Download - 63 Bus Law 729

Citation: 63 Bus. Law. 729 2007-2008

Content downloaded/printed from HeinOnline (http://heinonline.org)Mon Feb 13 21:01:42 2012

-- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and Conditions of the license agreement available at http://heinonline.org/HOL/License

-- The search text of this PDF is generated from uncorrected OCR text.

-- To obtain permission to use this article beyond the scope of your HeinOnline license, please use:

https://www.copyright.com/ccc/basicSearch.do? &operation=go&searchType=0 &lastSearch=simple&all=on&titleOrStdNo=0007-6899

Go-Shops vs. No-Shops in Private EquityDeals: Evidence and Implications

By Guhan Subramanian*

I"Look at this. Pet Products of America is trying

to use the Holiday Catalogue to satisfy their go-shop!"

From: Debevoise & Plimpton Private Equity Report (Fall 2006)

Go-shop provisions have changed the way in which private equity firms execute public-

company buyouts. While there has been considerable practitioner commentary on go-shops

over the past three years, this Article presents the first systematic empirical evidence on

the effect of this new dealmaking technology on deal pricing and deal process. Contrary to

the claims of prior commentators, I find that: (1) go-shops yield more search in aggregate

(pre- and post-signing) than the traditional no-shop route; (2) "pure" go-shop deals, in which

there is no pre-signing canvass of the marketplace, yield a higher bidder 17 percent of the

time; and (3) target shareholders receive approximately 5 percent higher returns through

* Joseph Flom Professor of Law and Business, Harvard Law School; Douglas Weaver Professor of

Business Law, Harvard Business School. Comments welcome at [email protected]. I thank LucianBebchuk, David Bernstein, John Coates, Larry Hamermesh, Victoria lvashina, Hon. Jack Jacobs, Hon.Leo Strine, David Walker, and seminar participants at Harvard Law School, Harvard Business School,and New York University School of Law for comments on earlier drafts; Ranjan Ahuja and Rebecca Matafor excellent research assistance; Mark Morton, Esq., and Roxanne L. Houtman, Esq., of Potter, Ander-son & Corroon LLP for access to their data on go-shops; and the Olin Foundation at the Harvard LawSchool and the Division of Research at the Harvard Business School for generous funding.

730 The Business Lawyer; Vol. 63, May 2008

the pure go-shop process relative to the no-shop route. I also find no post-signing competitionin go-shop management buyouts ('MBOs'), consistent with practitioner wisdom that MBOsgive incumbent managers a significant advantage over other potential buyers. Taken as awhole, these findings suggest that the Delaware courts should generally permit go-shops asa means of satisfying a sell-side board's Revlon duties but should pay close attention to theirstructure, particularly in the context of go-shop MBOs.

The "go-shop" clause has emerged as an important new deal-making technol-ogy during the private equity boom of 2005-2007. Under the so-called Revlon'duty, the seller's board of directors must obtain the highest possible price in the saleof the company Traditionally, the board would satisfy its Revlon duty by canvassingthe market (through investment bankers), identifying serious bidders, holding aformal or informal auction among them, and signing a deal with the winning bid-der. The merger agreement would typically include a "no-shop" clause, which wouldprevent the target from talking to potential "deal jumpers," unless the target board'sfiduciary duty required it to do so (a "fiduciary out").2 The go-shop clause turnsthis traditional approach on its head: rather than canvassing the marketplace first,the seller negotiates with a single bidder, announces the deal, and then has thirty tofifty days to "go shop" to find a higher bidder.3 At the highest level, then, the tra-ditional route involves a market canvass followed by exclusivity with the winning

bidder; while the go-shop route in its pure form involves exclusivity with a bidderfollowed by a market canvass.

While there have been numerous practitioner commentaries on go-shop provi-sions in the three years since companies first used them,4 go-shop clauses havebeen largely ignored by academic commentators to date. This Article presents thefirst systematic empirical evidence on the effect of go-shop provisions in pri-vate equity deals, which account for the vast majority of all go-shops. I con-

struct a new sample of all going-private deals between January 2006 and August2007 that include a private equity sponsor (n= 141) and examine the solicitationfeature (no-shop or go-shop) in each deal. The data reveals two different kindsof go-shops: a "pure" go-shop, in which the seller negotiates exclusively with asingle buyer and then shops after the deal is announced; and what I term an "add-

on" go-shop, in which the go-shop provision is included after the target has al-ready conducted a pre-signing canvass of the marketplace. While the contractuallanguage in the merger agreement is generally the same across pure and add-ongo-shops, the important difference in the deal context has implications for theeconometric analysis that follows. Specifically, when I compare go-shop deals tothe traditional no-shop deals in my sample, I find that the pure go-shops (but notadd-on go-shops) yield approximately 5% higher returns to target shareholders.This finding stands in sharp contrast to the weight of practitioner commentary todate, which generally views the go-shop process as simply "window dressing" and

1. Revlon v. MacAndrews & Forbes Holdings, Inc., 506 A.2d 173, 182 (Del. 1986).2. See infra Part 1.3. See id.4. See infra Part II.

Go-Shops vs. No-Shops in Private Equity Deals 731

"illusory."' I suggest an explanation for this finding that focuses on the potentialbenefits of exclusivity for both the buyer and the seller from the go-shop processrelative to the traditional no-shop route.

The narrow (doctrinal) implication of these findings is that go-shop provisions,appropriately structured, can satisfy a target board's Revlon duties. The broader (trans-actional) implication is that go-shop provisions can be a "better mousetrap" in dealstructuring-a "win-win" for both buyer and seller. This conclusion would be con-sistent with the increasing use of go-shops over the past two years that I find in mysample. The final (policy) implication is that private equity firms are not stealingcompanies from the public shareholders at low-ball prices through go-shops, assome commentators suggest;6 rather, the go-shop process induces a full price fromthe first bidder, which is meaningfully shopped post-signing.

