magnitude of acquisition premiums. Acquirer shareholders can object to a proposed
acquisition for a variety of reasons, including overpayment. Accordingly, if acquirer
shareholder voting is effective, acquisitions that require shareholder approval should
Hosted by The Berkeley Electronic Press
36
involve lower premiums. Premiums thus have both an advantage and a disadvantage
compared to acquirer returns as a measure of the effectiveness of shareholder voting. On
the one hand, premiums can reflect price differences that are too small to affect acquirer
returns. On the other hand, premiums do not reflect the effectiveness of shareholder voting
in blocking acquisitions for reasons other than price.
To test whether acquisitions requiring shareholder approval are associated with
lower premiums, I estimate for public target acquisitions an ordinary least squares model
in which the dependent variable is the ratio of deal value to stock market capitalization of
the target firm 30 trading days before the announcement (PREMIUM) and the independent
variables are VOTE and several controls: the percentage of consideration paid in stock
(PCTSTOCK), SIZE, LARGE, DEFENSIVE, a proxy for acquirer agency costs calculated,
following Hartzell, Ofek, and Yermack (2004), as the residual from regressing the acquirer
chief executive officer’s salary and bonus on two-digit SIC industry indicators, log of
market capitalization, year indicators, and the prior year stock return as a proxy for agency
costs, 2-digit SIC industry fixed effects, and year fixed effects.
Table 7 reports the results. Columns 1 and 2 present regression results for all
public target acquisitions. PCTSTOCK is the only variable whose coefficient estimate is
significantly different from zero. One interpretation of its negative sign is that acquirers
pay less when a larger portion of the consideration is tax-free.31 This interpretation is
consistent with Ayers, Lefanowicz, and Robinson (2003). Columns 4 and 5 report similar
31 Non-stock consideration is taxable even in tax-free acquisitions. See Internal Revenue Code, Section 356.
http://law.bepress.com/alea/16th/art64
37
results for a sample of tax-free acquisitions, which is unsurprising, since most of the
transactions in the full sample are tax free. Columns 3 and 6 omit all of the variables other
than VOTE and SIZE. In both columns, VOTE is negative and significant, as in
Hamermesh (2004). That VOTE loses significance and PCTSTOCK is highly significant
when both are included suggests that it is the percentage of stock in the consideration,
possibly reflecting tax savings, that drive the results in Hamermesh (2004). When a proxy
tax savings — the portion of consideration paid in stock — is included, the overall fit of
the model increases and the coefficient estimate of VOTE becomes insignificant.
Consistently, in unreported regressions I find that VOTE is positively related to
PCTSTOCK and SIZE.
4. The Effect of Voting on Deal Completion
Acquirer shareholders almost never vote acquisitions down.32 In my sample, they
did so once, in Centura Software’s failed attempt to acquire InfoSpinner in 1997. The
circumstances in that case were unusual. First, because the acquirer was incorporated in
California, the necessary majority for approving the acquisition was a majority of the votes
outstanding, rather than a majority of the votes cast, as required by stock exchange rules.
Second, the acquisition was opposed by the firm’s founder and largest shareholder, who
32 A celebrated example of a poorly received acquisition that received acquirer shareholder approval is the $25 billion acquisition of Compaq by Hewlett Packard in 2002, which was completed even though it was associated with cumulative abnormal returns of –0.18 in the five trading days around its announcement and was opposed by members of the founding family owning 18% of the stock. A more recent example is the $13.5 billion acquisition of Veritas by Symantec in 2005. See David Marcus, Split decision, Corporate Control Alert, July 2005, 4.
