!Корпоративное право 2023-2024 / Bargeron L._Do Shareholder Tender Agreements Inform or Expropriate Shareholders

Do Shareholder Tender Agreements Inform or Expropriate Shareholders?◊

Leonce Bargeron

Katz Graduate School of Business, University of Pittsburgh, Pittsburgh, PA 15260.

Draft: September 26, 2006

________________________________________________________________________

Abstract

By signing a Shareholder Tender Agreement (STA) a shareholder pre-commits to tender her shares to a particular bidder, forsaking the right to tender to any subsequent bidder. In a representative sample between 1995 and 2001, 60% of the tender offers contain an STA. The data indicate that STA deals are associated with lower premiums, greater ownership concentration, greater management ownership, and greater information asymmetry. These results suggest that STAs certify value to uninformed shareholders, thereby increasing the efficiency of the tender offer process. The evidence also distinguishes STAs from termination fees and lockup options, both empirically and operationally.

JEL classification: G34; D82

Keywords: Tender offer; Merger; Asymmetric information; Certification; Bid premium

◊ I am grateful to Paolo Fulghieri, Anil Shivdasani, Eitan Goldman, Matthias Kahl, Wayne Landsman, and Steve Slezak for their insights. I also thank Gregg Brown, David Guilkey, Rob Hansen, Ken Lehn, David Ravenscraft, Fredeik Schlingemann, Shawn Thomas, Harold Zhang, and seminar participants at Babson College, Boston College, the University of Calgary, the University of Cincinnati, Georgia Institute of Technology, the University of North Carolina, the University of Pittsburgh, the Securities and Exchange Commission, and the Stockholm School of Economics for their helpful comments. All remaining errors are my own.

Tel.: + 1 412 648 1642; fax: 1 412 648 1693. E-mail address: LLBargeron@katz.pitt.edu

Electronic copy available at: http://ssrn.com/abstract=939747

Given Buffett’s success as a value investor, could he now face resistance from investors who don’t want to sell to him at a depressed price? “The fact that it’s Warren Buffett who wants to buy from you should tell you that you shouldn’t sell, at least not at his price”, quips Dorr.

BusinessWeek - September 8, 2003

1. Introduction

This paper investigates an intriguing contract between the bidder in a tender offer

and at least one shareholder in the target firm. In the contract, which I refer to as a

Shareholder Tender Agreement (STA), a shareholder in the target firm commits to tender

her shares to a particular bidder regardless of future events.1 By signing the STA, the

shareholder relinquishes her right to tender to a dominating competing bid, and hence

forgoes potential future profits. In addition, while prior to signing the shareholder incurs

the expenses associated with valuing the deal and negotiating the contract, the

shareholder receives no additional compensation for signing an STA as, by law, the same

price must be paid for each tendered share.2 Notwithstanding, STAs are present in 60% of

the tender offers in a representative sample that spans the 1995 to 2001 period.

The following excerpt from Berkshire Hathaway Inc.’s Securities and Exchange

Commission (SEC) filing in its tender offer for XTRA Corporation illustrates the terms

typical of an STA:3

Concurrently with the execution of the Merger Agreement, Parent and Purchaser entered into the Stockholders Agreement [STA] with Tiger Management L.L.C. and Tiger Performance L.L.C. (the "Tiger Advisers"), and with Tiger Management Corporation and Julian H. Robertson, Jr. (collectively with the Tiger Advisers, the "Signatory Holders"). The Tiger Advisers are the beneficial owners of 3,175,594 Shares, or approximately 30% of the issued and outstanding Shares.

1Throughout the paper, to avoid confusion I refer to the bidder and the non-pivotal shareholders using the male gender and I refer to the STA signatory using the female gender.

2Regulation 14d-10 of the Securities Exchange Act of 1934 states that no bidder shall make a tender offer unless the consideration paid to any security holder pursuant to the tender offer is the highest consideration paid to any other security holder during such tender offer.

3The complete description of the agreement as given in the 14-D filing with the SEC is included in Appendix A.