While the evidence presented here suggests no reason for categorical skep-ticism of go-shops, the data does indicate some reason to be wary in the spe-cific context of management buyouts ("MBOs"). Non-MBOs with a pure go-shopclause are jumped 23% of the time, while MBO go-shops are never jumped. Whilethe sample is small, this finding is consistent with practitioner impressions thatpotential bidders are generally unwilling to bid when management has publiclysigned on with a preferred buyout partner.7

Taken as a whole, these findings have implications for how sell-side boardsshould structure a meaningful go-shop process, and where the Delaware courtsshould focus their attention in determining whether a particular go-shop satis-fies the selling board's Revlon duties. To date, practitioners and courts have fo-cused on the length of the go-shop window and the magnitude of the breakupfee in assessing the viability of the go-shop process.8 The analysis presented heresuggests additional features that boards should negotiate for and courts shouldlook for, particularly in the context of MBOs: bifurcated breakup fees, no con-tractual match right or (even better) no ability to participate in the post-signingauction, a contractual commitment for the initial bidder to sell in to any higheroffer that emerges during the go-shop period, and ex ante inducement fees forsubsequent bidders, among other deal features. This proposal tracks the Delawarecourts' general approach to conflict transactions, which begins with substantivefairness review 9 but gives up fairness review if appropriate procedural protectionsare in place.'0

The remainder of this Article proceeds as follows. Part I provides backgroundon the private equity industry and the market factors that have influenced dealprocess over the past two years. Part II provides a chronology of the relevant

5. See infra note 74 and accompanying text.6. See id.7. See infra note 72 and accompanying text.8. See infra Parts II & Ill.9. See Weinberger v. UOP, Inc., 457 A.2d 701, 710-11 (Del. 1983); Kahn v. Lynch Commc'ns Sys.,

Inc., 638 A.2d 1110, 1115 (Del. 1994).10. See Faith Stevelman, Going Private at the Intersection of the Market and the Law, 62 Bus. LAw. 777,

783 (2007).


Delaware case law Part III reviews the prior literature. Part IV presents the meth-odology and empirical results. Part V suggests implications of the empirical find-ings for sell-side boards of directors, transactional lawyers, and the Delawarecourts. Part VI concludes.

I. BACKGROUND

It is well-known that private equity ("PE") has become a major factor in globalmergers and acquisitions ("M&A") over the past seven years. According to datafrom Dealogics M&A database, P/E deal volume has grown at an astonishing 45%rate peryear between 2001 and 2007. During this period P/E has gone from 6% oftotal global M&A deal volume to 24%. Despite this ferocious pace of investment,there remains a massive "overhang" of uninvested P/E capital, estimated to be$140 billion in November 200611 and $250 billion in April 2007.12 To the extentthat debt-to-equity ratios of three-to-one or four-to-one are still feasible today, thisoverhang means that P/E firms collectively have a purchasing power of approxi-mately $1.0 trillion. To put this number in perspective, the total public capital inthe United States is roughly $14 trillion, 3 suggesting that P/E investors have theability to take private 5-6% of current U.S. public capital (depending on assump-tions about premiums paid) before raising a single new dollar.

Of course, P/E firms are continuing to raise money, and investors are still clam-oring to give it to them despite the recent slowdown in the credit markets. A recentsurvey of managers of pension funds, foundations, and endowments indicatesthat these funds plan to increase their allocation to private equity over the nextthree years, from 4.0% to 6.3% of their total funds under management. 14 Theincrease translates into an additional $400 billion that is expected to flow intoprivate equity from baseline allocation levels that are already at unprecedentedhighs. This trend perhaps reflects the growing consensus among academics andpractitioners that P/E firms, unlike hedge funds," create significant value, on av-erage, for the companies in which they invest by attracting superior managers

11. See Erin White & Gregory Zuckerman, The Private-Equity CEO, WALL ST. J., Nov. 11, 2006, atC1 ("More deals are coming: Private-equity firms have raised more than $199 billion since the start of2005 but have spent only $56 billion.").

12. See Bank of America Business Capital, CapitalEyes, Top Tips to Make the Best Acquisition-or Not(Mar./Apr. 2007), http://corp.bankofamerica.com/public/public.portal?-pd-page-label=products/abf/capeyes/archive-index&dcCapEyes=indCE&id=349 (reporting "more than $250 billion" overhang).

13. The Dow Jones Wilshire 5000 index is generally considered to be the best measure for totalU.S. stock market capitalization. See DowJones Indexes, DowJones Wilshire Index Family, DowJonesWilshire 5000, http://www.djindexes.com/wilshire/index'cfm?go=key-benefits (last visited Apr. 13,2008). This index stood at approximately $14 trillion as of January 13, 2008. See Wilshire Indexes,http://www.wilshire.com/Indexes/broad/Wilshire5000 (last visited Jan. 13, 2008).

14. See Jenny Anderson, For Private Equity, the Party Isn't Over, N.Y. TIMES, Oct. 10, 2007, at C1(discussing Citigroup study of close to 50 pension managers).