Hosted by The Berkeley Electronic Press
38
then resigned from the board and cast 11.7% of the outstanding votes against the
acquisition.33
The rarity of instances in which acquirer shareholders vote acquisitions down can
have three different interpretations. The first interpretation is that voting does not
constrain acquirer management because other disciplining devices render voting redundant
or because collective action costs prevent shareholders from voting effectively.34 The
second interpretation is that voting does constrain management and therefore only
acquisitions acceptable to shareholders are announced. The third interpretation is that
managers abandon acquisitions that meet with unexpected opposition before they are voted
down.35
33 See Centura Software Corporation Registration Statement on Form S-4 Dated January 27, 1997. In a March 10 filing, Centura announced that the voting would take place on March 31 and that the record date for determining who can vote would be February 28. See Centura Software Corporation Amendment No. 1 to Registration Statement on Form S-4 Dated March 10, 1997. In a March 31 filing, Centura postponed the shareholder meeting to April 17 and the record date to March 31. See Centura Software Corporation Amendment No. 2 to Registration Statement on Form S-4 Dated March 31, 1997. See Centura Software Corporation Amendment No. 2 to Registration Statement on Form S-4 Dated March 31, 1997. In a May 15 filing, Centura reported that the acquisition was not approved by the required majority. See Centura Software Corporation Quarterly Report for the Quarterly Period Ended March 31, 1997. Ironically, the initial market reaction to the acquisition was enthusiastic: The five-trading-day cumulative abnormal returns to Centura’s shareholders at announcement were 0.278.
34 A newspaper article discussing the aborted acquisition of IMS Health by VNU notes: “It is extremely rare for a big company such as VNU to fall afoul of its shareholders in such a dramatic fashion. While shareholders often complain about mergers and acquisitions, those who disapprove usually sell their stock rather than fight.” See Jason Singer, VNU investors maintain pressure, Wall Street Journal, November 18, 2005, B2.
35 Acquirers can abandon acquisitions without disclosing why. Pharmacopeia’s termination of its Eos acquisition in 2002 is an example. See Pharmacopeia, Inc. Report on Form 8-K dated January 17, 2002. Nevertheless, the trail of filings that preceded that termination, including filings by a 10% shareholder that solicited proxies against the deal, made the reason for the termination abundantly clear. See Pharmacopeia, Inc. Notice of Exempt Solicitation by OrbiMed Advisors LLC dated January 8, 2002. A rare example of an acquirer terminating a definitive acquisition agreement while citing shareholder opposition is Landry’s Seafood Restaurants’ failed acquisition of Consolidated Restaurant Companies, which was called off five days after its announcement amid strong shareholder opposition and five-trading-day cumulative abnormal returns of –0.29. It is unclear from the filing whether the acquisition depended on acquirer shareholder approval. See Landry’s Seafood Restaurants, Inc. Current Report on Form 8-K dated March 2, 1999.
http://law.bepress.com/alea/16th/art64
39
In this section, I test the third interpretation. Specifically, I test whether poorly
received acquisitions are more likely to be abandoned when they require shareholder
approval. Acquisitions can be withdrawn for a variety of reasons, including the sale of the
target to another bidder, failure to obtain regulatory approval, a material adverse change in
the business of the acquirer or the target, a decline in the value of the stock portion of the
consideration, or a rejection of the deal by target shareholders. Acquirer shareholder
opposition can be another reason for abandoning an acquisition.
To be sure, acquirers can abandon poorly received acquisitions to avoid market
discipline even if shareholder approval is not required. Mitchell and Lehn (1990) find that
acquirers whose acquisitions are associated with lower announcement returns are likely to
become takeover targets. Lehn and Zhao (2006) find that managers of acquirers whose
acquisitions are associated with lower announcement returns are likely to be replaced,
especially if they complete the acquisition. Investigating whether this discipline affects
acquirer behavior, Jennings and Mazzeo (1991) find no relation between acquirer
announcement returns and completion rates. Luo (2005), however, finds that completion
rates are positively related to acquirer announcement returns in a larger sample. Chen,
Harford, and Li (2005) find this relationship only among acquirers with concentrated
ownership by long-term, independent institutional investors.36 Paul (2005) finds that
independent boards and outside blockholders are negatively related to completion rates of
poorly received acquisitions. She also finds that acquirers that complete poorly received
Another example is the aborted acquisition of IMS Health by VNU. See IMS Health Incorporated Current Report on Form 8-K dated November 17, 2005.
36 If the main concern is meeting a quorum, institutions can help because they are obligated to vote (Del Guercio (1996)).
Hosted by The Berkeley Electronic Press
40
acquisitions are likely to downsize, experience chief executive turnover, become takeover
targets, or go private, and that the likelihood of downsizing or experiencing chief executive
turnover is higher for acquirers with independent boards.