Electronic copy available at: http://ssrn.com/abstract=939747

Pursuant to the Stockholders Agreement, the Signatory Holders have agreed to tender in the Offer, prior to the initial expiration date of the Offer, all Shares owned beneficially and of record by the Tiger Advisers. The Signatory Holders have also agreed to vote the Tiger Advisers' Shares in favor of the Merger and the Merger Agreement …

As stated, just prior to the offer the Tiger Advisers committed to tender their entire 30% stake in XTRA to Berkshire Hathaway. But the bid was only 5% above the stock price prior to the offer. This raises the following question: why does a sophisticated investor surrender the option to tender to the highest bidder?

To answer the above question, this paper analyzes a sample of 525 tender offers announced between 1995 and 2001. I find that the characteristics of offers that include an STA differ substantially from those that do not. On average, non-STA offer premiums are 20% higher than STA offer premiums. In addition, the typical target firm in an STA offer is younger and has both greater ownership concentration and greater management ownership than the typical non-STA target firm.

Comparing these results to the predictions of three hypotheses helps discern the primary motivation for STAs. First, the certification hypothesis, proposed in Bargeron (2006), asserts that an STA reduces adverse selection caused by information asymmetry between an informed bidder and uninformed shareholders, increasing the efficiency of the tender offer process. This hypothesis predicts that in turn, the STA benefits all parties. According to the certification hypothesis, Tiger Management signed the costly STA to convince the less informed shareholders that the bid, although only 5% above the preannouncement price, was indeed fair.

Second, the bargaining hypothesis suggests that shareholders receive a higher bid premium in exchange for pre-committing shares to a particular bidder. This hypothesis predicts that all shareholders benefit from the STA. According to the bargaining

1

hypothesis, Tiger Management used the STA as a bargaining chip in an effort to secure a higher bid from Berkshire Hathaway.

Finally, the expropriation hypothesis posits that the bidder and the signatory shareholders use an STA to expropriate value from the remaining shareholders: the bidder offers private benefits to the shareholders who sign the STA, and in return the signatories agree to a lower premium. This hypothesis predicts that the STA enriches a small subset of shareholders at the expense of the remaining shareholders. According to the expropriation hypothesis, Berkshire Hathaway and the Tiger Advisers conspired to expropriate value from the owners of the remaining 70% of XTRA.

The evidence lends support to the certification hypothesis. Specifically, STAs are associated with lower premiums, greater information asymmetry, greater ownership concentration, and greater management ownership, each of which is consistent with certification. In contrast, the lower premiums are inconsistent with the bargaining hypothesis and the ownership evidence is inconsistent with the expropriation hypothesis. I therefore conclude that an STA facilitates an efficient takeover by certifying a fair bid, with the costly public commitment serving to convince uninformed target shareholders that tendering is optimal. Thus, notwithstanding Mr. Dorr’s BusinessWeek conjecture, when a well-informed investor whose interests are aligned with yours commits to tender, perhaps you should sell, even at Mr. Buffett’s price.

Note that the analysis also distinguishes STAs from termination fees and lockup options. Indeed, the results suggest that STAs serve an altogether different purpose than termination fees and lockup options. An STA reduces information asymmetry between the bidder and the target firm’s shareholders (certification). In contrast, termination fees and lockup options serve to enhance target bargaining power (Burch, 2001), to

2

compensate the bidder for externalities resulting from the bid (Bates and Lemmon, 2003 and Officer, 2003), to resolve a holdup problem (Hotchkiss, Qian, and Song, 2005), or to reward a bidder for forgoing a toehold (Betton, Eckbo, and Thorburn, 2005). Moreover, STAs are empirically quite distinct from termination fees and lockup options. In particular, STAs are associated with lower average takeover premiums while prior literature finds that termination fees and lockup options are associated with higher takeover premiums.

The remainder of the paper proceeds as follows. Section 2 provides institutional background and describes the disclosure requirements for tender offers. Section 3 defines an STA and develops the predictions of the three hypotheses. Section 4 describes the data and Section 5 tests the hypotheses. Finally, Section 6 concludes.