15. See, e.g., Robin Greenwood & Michael Schor, Hedge Fund Investor Activism and Takeovers 4-5(Harvard Business School, Working Paper No. 08-004, July 2007) (reporting abnormal returns not dif-ferent from zero for hedge fund targets that remain independent a year after the activist investment).


with high-powered incentives 6 and then monitoring these managers with expe-rienced, repeat-play directors drawn from P/E firms that have significant "skin inthe game."17 The value proposition of this business model has only increased withthe enhanced disclosure and monitoring obligations imposed on public compa-nies by the Sarbanes-Oxley Act of 2002. t8

Another important feature of the P/E marketplace is the typical compensationstructure for the P/E general partners ("GPs"). P/E funds typically charge theirlimited partner ("LP") investors an annual management fee of 2% of assets undermanagement, plus a "carried interest" of 20-25% of any profits from the fund's in-vestments. 19 In the early days of private equity, when funds were relatively small,the 2% management fee was intended to pay the operating expenses of the P/Efirm ("keeping the lights on") and all of the "serious money" was to be in carriedinterest,20 thereby aligning the incentives of the P/E GPs with their LP investors.As fund sizes grew dramatically over time, management fee percentages droppedslightly2 but did not fully reflect the economies of scale that result from distribut-ing overhead expense over a larger fund base. The result is that management feestoday go well beyond "keeping the lights on" to become a large piece of overallcompensation for P/E firms. A recent study confirms this point, finding that P/E firmson average gained 60% of their overall compensation from management fees and only40% from the carried interest. 22 The implication is that P/E firms are unlikely to re-lease committed capital back to their limited partners, even if the recent downturn

16. See, e.g., Emily Thornton, Going Private, Bus. WK., Feb. 27, 2006, at 53, 54 ("The attractionsIfor top managers] are two-fold: money and freedom.").

17. See, e.g., Felix Barber & Michael Goold, The Strategic Secret of Private Equity, HARv. Bus. REv.,Sept. 2007, http://harvardbusinessonline.hbsp.harvard.edu/hbsp/hbr/articles/article.jsp?articlelD=R0709B&ml-action=get-article&print=true; Robert C. Pozen, If Private Equity Sized Up Your Business,HARv. Bus. REv., Nov. 2007, http://harvardbusinessonline.hbsp.harvard.edu/hbsp/hbr/articles/article.jsp?ml action=get-article&ml-issueid=BR071 1&articlelD=R0711D; Diana Farrell, Private Equity Isn'tFading Away, Bus. WK., Nov. 20, 2007, http://www.businessweek.com/globalbiz/content/nov2007/gb20071120276791.htm. For a vivid (and perhaps jarring) example of the contrast between public-company and private equity boards, see, for example, White & Zuckerman, supra note 11, at C1 ("AtMr. Conde's first meeting with his new bosses [at SunGard Data Systems], SunGard director DavidRoux, co-founder of one of the firms that bought the company, offered advice on how to train newclients. In three years running SunGard as a publicly traded company, Mr. Conde says he rarely heardsuch specific suggestions from directors.").

18. Pub. L. No. 107-204, 116 Stat. 745 (codified in scattered sections of 11, 15, 18, 28, and 29U.S.C.). A point perhaps best put by David Bonderman of Texas Pacific Group: "Our best friendshave been Messrs. Sarbanes, Oxley, and Spitzer .... When it becomes uncomfortable for executives inpublic markets, we are a source of alternative capital, but a more expensive one." Deborah Orr, TheNew Titans, FORBEs, April 19, 2004, at 68. For similar comments made by former HCA chief executiveofficerJack 0. Bovenderjr. in an interview, see Theo Francis, HCA Chief Enjoying the Private Life, WALLST. J., Jan. 7, 2008, at B1.

19. See, e.g., Thomas 0. Wells & Samantha A. Carter, Profits Interest-Converting Compensation toCapital Gains and Other Planning Ideas, FLA. B.J., Dec. 2007, at 52, 52.

20. Tennelle Tracy, It's the Fees, Not the Profits, Wall St. J., Sept. 13, 2007, at Cl.21. See, e.g., Barber & Goold, supra note 17, at 55 (reporting management fees of "1.5% to 2% of

assets under management" for "large" buyout funds).22. See Andrew Metrick & Ayako Yasuda, The Economics of Private Equity Funds 5, 22 (University of

Pennsylvania, Wharton School, Department of Finance, Working Paper, Sept. 9, 2007).


in the debt markets significantly reduces the number of attractive investments, 23

because of the firms' incentives to maximize assets under management and therebymaximize management fees.

The massive pools of cash controlled by PIE firms have yielded frenzied compe-tition among these firms to get deals. Stephen Schwartzman, head of private equitypowerhouse Blackstone Group, reports that prices in P/E deals have risen to "nose-bleed territory" 24 Price-earnings trends confirm this impression. 2

1 With too muchmoney chasing too few deals, full-blown auctions have become ubiquitous, lead-ing P/E firms increasingly to search for ways to gain exclusivity-that is, one-on-

one negotiations with a prospective target .2 Not only does exclusivity potentiallyallow a P/E buyer to pay a lower price, but it also reduces the possibility that theP/E buyer will walk away empty-handed (i.e., not closing the deal) after makingsignificant investments of both time and money in the due diligence process.

Exclusivity, though, has been elusive. In private-company deals, P/E firms can-vass the marketplace relentlessly to identify potential sellers. One Boston-based

P/E firm, for example, employs dozens of recent college graduates to cold-callentrepreneurs in the hopes of finding a potential seller before other buyers ap-pear. 7 Even in the needle-in-a-haystack situation where this firm identifies sucha seller, a common next step is for the seller to retain an investment banker whowill canvass the marketplace-thus eliminating or at least severely diminishing

any first-mover advantage that the P/E firm might have had.In public-company deals, exclusivity is even more difficult to achieve because

the target board has a fiduciary duty to maximize the price that it receives in asale of the company The traditional way in which boards fulfill this "Revlon duty"(named after the 1986 Delaware Supreme Court opinion in which the courtmost squarely articulated the duty) 28 is by canvassing the market, then signing amerger agreement with the highest bidder.2 9 In public-company deals, then, theconventional wisdom historically has been that the board of directors has a legalobligation not to grant exclusivity in a P/E buyout.3" Once a full canvass of themarket has taken place and the deal has been announced, the merger agreement

23. See infra note 79 and accompanying text.24. Peter Smith, Buy-out Industry Warned on Prices, FTcom, Feb. 21, 2006, http://www.ft.com/cms/

s/O/Of577274-a2fc- 1 ida-ba72-0000779e2340.html25. See Thornton, supra note 16, at 58 (reporting average price-to-eamings multiples of nine-to-

one in in the fourth quarter of 2005, compared to six-to-one in 2000).26. See, e.g., Robert L. Friedman, The Blackstone Group, Louis J. D'Ambrosio, Avaya Inc., John C.