While none of these studies singles out acquisitions that need shareholder approval,
Luo (2005) reports that the positive relation between announcement returns and
completion rates is present only for the 302 acquisitions in his sample that were announced
without a binding agreement, and interprets this as evidence that acquirers abandon poorly
received acquisitions when the cost is low. That these acquisitions were abandoned, rather
than structured in a way that would obviate the need for acquirer shareholder approval,
suggests that what motivated management was the fear of market discipline, rather than the
fear of losing the vote.37
If voting is an effective constraint on acquisitions, and managers do not perfectly
predict shareholder reaction to acquisition announcements, the relation between acquirer
announcement returns and acquisition completion should be more pronounced in
acquisitions requiring shareholder approval. I test this hypothesis using a probit model in
which dependent variable is an indicator for whether the acquisition was completed, and
the depended variables are CAR, VOTE, and their interaction. In a modified specification,
37 A reminder of the force of market discipline is the 2005 ouster of Deutsche Börse’s chief executive officer following his insistence on making a bid for the London Stock Exchange which, under German law, did not require acquirer shareholder approval. Notably, the ouster was prompted by shareholder threat to vote the entire board out of office in the annual shareholder meeting (Cohen and Jenkings (2005)). In the United States, voting the board out of office is costly because it requires the dissident shareholders to nominate a competing slate of director candidates in a proxy statement of their own. A contemporaneous example is the threat by Euronext shareholders to oust the chief executive officer if he proceeded with a proposed bid for the London Stock Exchange (Cohen (2005)). That acquisition, however, would also need shareholder approval (Reilly (2005)).
http://law.bepress.com/alea/16th/art64
41
I add variables likely to affect the probability of shareholder rejection of the acquisition:
an indicator for acquisitions requiring approval by a majority or a supermajority of the
outstanding votes (ABSOLUTE), its interaction with CAR, and an indicator for
acquisitions in which acquirer shareholders holding at least 40% of the voting committed
their support (GUARANTEED).
Table 8 summarizes the results. As suggested by the fact that 81% of withdrawn
acquisitions require shareholder approval, there is a significant negative relation between
shareholder approval and the probability of completion in all specifications. This finding
is consistent with the approval requirement causing deal withdrawal due to shareholder
opposition or delay. It is also consistent, however, with the possibility that another
characteristic of acquisitions requiring approval, such as acquirer financial condition,
causes the withdrawal. Announcement returns are not significantly related to the
probability of completion regardless of whether shareholder approval is required,
suggesting that if shareholder opposition causes the withdrawal, the opposition does not
form immediately after the announcement. The probability of completion is not
significantly related to the presence of guaranteed shareholder support, nor is it
significantly related to an absolute majority requirement, failing to support the shareholder
opposition hypothesis. The results do not materially change when, in addition to including
acquirer returns and a vote indicator as standalone regressors, I interact each with the
portion of acquirer stock held by the five largest mutual funds, independent investment
advisors, or public pension funds.
Hosted by The Berkeley Electronic Press
42
VI. Conclusion
[It is too early to conclude. Add a conclusion after completing data collection and
analysis.]
http://law.bepress.com/alea/16th/art64
43
References
American Bar Association, 2006, Changes in the Model Business Corporation Act —
Proposed amendments to Chapters 8 and 10 relating to voting by shareholders for
the election of directors, Business Lawyer (forthcoming 2006).
Amihud, Yakov, Baruch Lev, and Nickolaos G. Travlos, 1990, Corporate control and the
choice of investment financing: The case of corporate acquisitions, Journal of
Finance 55, 603–616.
Anabtawi, Iman, 2006, Some skepticism about increasing shareholder power, UCLA Law
Review 53, 561–599.
Ang, James S., and Yingmei Cheng, 2006, Direct evidence on the market-driven
acquisitions theory, Journal of Financial Research 29, 199–216.
Asquith, Paul, Robert F. Bruner, and David W. Mullins, Jr., 1987, Merger returns and the
form of financing, Proceedings of the Seminar on the Analysis of Security Prices
34, 115–146.