2. Institutional background and disclosure requirements for tender offers

The Securities and Exchange Commission (SEC) regulates the tender offer process. As prescribed by law, a tender offer must remain open for at least 20 business days. This waiting period ensures that all interested parties, including potential competing bidders, have time to evaluate the offer and respond accordingly. Further, during the offer period a shareholder can withdraw any shares that he previously tendered. A mandatory, publicly available 14D-1 filing with the SEC explains the specifics of each tender offer. This filing must include details such as offer price, method of payment, and conditions for acceptance, as well as a description of all contracts and arrangements concerning the deal.

In addition to federal laws and disclosure requirements, various state laws also govern the tender offer process. Of particular relevance for many tender offers, state law

3

determines the ownership threshold required for a short-form merger. In a short-form merger a bidding firm acquiring at least the threshold percentage of the target firm can actually consummate the merger without a shareholder vote. In Delaware, for example, this threshold is 90%. Note that the possibility of a short-form merger motivates many tender offers. In fact, contrary to the view that tender offers permit acquirers to avoid negotiations with target management, approximately 90% of the offers in my sample include a merger agreement that is pre-approved by the boards of directors of both the bidder and the target. Generally, as stated in the SEC filing, a primary goal of these tender-mergers is for the acquirer to purchase enough shares to execute a short-form merger. This enables the acquirer to hasten the merger process, decrease takeover expenses, and mitigate the uncertainty involved in a shareholder vote.4

Merger and tender offer documents frequently include the details of other contracts that share some of the features of STAs. For instance, both lockup options and termination fees confer an advantage to one bidder over other bidders, as does an STA, that is, in the event of a successful competing bid, all three contracts transfer value from the shareholders to the initial bidder. However, the shareholders that bear these costs vary across the contract types. With lockup options and termination fees all shareholders bear the costs in proportion to their ownership. Essentially, by negotiating these contracts management is risking all the shareholders’ money. With an STA, on the other hand, only the shareholders signing the agreement stand to lose. By signing an STA, the signatories are “putting their own money where their mouths are.”

Finally, there is a distinction between an STA and direct bidder ownership in the target firm (toehold). Although the bidder controls the shares in both cases, with a

4 However, structuring the deal as a tender-merger can create financing difficulties. See Reed and Lajoux (1998).

4

toehold the bidder realizes both the appreciation of target shares caused by his own bid and any decrease in target value in the event of a failed takeover,5 whereas with an STA the bidder does not realize the resulting share appreciation nor is he exposed to any depreciation in target value in the event of a failed takeover.

3. STA hypotheses and predictions

In this section I formally define an STA. I then develop three hypotheses motivating the existence of STAs and derive testable implications for each. Because differences in testable implications empirically distinguish one hypothesis from the others, one hypothesis is likely to be more consistent with the data than the others.

3.1. Definition of an STA

An STA is among the agreements that must be disclosed in a 14D-1 tender offer SEC filing. However, the name given to an STA within the 14D-1 varies across deals. Most commonly, STAs are referred to as Voting Agreements, Shareholder Agreements, or Tender Agreements. In this paper I define an STA as any agreement that satisfies all of the following three criteria:

1.Prior to the tender offer, the agreement is signed by the bidder and at least one shareholder in the target firm.

2.The signatory shareholder commits to tender her shares to the bidder and forgo all withdrawal rights.6

5Goldman and Qian (2005) model the optimal toehold choice considering the cost of an unsuccessful takeover. Bradley, Desai, and Kim (1983) find that target firm value decreases in the event of an unsuccessful takeover.

6This condition is also met when the signatory shareholder commits to sell her shares to the bidder at the tender offer price but consummation of the sale is conditional on the outcome of the tender offer.

5

3. Until the deal is consummated or the bidder withdraws the offer, the signatory shareholder commits to support the deal in all shareholder votes.

Note that the STA shareholder must tender her shares to the STA bidder even if a higher competing bid wins the takeover contest. In this case, if the STA bidder does not receive enough shares to satisfy his minimum condition, he still has the option to accept any tendered shares in exchange for the STA bid price.