Finley, Simpson Thacher & Bartlett LLP, Eileen T Nugent, Skadden, Arps, Slate, Meagher & Flom LLP,Private Equity Buy-outs, Presentation to Harvard Law School Mergers, Acquisitions, and Split-Ups(Nov. 13, 2007) (comments of Robert L. Friedman, chief legal officer, The Blackstone Group) (from56:40 of video transcript available at http://blogs.law.harvard.edu/corpgov/category/mergers-and-acquisitions/) ("We rarely have a one-off opportunity without competition. So we are competing withKKR and TPG.").

27. This fact is within the personal knowledge of the author.28. See Revlon v. MacAndrews & Forbes Holdings, Inc., 506 A.2d 173 (Del. 1986).29. For more detail on the traditional deal process, see generally Guhan Subramanian, Bargaining

in the Shadow of Takeover Defenses, 113 YALE LJ. 621 (2003).30. See, e.g., Joseph L. Morrel, Note, Go Shops: A Ticket to Ride Past a Target Board's Revlon Duties?,

86 TEx. L. REv. 1123, 1124 (2008).


typically includes a "no-shop" clause, in which the target agrees not to activelysolicit other buyers.31 The target board will nevertheless insist on a "fiduciaryout," which allows the board to negotiate with a third party who might be ableto make a superior offer.3 1 If a higher bidder emerges during this post-announce-ment period, the target will be required to pay the initial bidder a breakup fee,typically amounting to 2-4% of the deal value.33

Enter the go-shop clause, which turns the traditional negotiation process forpublic-company buyouts on its head. A P/E buyer will approach a target with anindication of interest. Rather than canvassing the marketplace at this point, thetarget will grant the buyer an exclusive negotiating period in exchange for signedconfidentiality and standstill agreements. 34 If the parties reach agreement duringthis exclusive negotiating period, the deal is announced. The go-shop provision inthe merger agreement then provides the seller with the right to solicit other buyersfor thirty to sixty days, in sharp contrast to the traditional "no-shop" deal. 35 Themerger agreement will usually include a "bifurcated" breakup fee: a lower amount,typically 1-2% of the equity value of the deal, if a higher offer emerges during thego-shop period; and a higher amount, typically 2-4% of the deal value, if a higheroffer emerges after the go-shop period expires but before the initial deal closes.36

If a higher offer emerges before the go-shop period expires, the target board willdesignate the new bidder as an "Excluded Party" ("excluded" because it is exemptfrom paying the higher breakup fee if it reaches a deal to buy the target). 3

1 Underthe "match right" that is typically included in the initial merger agreement, the targetboard must then negotiate for either three or five days "in good faith" with the ini-tial bidder to see if the initial bidder can match the terms offered by the ExcludedParty 35 The seller may go back and forth several more times, but each new bid bythe Excluded Party triggers a new three or five day match right for the initial bid-der.39 If the Excluded Party wins the auction in the end, the target board will pay

31. Seeid. at 1130.32. See id.33. See John C. Coates IV & Guhan Subramanian, A Buy-Side Model of M&A Lockups: Theory &

Evidence, 53 STAN. L. REV. 307, 331-35 (2000) (providing empirical evidence on the magnitude ofbreakup fees in a sample of public-company acquisitions announced between 1988 and 1998).

34. See Morrel, supra note 30, at 1131-32.35. See id.36. See, e.g., Coates & Subramanian, supra note 33, at 331-35; Morrel, supra note 30, at 1132-33.

It is interesting to note a parallel between the go-shop process and section 363 of the BankruptcyCode. Section 363 requires a debtor to obtain the highest price available in the sale of assets, but indoing so, the debtor is permitted to sign-up a deal with a so-called stalking horse bidder, who receivesmodest deal protection and other preferential treatment including a breakup fee, before the debtorconducts the full-scale auction. See 11 VS.C. § 363 (2000 & Supp. V 2005). For an example of asuccessful auction under section 363, in which a stalking horse bid was then subject to intense com-petition from eight bidders in a full-blown auction, see GuHAN SUBRAMANIAN & ELIOT SHERMAN, CABLE &WIRELEss AMERICA (Harvard Business School Case Study 9-908-004) (July 5, 2007) (describing sale ofCable & Wireless America under section 363).

37. See Memorandum from Mark A. Morton & Roxanne L. Houtman, Potter, Anderson & CorroonLLP, Go-Shops: Market Check Magic or Mirage? 5-6 (May 2007).

38. See Morrel, supra note 30, at 1133.39. See generally Morton & Houtman, supra note 37.


the reduced breakup fee to the initial bidderY'0 The length of the go-shop period,the size of the two breakup fees, and whether the initial bidder gets a match rightfor any subsequent offer are all heavily negotiated points between the P/E buyerand the target during the exclusive negotiating period.

At the highest level, in both the no-shop and go-shop processes, the target can-vasses the market to see if there is a higher-value bidder. The critical differencelies in when this market check takes place: in the traditional no-shop route, themarket check takes place before signing; while in the go-shop process, the marketcheck takes place after the deal is signed and announced. The timing difference hasthree important implications for the deal process. First, a pre-signing market checkplaces all bidders on a level playing field with respect to the economics of thetransaction; in contrast, a post-signing market check gives the announced bidder aslight leg up because of the breakup fee. Even though the breakup fee is a modestfraction of deal value, the combination of the fee and the first bidder's match rightmay deter a prospective bidder.4' Second, a pre-signing market check gives all bid-

ders the same timeline for making bids; in contrast, a post-signing market checkrequires potential bidders to demonstrate a reasonable likelihood of making a su-perior proposal before the expiration of the go-shop period, which in some casescan be a tight timeframe. Third, and perhaps most importantly, in a pre-signingmarket check the board and management have a legal obligation to maintain alevel playing field among all bidders; in contrast, in a post-deal market check pro-cess the board has explicitly given a preferred status to the announced bidder. Thisinside track is particularly valuable when management is part of the buyout groupand/or has announced its intention to stay on post-buyout to run the companyBidders who may have been willing to make a bid during a pre-bid market checkmay be deterred when the initial bidder is perceived to have management "locked

up. ' 42 For any or all of these reasons, then, the timing of the market check (pre-deal versus post-deal) can have implications for whether the highest-value buyer isidentified and, related, whether the target achieves the highest feasible price.