Avery, Christopher, Judith A. Chevalier, and Scott Schaefer, 1998, Why do managers
undertake acquisitions? An analysis of internal and external rewards for
acquisitiveness, Journal of Law, Economics, and Organization 14, 24–43.
Ayers, Benjamin C., Craig E. Lefanowicz, and John R. Robinson, 2003, Shareholder taxes
in acquisition premiums: The effect of capital gains taxation, Journal of Finance
57, 2783–2801.
Bainbridge, Stephen M., 2006, The case for limited shareholder voting rights, UCLA Law
Review 53, 601–636.
Hosted by The Berkeley Electronic Press
44
Balachandran, Sudhakar V., Peter Joos, and Joseph Weber, 2004, Do voting rights matter?:
Evidence from the adoption of equity-based compensation plans, Working paper,
Columbia University.
Barclay, Michael J., and Clifford G. Holderness, 1991, Negotiated block trades and
corporate control, Journal of Finance 46, 861–878.
Bebchuk, Lucian Arye, 2005, The case for increasing shareholder power, Harvard Law
Review 118, 833–917.
___________, 2005a, The myth of the shareholder franchise, Working paper, Harvard
University.
Bebchuk, Lucian A., and Yaniv Grinstein, 2005, Firm expansion and CEO pay, Working
paper, Harvard University.
Berman, Dennis K., and Jason Singer, 2005, VNU’s largest shareholders deride plan to
buy IMS Health, Wall Street Journal, October 1, 2005, B3.
_____________, 2005, VNU, IMS cancel acquisition pact, Wall Street Journal, November
17, 2005, A10.
Bethel, Jennifer E., Julia Porter Liebeskind, and Tim Opler, 1998, Block share purchases
and corporate performance, Journal of Finance 53, 605–634.
Bethel, Jennifer E., and Stuart L. Gillan, 2002, The impact of the institutional and
regulatory environment on shareholder voting, Financial Management 31, 29–54.
Black, Bernard S., 1989, Bidder overpayment in takeovers, Stanford Law Review 41, 597–
660.
http://law.bepress.com/alea/16th/art64
45
Brickley, James A., Ronald C. Lease, and Clifford W. Smith, Jr., 1994, Corporate voting:
Evidence from charter amendment proposals, Journal of Corporate Finance 1, 5–
31.
Burch, Timothy R., Angela G. Morgan, and Jack G. Wolf, 2004, Is acquiring-firm
shareholder approval in stock-for-stock mergers perfunctory?, Financial
Management 33, 45–69.
Chang, Saeyoung, 1998, Takeovers of privately held targets, methods of payment, and
bidder returns, Journal of Finance 53, 773–784.
Chemmanur, Thomas, and Imants Paeglis, 2003, The choice of the medium of exchange in
acquisitions: A direct test of the double-sided asymmetric information hypothesis,
Working paper, Boston College.
Chen, Xia, Jarrad Harford, and Kai Li, 2005, Monitoring: Which institutions matter?,
Working paper, University of British Columbia.
Christofferson, Susan E. K., Christopher C. Géczy, David K. Musto, and Adam V. Reed,
2005, Vote trading and information aggregation, Working paper, McGill
University.
Coffee, John C., Jr., 1984, Regulating the market for corporate control: A critical
assessment of the tender offer’s role in corporate governance, Columbia Law
Review 84, 1145–1296.
Cohen, Norma, and Patrick Jenkins, 2005, Comment & Analysis, Financial Times, May
25, 2005, 17.
Cohen, Norma, 2005, Euronext investors warn chief on LSE bid, Financial Times, October
vote for Sovereign deal, Wall Street Journal, November 23, 2005, C1.
Louis, Henock, 2004, The cost of using bank mergers as defensive mechanisms against
takeover threats, Journal of Business 77, 295–310.
Luo, Yuanzhi, 2005, Do insiders learn from outsiders? Evidence from mergers and
acquisitions, Journal of Finance 60, 1951–1982.
Maksimovic, Vojislav, and Gordon Phillips, 2001, The market for corporate assets: Who
engages in mergers and asset sales and are there efficiency gains?, Journal of
Finance 61, 2019–2065.