3.2. Certification hypothesis

The first of the three hypotheses, the certification hypothesis, focuses on the information asymmetry between an informed bidder and uninformed target shareholders. Grossman and Hart (1981) and Shleifer and Vishny (1986) demonstrate that a bidder can overcome the free rider problem in tender offers even when the bidder’s information is superior to the shareholders’ information. However, the resulting takeover process is not efficient. In each of these models, low value bidders cannot afford the resulting equilibrium pooling bid and cannot credibly separate themselves from the higher value bidders. Therefore, low value bidders do not bid. The certification hypothesis suggests that STAs evolved to alleviate this inefficiency.

An STA credibly certifies a bidder’s takeover valuation, reducing the information asymmetry and the accompanying adverse selection. In effect, a bid certified by an STA allows the uninformed shareholders to infer the underlying value of the deal, that is, the uninformed shareholders become informed. Thus, bidders use an STA to signal their true type. Low value bidders, who could not profitably bid otherwise, negotiate an STA to overcome adverse selection and successfully take over the target; in contrast, high value

6

bidders overcome adverse selection without incurring the costs of an STA. Appendix B presents a simple example of the certification hypothesis.

Note that the certification hypothesis depends critically on the assumption of no side payments between the bidder and the shareholder who signs the STA.7 Without this assumption the interests of the informed and uninformed shareholders do not necessarily coincide, in which case the STA would no longer credibly certify the underlying value of the deal to the uninformed shareholders and the certification hypothesis would unravel. In the following subsection, I introduce the expropriation hypothesis, which relaxes the assumption of no side payments, to test the validity of this assumption.

The certification hypothesis generates three testable predictions. First, STA offers are associated with lower bid premiums than non-STA offers. This is attributable to selfselection: in equilibrium only low synergy value bidders negotiate an STA. Second, information asymmetry is positively related to the probability of an STA. Greater information asymmetry leads to greater adverse selection, increasing the probability that an STA will be necessary to overcome the adverse selection. Third, ownership concentration is positively related to the probability of an STA. Increasing a shareholder’s ownership increases her profits in the event of a successful takeover and hence the greater a shareholder’s ownership the more willing she is to incur the costs of an STA.

3.3. Bargaining hypothesis and expropriation hypothesis

7 Laws against defrauding shareholders support this assumption. The expected costs and penalties of such fraud are substantial. This is particularly true when board members, who have professional reputation concerns and are subject to greater shareholder scrutiny, are among the signatories of the STA.

7

The remaining two hypotheses are closely related to the hypotheses regarding termination fees and lockup options developed in Bates and Lemmon (2003), Burch (2001), and Officer (2003). Both hypotheses result from the advantage an STA confers on a particular bidder. For the bidder signing the contract, an STA reduces the portion of uncommitted shares necessary for a successful takeover. However, for all other bidders the STA increases the portion of uncommitted shares necessary for a successful takeover. As a result, an STA confers an advantage to the signatory bidder. Determining which shareholders benefit from the assignment of this advantage distinguishes the two hypotheses.

The bargaining hypothesis posits that all shareholders benefit from the STA. In particular, the signatory shareholder extracts a higher bid from the bidder in return for an STA, and thus all shareholders receive larger payoffs from the takeover. This hypothesis yields the following testable prediction: STA offers are associated with higher bid premiums than non-STA offers.

Alternatively, under the expropriation hypothesis the majority of shareholders do not benefit from the STA. Instead, the large shareholder negotiates an STA with a “sweetheart” bidder in return for private benefits at the expense of the other shareholders. These benefits can take the form of job security agreements, such as employment contracts, monetary compensation, such as non-competition agreements and golden parachutes, or even non-pecuniary compensation. This hypothesis relaxes the assumption of no side payments that underlies the certification hypothesis.

The expropriation hypothesis generates two testable predictions. First, STA offers are associated with a lower bid premium than non-STA offers. Because the parties involved in the STA expropriate a portion of the surplus, less of the surplus is available

8