II. EVOLUTION OF DEAL PROCESS CASE LAW

In 1985, the Delaware Supreme Court held in Revlon v. MacAndrews & Forbes

Holdings, Inc. that when the sale or "break-up" of a company becomes "inevitable,"the directors' duties shift from "defenders of the corporate bastion to auctioneerscharged with getting the best price for the stockholders in the sale of the com-pany"4 3 Virtually all P/E buyouts are subject to examination under the Revlon stan-dard because target shareholders are typically getting cashed out of the company

40. See id. at 5.41. See generally Guhan Subramanian, The Drivers of Market Efficiency in Revlon Transactions, 28

J. CORP. L. 691 (2000).42. See, e.g., M & A Law Prof Blog, http://lawprofessors.typepad.com/mergers/ (Apr. 26, 2007)

("Harman's Shopping Spree") ("Illn neither case [Columbia HCA and Freescale] did competitorsemerge, likely due to the involvement of management in the initial deal.").

43. 506 A.2d 173, 182 (Del. 1986).


In addition, management buyouts are subject to entire fairness review, requiringthe buyer to show both fair price and fair process. 4

Despite the auctioneering language in Revlon, courts in subsequent cases clari-fied that the basic duty is to maximize immediate shareholder value but that thereis no "standard formula" 45 that a board had to follow. In 1989, the Delaware Su-preme Court in Barkan v. Amsted Industries, Inc.46 provided further clarification onthe substantive requirements imposed by Revlon: first, a level playing field amongbidders;47 and second, a "market check" to ensure that the board was getting thebest possible deal for its shareholders, 48 with a narrow exemption for situations inwhich the board already had "a body of reliable evidence with which to evaluatethe fairness of the transaction."49

Until very recently, the trajectory of the case law has been toward greater flex-ibility in the process that a board may use to full its Revlon duties (or, put differ-ently, a watering down of the requirements imposed by Revlon).50 The evolutionbegan with In re Fort Howard Corp. Shareholders Litigation,1 two years after Revlon,in which the sell-side board did not engage in a pre-signing market check, but thepress release announcing the deal invited alternative proposals and the breakupfee was a relatively modest 1.9% of the equity value of the deal. The DelawareCourt of Chancery rejected the plaintiffs' claim that this deal process violatedthe selling board's Revlon duties, thereby endorsing the possibility of an implicit"market check." 2

Practitioners since then have tested the boundaries of what constitutes a validimplicit market check.53 In In re Pennaco Energy, Inc. Shareholders Litigation54 andIn re MONY Group, Inc. Shareholder Litigation,51 the target board conducted no pre-signing canvass of the marketplace, following the Fort Howard blueprint, but the

44. See Weinberger v. UOP, Inc., 457 A.2d 701, 710-11 (Del. 1983); Kahn v. Lynch Commc'nsSys., Inc., 638 A.2d 1110, 1115 (Del. 1994).

45. Mills Acquisition Co. v. Macmillan, Inc., 559 A.2d 1261, 1286 (Del. 1989).46. 567 A.2d 1279 (Del. 1989).47. See id. at 1286-87 ("[Wlhen several suitors are actively bidding for control of a corporation,

the directors may not use defensive tactics that destroy the auction process .... When multiple biddersare competing for control ... fairness forbids directors from using defensive mechanisms to thwart anauction or to favor one bidder over another.").

48. Id. at 1287 ("When the board is considering a single offer and has no reliable grounds uponwhich to judge its adequacy... fairness demands a canvas of the market to determine if higher bidsmay be elicted.").

49. Id..50. See WILIAM T. ALLEN, REINIER KRAAKMAN & GuHAN SUBRAMANIAN, COMMENTARIES AND CASES ON THE

LAw OF BuSINESS ORGANIZATION 558-59 (2d ed. 2007).51. Civ. A. No. 9991, 1988 WL 83147, at *7 (Del. Ch. Aug. 8, 1988).52. Id. at *11-14.53. Cf. Coates & Subramanian, supra note 33, at 334 n.90 ("The percentage that is okay has slowly

risen. A year ago, two years ago, people were talking about two percent, two-and-a-half percent. Now,you hear them talking about three, three-and-a-half percent. Some are even saying four percent. Yousit there and ask, 'On what basis are you doing that? Where did you get that number?' There hasn'tbeen a specific challenge, so everybody pushes the envelope." (quoting Richard Beattie, managingpartner, Simpson, Thacher & Bartlett)).