Maloney, Michael T., Robert E. McCormick, and Mark L. Mitchell, 1993, Managerial
decision making and capital structure, Journal of Business 66, 189–217.
Malmendier, Ulrike, and Geoffrey Tate, 2005a, CEO overconfidence and corporate
investment, Journal of Finance 60, 2661-2700.
Hosted by The Berkeley Electronic Press
50
__________, 2005b, Who makes acquisitions? A test of the overconfidence hypothesis,
Working paper, Stanford University.
Martin, Kenneth J., 1996, The method of payment in corporate acquisitions, investment
opportunities, and management ownership, Journal of Finance 51, 1227–1246.
Maquieira, Carlos P., William L. Megginson, and Lance Nail, 1998, Wealth creation
versus wealth redistribution in pure stock exchange mergers, Journal of Financial
Economics 48, 3–33.
Mikkelson, Wayne H., and Richard S. Ruback, 1985, An empirical analysis of the
interfirm equity investment process, Journal of Financial Economics 14, 523–553.
Mitchell, Mark, Todd Pulvino, and Erik Stafford, 2004, Price pressure around mergers,
Journal of Finance 59, 31–63.
Moeller, Sara B., Frederik P. Schlingemann, and René M. Stulz, 2004, Firm size and the
gains from acquisitions, Journal of Financial Economics 73, 201–228.
__________, 2005, Wealth destruction on a massive scale? A study of acquiring-firm
returns in the recent merger wave, Journal of Finance 60, 757–782.
Morck, Randall, Andrei Shleifer, Robert W. Vishny, 1990, Do managerial objectives drive
bad acquisitions?, Journal of Finance 45, 31–48.
Morgan, Angela G., and Annette B. Poulsen, 2001, Linking pay to performance —
compensation proposals in the S&P 500, Journal of Financial Economics 62, 489–
523.
Paul, Donna L., 2005, Board composition and corrective action: Evidence from corporate
responses to bad acquisition bids, Working paper, Washington State University.
http://law.bepress.com/alea/16th/art64
51
Qiu, Lily, 2004, Which institutional investors monitor? Evidence from acquisition activity,
Working Paper 2004-21, Brown University.
Rau, P. Raghavendra, and Theo Vermaelen, 1998, Glamour, value and the post-acquisition
performance of acquiring firms, Journal of Financial Economics 49, 223–253.
Reilly, David, 2005, Exchanges tango in Europe, Wall Street Journal, September 19, 2005,
C12.
Rhodes–Kropf, Matthew, David T. Robinson, and S. Viswanathan, Valuation waves and
merger activity: The empirical evidence, Journal of Financial Economics 77, 561–
603 (2005).
Roll, Richard, 1986, The hubris hypothesis of corporate takeovers, Journal of Business 59,
197–216.
Rosen, Richard J., 2006, Merger momentum and investor sentiment: The stock market
reaction to merger announcements, Journal of Business 79, 987–1017.
Servaes, Henri, 1991, Tobin’s Q and the gains from takeovers, Journal of Finance 66,
409–419.
Shleifer, Andrei, and Robert W. Vishny, 2003, Stock market driven acquisitions, Journal
of Financial Economics 70, 295–311.
Shabelman, David, 2005, eBay pays cash for Rent.com, Corporate Control Alert, March
2005, 7–8.
Singer, Jason, 2005, VNU investors maintain pressure, Wall Street Journal, November 18,
2005, B2.
Schwert, G. William, 1996, Markup pricing in mergers and acquisitions, Journal of
Financial Economics 41, 153–192.
Hosted by The Berkeley Electronic Press
52
Slovin, Myron B., Marie E. Sushka, and John A. Polonchek, 2005, Methods of payment in
asset sales: Contracting with equity versus cash, Journal of Finance 60, 2385–
2407.
Officer, Micah S., 2004, Collars and renegotiation in mergers and acquisitions, Journal of
Finance 59, 2719–2743.
Officer, Micah S., 2006, The price of corporate liquidity: Acquisition discounts for
unlisted targets, Journal of Financial Economics, forthcoming.
Travlos, Nickolaos G., 1987, Corporate takeover bids, method of payment, and bidding
firms’ stock returns, Journal of Finance 42, 943–963.