54. 787 A.2d 691, 702 (Del. Ch. 2001).55. 852 A.2d 9, 18 (Del. Ch. 2004).


breakup fees were 3.0% and 3.3% of equity value, respectively, and there were noinvitations in the press releases for alternative bids. In Pennaco, the merger agree-ment also included a "match right," a relatively new deal term at the time, whichgave the buyer a three-day exclusive negotiation period with Pennaco if a higherbidder emerged. 56 The Delaware Court of Chancery nevertheless approved bothdeals under the Revlon standard. 5

1

Continuing in this line of cases, In re Toys "R" Us, Inc. Shareholder Litigation51 in-

volved a fifteen-month strategic review process at the well-known toy retailer. TheToys "R" Us board initially focused on the sale of the toy business, then shifted toa sale of the entire company (consisting of the lagging toy business and the hugelysuccessful Babies "R" Us business).59 The board solicited bids from the same fourbuyers that had been identified as interested in buying the toy business, eventhough the set of potential buyers for the entire company was likely to be differ-ent.60 The deal contained a 3.75% breakup fee (amounting to $247.5 million) anda match right against any potential third-party bidder.6' Despite the convolutedsearch process and the significant deal protection, the Delaware Court of Chan-cery held that the board had satisfied its Revlon duties. 62

The latest development in the articulation of Revlon duties reverses this trajec-tory, albeit in a potentially narrow way In re Netsmart Technologies, Inc. ShareholdersLitigation63 involved the sale of Netsmart, a micro-cap company in the behavioralhealthcare information technology marketplace. The board sold the company toa P/E group after an auction among seven PIE investors but no strategic buyers.64

Plaintiffs challenged the deal under Revlon, claiming that the board should havesolicited strategic buyers as well; defendants responded that the relatively smallbreakup fee facilitated an implicit market check for potential strategic buyers.6 5

Surprisingly, the Delaware Court of Chancery granted a preliminary injunctionagainst the deal, acknowledging that the implicit market check was a well-acceptedtechnique but drawing a distinction between a large-cap deal and a micro-capdeal such as Netsmart: "The mere fact that a technique was used in different mar-ket circumstances by another board and approved by the court does not meanthat it is reasonable in other circumstances that involve very different marketdynamics.

66

56. See AGREEMENT AND PLAN OF MERGER BETWEEN MARATHON OIL AND PENNACO ENERGY § 5.02(a)(Dec. 22, 2000).

57. Pennaco Energy, 787 A.2d at 707; MONY Group, 852 A.2d at 24.58. 877 A.2d 975, 983 (Del. Ch. 2005).59. Id. at 990.60. See id.61. Id. at 997.62. See id. at 1021. 1 served as an expert witness for the plaintiffs in this litigation, providing em-

pirical evidence that the breakup fee was large and theoretical reasons why a match right should deterpotential bidders.

63. 924 A.2d 171, 175 (Del. Ch. 2007).64. id.65. See id. at 175-76.66. Id. at 197.


Against this general backdrop, the Delaware Court of Chancery in two recentdecisions examined go-shop provisions in P/E deals, though the court in neitherof these decisions articulated general principles that might provide general guid-ance for transactional lawyers and bankers who structure go-shops. In In re ToppsCo. Shareholders Litigation,7 Vice Chancellor Strine held that a forty-day go-shopperiod, despite the existence of a match right and a 4.3% breakup fee after the go-shop period expired, did not violate the target board's Revlon duties. ("For 40 days,the Topps board could shop like Paris Hilton." ) In contrast, in In re Lear Corp.Shareholder Litigation,69 issued just one day later, Vice Chancellor Strine lookedfar more skeptically on the go-shop provision because the Lear go-shop requiredthe board to "get the whole shebang done [i.e., discovery of a superior proposal,expiration of the ten-day match right, termination of the initial agreement, and asigned-up agreement with the new bidder] within the forty-five-day window.'7 °

In both Topps and Lear, the court refused to grant a preliminary injunction againstthe deal on the Revlon claims but granted the plaintiffs' request for a preliminaryinjunction until certain disclosure problems were cured.7

In short, the current state of play regarding go-shop provisions and Revlon du-ties seems to be this: an evolution toward greater deference to the target board'sdiscretion in managing the deal process, with a slight "pinprick" in this reason-ableness assessment in Netsmart; and, applying this overall approach to the spe-cific context of go-shops, a mixed bag in Topps and Lear, in which the go-shop wasunproblematic in Topps but apparently closer to the line in Lear.

III. PRIOR LITERATURE

Mirroring the case law, the prior literature is divided on go-shop clauses. On onehand, some commentators argue that the proliferation of go-shop clauses is amanifestation of seller bargaining power in a world of frenzied buy-side competi-tion.72 Under this view, the go-shop clause is seller-friendly because it replaces the

67. 926 A.2d 58, 86-87 (Del. Ch. 2007).68. Id. at 86.69. 926 A.2d 94 (Del. Ch. 2007).70. Id. at 119. Vice Chancellor Strine continued in his usual entertaining way: "It is conceivable, I

suppose, that this could occur if a ravenous bidder had simply been waiting for an explicit invitationto swallow up Lear. But if that sort of Kobayashi-like buyer existed, it might have reasonably been ex-pected to emerge before the Merger Agreement with Icahn was signed based on Lear's lack of a rightsplan and the publicity given to Icahn's prior investments in the company" Id. at 120.

71. See Topps, 926 A.2d at 87, 93; Lear, 926 A.2d at 122-23.72. See, e.g., Bank of America Business Capital, CapitalEyes, Why More Sellers Are Using a "Go-

Shop" Clause (May/June 2007), http://corp.bankofamerica.com/public/public.portal?-pd-pagejlabel=products/abf/capeyes/index&dcCapEyes=indCE&id=353 ("M&A lawyers say Igo-shop clausesl arelikely to remain popular as sellers seek to include this provision in more deals."); Steve Rosenbush,Private Equity Slugfest, BusINEssWEEK.COM, Feb. 13, 2007, http://www.businessweek.com/bwdaily/dn-flash/content/feb2007/db20070212-956645.htm ("Target companies are figuring out ways to elicitmore competition. One technique is the so-called 'go-shop' provision."); Mark W Peters, Mark A.Metz, Douglas S. Parker, Priya M. Doornbos, Commentary, The Increasingly Popular 'Go-Shop' Provi-sion-Why Now? (May 2007), http://www.dykema.com/publications/docs/mandALitiRep5-31-07.pdf("The existing seller's market environment may help account for the recent popularity of the go-shop


usual "no-shop" with a "go-shop," which can only work to the seller's advantage.7 3

This view, if correct, would suggest higher returns for target shareholders in dealswith go-shop clauses, either because higher prices and go-shop clause incidenceare both positively correlated with seller bargaining power, or because the go-shop increases the likelihood of a subsequent (higher) bid, or both.