Wansley, James W., William R. Lane, and Ho C. Yang, 1987, Gaines to bidder firms in
cash and securities transactions, Financial Review 22, 403–414.
Yook, Ken C., 2003, Larger return to cash acquisitions: Signaling effect or leverage
effect?, Journal of Business 76, 477–498.
http://law.bepress.com/alea/16th/art64
53
Table 1: Distribution of Acquisitions by Industry, Voting, and Acquirer Type
This table reports the number of successful and unsuccessful acquisitions announced between January 1, 1995 and December 31, 2003 for which voting information is available. Agriculture includes targets whose primary 2-digit SIC code is 2. Mining includes targets whose primary 2-digit SIC code is 10 or 12. Oil and gas extraction includes targets whose primary 2-digit SIC code is 13. Construction includes targets whose primary 2-digit SIC code is 15, 16, or 17. Manufacturing includes targets whose primary 1-digit SIC code is 2 or 3. Transportation includes targets whose primary 2-digit SIC code is 40, 42, 44, 45, 46, or 47. Communications includes targets whose primary 2-digit SIC code is 48. Utilities includes targets whose primary 2-digit SIC code is 49. Banks includes targets whose primary 2-digit SIC code is 60 or whose primary 4-digit SIC code is 6712. Insurance includes targets whose primary 2-digit SIC code is 63 or 64. Other financial includes targets whose primary 2-digit SIC code is 62 or 67, or whose primary 4-digit SIC code is 6719, 6726, 6794, 6798, or 6799. Real estate includes targets whose primary 2-digit SIC code is 65. Services includes targets whose primary 1-digit SIC code is 7 or 8.
This table reports mean and median values of transaction characteristics for acquisitions requiring shareholder approval and acquisitions not requiring shareholder approval. Acquirer capitalization is the stock market capitalization of the acquirer one day before the announcement. Deal value is the value of consideration reported by SDC. Size is the ratio of the deal value to acquirer capitalization one trading day before the announcement. All-stock is an indicator for payment only in common stock. Percent Stock is the percentage of acquirer common stock of the consideration. Tax-free is an indicator for tax-free acquisitions. Registered is an indicator for acquisitions involving the issuance of registered stock. Duration is deal duration from announcement to closing. Completed is an indicator for completed acquisitions. Tender is an indicator for tender offers. Private is an indicator for private targets. CAR is the cumulative abnormal returns to acquirer shareholders in the 5-trading day window around the announcement date. Premium is the ratio of deal value to target stock market capitalization. Asterisks indicate the significance of the difference between means of the two samples in a Wilcoxon test for interval variables and a Chi-square test for indicator variables. Significance (p-value): * 10%, ** 5%, *** 1%.
This table reports coefficient estimates and, in parentheses, the significance (p-value) from estimating an ordinary least squares model in which the dependent variable is deal duration from announcement to closing. All regulated is an indicator for acquisitions whose primary 2-digit SIC code is 40, 44, 45, 47, 48, 49, 60, 63, or 64, or whose primary 4-digit SIC code is 6712. Bank is an indicator for acquisitions of targets whose primary 2-digit SIC code is 60 or whose primary 4-digit SIC code is 6712. Communications is an indicator for acquisitions of targets whose primary 2-digit SIC code is 48. Other regulated is an indicator for acquisitions of targets whose primary 2-digit SIC code is 40, 44, 45, 47, 49, 63, or 64 (railroad transportation, water transportation, transportation by air, transportation services, electric, gas, and sanitary services, insurance carriers, and insurance agents, brokers, and service). Vote is an indicator for acquisitions that need shareholder approval. Estimates are based on Huber-White robust standard errors. Significance (p-value): * 10%, ** 5%, *** 1%.