On the other hand, the more common view is that go-shop clauses work primar-ily to the buyer's advantage because the addition of a go-shop clause gives the targetboard an excuse to curtail (or eliminate completely) pre-deal shopping and thesubsequent go-shop period is illusory.7 4 The go-shop period can be illusory if thesell-side investment bankers do not do an adequate job of canvassing the market-place during the go-shop period.75 Alternatively, or in addition, the go-shop periodcan be illusory if potential bidders are deterred by the breakup fee, the relativelyshort time period to make a bid, or the existence of an inside bidder who will typi-cally have the right to match any higher offer. Under one version of this view, theP/E buyer promises management lucrative post-buyout employment agreementsand management puts a go-shop clause into the merger agreement as "windowdressing" to provide cover against judicial scrutiny76 This view suggests lower re-turns for target shareholders in deals with go-shop clauses, either because the lu-crative management contracts come at the expense of shareholders, or because thebidder who gains exclusivity is unlikely to be the highest bidder, or both.

provision."); MergerMetrics.com, A Review of Merger Agreement Provisions in Going Private Transac-tions (March 23, 2007), https://www.mergermetrics.com/pub/rs 20070326.html (arguing that prolif-eration of go-shop clauses indicates "a strengthening of the target's negotiating position"). See generallyDEBEVOISE & PLIMPTON, PRIVATE EQUITY REPORT (Fall 2006), http://www.debevoise.com/newseventspubs/publications/detail.aspxid=84bc208f-033a-482a-9ae3-9fc0e104be73 ("The business press perceivesgo-shop provisions as another example of how the competitive M&A market is changing acquisitionagreements in favor of targets.").

73. See supra note 72.74. See, e.g., Andrew Ross Sorkin, Looking for More Money, After Reaching a Deal, N.Y. TIMEs, Mar. 26,

2006, http://www.nytimes.com/2006/03/26/business/yourmoney/26dealbook.html ("The question,though, is why a company wouldn't hold an open auction to begin with?... [Mlore often than not, itwould seem that an open auction is the best way to go .... [Tihe biggest problem with a go-shop pro-vision is that by default, other potential bidders start at a huge disadvantage."); Christina M. Saut-ter, Shopping During Extended Store Hours: From No Shops to Go-Shops-The Development, Effectiveness,and Implications of Go-Shop Provisions in Change of Control Transactions, 73 BROOKLYN L. REV. 525 (2008);M & A Law Prof Blog, http://lawprofessors.typepad.com/mergers/ (Apr. 26, 2007) ("Harman's Shop-ping Spree") ("llinvestors have increasingly come to see "go-shop" provisions as cover for unduly largebreak-up fees and the significant advantage and head-start provided by management participation.");Posting of Tom Taulli to BloggingBuyouts, http://www.bloggingbuyouts.com/ (June 12, 2007, 10:23EST) ("ACS Buyout: A Second Bite at the Apple?") ("[lit's usually the case that a 'go shop' doesn'tamount to much anyway."). Cf. Morton & Houtman, supra note 37, at 1, 7 ("[Our practical experi-ence suggests that while go-shops may be beneficial in some circumstances, they may serve as merewindow dressing in other cases.... [Wihy should one assume that a go-shop will serve to canvass themarket., if no one ever makes a competing bid?"). See generally Memorandum from M. Lipton, T. N.Mervis & P K. Rowe, Wachtell, Lipton, Rosen & Katz, Private Equity and the Board of Directors 4(noting that "some skepticism has been expressed about the effectiveness of go-shops given a percep-tion that private equity buyers may be reluctant to compete against signed-up deals.") [hereinafter"Lipton, Mervis & Rowe"].

75. Andrew Ross Sorkin, Investment Banks' Brave, New, Conflicted World, N.Y. TiMEs DEALBOOK,Mar. 29, 2007, http://dealbook.blogs.nytimes.com/category/tulane-2007/.

76. See supra note 74.


Despite the large practitioner literature that has emerged over the past two yearson go-shops, there has been no systematic investigation of go-shop provisions be-yond basic descriptive statistics. 7 Fundamental empirical questions remain unan-swered: Are go-shop deals different from no-shop deals? What fraction of go-shopdeals yield a higher bidder? Is the presence of a go-shop clause correlated eitherpositively or negatively with target shareholder returns? In the next Part I seek toanswer these and other questions.

IV. EMPIRICAL FINDINGS

I use Thomson Financial's M&A database to identify all buyouts of U.S. publiccompanies larger than $50 million in value announced between January 1, 2006,and August 31, 2007, in which at least one private equity firm was part of the buy-out group. I then cross-check this sample against monthly reports from CorporateControl Alert's "The Lineup" column to ensure full coverage of buyouts during thesample period. I exclude acquisitions out of bankruptcy because of the mandatoryprocess rules in this context and the oversight required by the U.S. BankruptcyCourt. I also exclude deals with a controlling shareholder, using a control thresholdof 35%,78 because any shopping process would not be meaningful in these dealsunless the controller agreed to sell its shares into a higher-value competing bid(which is rare). The final sample includes 141 buyouts.

This timeframe has the desirable feature of capturing all P/E deals in a "chapter"of PIE history: from the very beginning of the regular usage of the go-shop tech-nology79 to what many perceive to be the meltdown of the P/E marketplace in thesecond half of 2007 from the ripple effects of the sub-prime mortgage crisis onthe credit markets overall.8 0 One implication of this feature is that the sample in-cludes the full set of comparable transactions. While there will of course be moreP/E deals and go-shop deals in the future, these future deals may not be directlycomparable to the deals under examination here because of the significantly dif-ferent market environment going forward.