This table reports marginal effects and, in parentheses, the significance (p-value) from estimating a probit model in which the dependent variable is the presence of a shareholder approval requirement. Bank is an indicator for targets whose primary 2-digit SIC code is 60 or whose primary 4-digit SIC code 6712. Private is an indicator for private targets. Unregistered is an indicator for acquisitions involving the issuance of unregistered stock. Tender is an indicator for tender offers. Leverage is the sum of the acquirer’s face value of debt before the announcement plus the deal value (including assumed liabilities) divided by the sum of the book value of total assets before the announcement plus the deal value (including assumed liabilities). Size is the ratio of the deal value to acquirer capitalization one trading day before the announcement. Institutions is the portion of acquirer outstanding common stock held by institutional investors. Defensive is an indicator for acquirers that were actual or rumored targets in the two years preceding the announcement. Estimates are based on Huber-White robust standard errors. Significance (p-value): * 10%, ** 5%, *** 1%.
This table reports coefficient estimates and, in parentheses, the significance (p-value) from estimating an ordinary least squares model in which the dependent variable is deal duration from announcement to closing. The first column presents results for all acquisitions combined. The second column presents results for acquisitions of non-banks. The third column presents results for bank acquisitions. Banks are firms whose primary 2-digit SIC code is 60 or whose primary 4-digit SIC code is 6712. Vote is an indicator for acquisitions that need shareholder approval. Private is an indicator for private targets. Unregistered is an indicator for acquisitions involving the issuance of unregistered stock. Tender is an indicator for tender offers. Estimates are based on Huber-White robust standard errors. Significance (p-value): * 10%, ** 5%, *** 1%.
This table reports coefficient estimates and, in parentheses, significance levels (p-values) from estimating an ordinary least squares model (columns (1) and (2)) and a two stage least squares model (columns (3) and (4)) in which the dependent variable is the cumulative abnormal returns to acquirer shareholders in the 5-trading day window around the announcement date. Vote is an indicator for acquisitions that need shareholder approval. In columns (3) and (4), it is instrumented by the sum of the acquirer’s face value of debt before the announcement plus the deal value (including assumed liabilities) divided by the sum of the book value of total assets before the announcement plus the deal value (including assumed liabilities), and by the ratio of the deal value to acquirer capitalization one trading day before the announcement. Portion is the portion of acquirer outstanding common stock issued in the acquisition. Percent Stock is the percentage of acquirer common stock in the consideration. Private is an indicator for private targets. Large is an indicator for acquirers whose market capitalization one trading day before the announcement is in the top quartile of firms listed on the New York Stock Exchange on the last trading day of the announcement year. Defensive is an indicator for acquirers that were mentioned in the financial press as actual or rumored targets within two years before the announcement. Diversify is an indicator for targets whose two-digit SIC code is different from the acquirer’s. All regressions include year fixed effects. Estimates are based on Huber-White robust standard errors. Significance (p-value): * 10%, ** 5%, *** 1%.
This table reports coefficient estimates and, in parentheses, significance levels (p-values) from estimating an ordinary least squares model in which the dependent variable is the ratio of deal value to target stock market capitalization. Vote is an indicator for acquisitions that need shareholder approval. Percent Stock is the percentage of acquirer common stock in the consideration. Private is an indicator for private targets. Size is the ratio of the deal value to acquirer capitalization one trading day before the announcement. Large is an indicator for acquirers whose market capitalization one trading day before the announcement is in the top quartile of firms listed on the New York Stock Exchange on the last trading day of the announcement year. Defensive is an indicator for acquirers that were mentioned in the financial press as actual or rumored targets within two years before the announcement. Excess pay is the residual from regressing the acquirer chief executive officer’s salary and bonus on two-digit SIC industry indicator variables, the log of market capitalization, year indicators, and the prior year stock return. Announcement year fixed effects and industry fixed effects based on 2-digit SIC code of the target are included in the regression but not reported. Estimates are based on Huber-White robust standard errors. Significance (p-value): * 10%, ** 5%, *** 1%.
This table reports marginal effects and, in parentheses, the significance (p-value) from estimating a probit model in which the dependent variable is an indicator for completed acquisitions. CAR is cumulative abnormal returns to acquirer shareholders in the 5-trading day window around the announcement date. Vote is an indicator for acquisitions that need shareholder approval. Guaranteed is an indicator for acquisitions in which acquirer shareholders holding at least 40% of the votes commit to support the deal. Absolute is an indicator for acquisitions requiring a majority or a supermajority of the outstanding votes. Estimates are based on Huber-White robust standard errors. Significance (p-value): * 10%, ** 5%, *** 1%.