For each deal I examine the seller's press release, the merger agreement, the proxystatement issued in conjunction with the merger, and other SEC filings (primarily

77. See, e.g., Morton & Houtman, supra note 37, at 1 ("[Tlhe authors are not aware of any empiricalanalysis of go-shops .. .. "). See also Lipton, Mervis & Rowe, supra note 74, at 4 ("ITlhere is no empiri-cal evidence supporting the view that go-shops are ineffective, and there is at least anecdotal evidence(Maytag, Catalina Marketing, EGL) that private equity firms are willing to top publicly announceddeals." (emphasis in original)). More systematic evidence on this question is presented in Table 3.

78. See Guhan Subramanian, Post-Siliconix Freeze-outs: Theory & Evidence, 36 J. LEG. STUo. 1, 8(2007) (providing theoretical justification and case law supporting 35% threshold for control stake).

79. According to data from Morton & Houtman, supra note 37, there were only seven go-shopdeals before the beginning of my sample period, with the first being Welsh, Carson, Anderson &Stowe's buyout of U.S. Oncology in March 2004. Because the private equity deal environment becameconsiderably more heated during 2006-07, 1 use the shorter timeframe to ensure comparability be-tween the go-shop and no-shop samples.

80. See, e.g., Friedman et al., supra note 26, at 12 (Morgan Stanley/JPMorgan slide titled "Global DealFlow Has Dwindled in Recent Months") (on file with The Business Lawyer). The study notes that Augustand September were the first months in 2007 that had fewer deals larger than $100 million than thecorresponding months in 2006.


14A and 14D-9 filings) to capture various features of the deal process, including,most importantly, the seller's post-deal solicitation rights (no-shop or go-shop)and the pre-signing and post-signing market canvasses that were conducted. Dailystock price data comes from Datastream.

A. Go-SHOP INCIDENCE AND STRUCTURING

Table 1 shows the number of deals that were announced in each month of thesample period, broken down by the deal process used. Table 1 reveals that go-shops have appeared in one-third of deals (48 out of 141), roughly consistent withpractitioner estimates,"' though this aggregate statistic masks significant growthin the use of the go-shop technology during the sample period. For example, go-shops appeared in 18% of deals (8 out of 45) during the first half of the sampleperiod (January to September 2006) and 42% of deals (40 out of 96) in thesecond half of the sample period (October 2006 to August 2007). In addition,Table 1 shows that, because go-shop deals tend to be larger than no-shop deals,go-shop deals account for almost half of private-equity deal volume (46%) duringthe sample period.

Merging my sample with recent practitioner data8 2 reveals that private equityfirms account for the vast majority of go-shop provisions-48 out of 57 dealsduring the sample period (84%), which is far greater than private equity's shareof M&A deals overall. If deal volume rather than number of deals is considered,Table 1 shows that private equity accounts for virtually all (96%) of go-shops.One possible explanation for PIE's disproportionate use of go-shops is that thesefirms might value pre-signing exclusivity more than strategic buyers because ofthe sense among PIE buyers that they have no natural edge over other P/E firmsand therefore are playing a common value game. If PIE firms are generally usingthe same valuation models and have the same discount rates, the only source ofdifferentiation across firms would be access to management or expectations aboutfuture cash flows (the latter source of differentiation of course raising winner'scurse concerns). In contrast, strategic buyers typically have synergies with thetarget that provide at least the possibility for private value, thereby making thesefirms less concerned about pre-signing competitors. Another potential explana-tion is that strategic buyers are more concerned about post-signing competitionthan private equity buyers83-that is, publicly losing the deal is far worse than

81. See, e.g., Bank of America Business Capital, supra note 72 (quoting Robert Townsend, a partnerat Morrison & Forester, who estimates a 25% incidence of go-shops in P/E deals); A Review of MergerAgreement Provisions in Going Private Transactions MergerMetrics.com, (Mar. 23, 2007), https://wwwmergermetrics.com/pub/rs_20070326.html (reporting 32% go-shop incidence in private equitydeals announced betweenjanuary and March 2007). See also WEIt GOTSHAt & MANGES, SPONSOR-BACKEDGOING PRIVATE TRANSACTIONS REPORT 10 (2007) [hereinafter "WEIL GOTSHAL, GOING PRIVATE TRANSACTIONS"](reporting 24% go-shop incidence among eighty-five P/E deals announced between October 1, 2005,and December 31, 2006).

82. See Morton & Houtman, supra note 37.83. 1 thank Steve Munger, Co-Chairman of the Global Mergers & Acquisitions group at Morgan

Stanley, for this point.


Table 1Go-Shop Incidence

This table provides statistics on go-shop incidence by month. The sample includes all

buyouts of U.S. companies larger than $50 million in value, announced between January2006 and August 2007. Go-shop incidence is determined by examining the "Solicitation"section of the merger agreement. Statistics on total go-shop incidence come from Mark A.Morton & Roxanne L. Houtman, Potter, Anderson & Corroon LLP, Go-Shops: MarketCheck Magic or Mirage? (May 2007).

By # of Deals By Deal Volume ($MM)

Total TotalGo-Shops No-Shops Go-Shop Go-Shops No-Shops Go-Shopin Sample in Sample Deals in Sample in Sample Deals

2006January 1 3 1 $1,021 $13,383 $1,021February 0 5 0 $2,425March 0 5 1 $8,272 $3,181April 0 2 0 $516May 2 4 2 $3,793 $22,148 $3,793June 1 5 1 $453 $15,674 $453July 2 3 2 $22,877 $1,826 $22,877August 1 5 2 $1,707 $4,170 $2,834September 1 5 1 $17,704 $1,542 $17,704October 2 2 2 $18,055 $2,385 $18,055November 2 7 3 $22,674 $29,856 $22,881December 2 5 2 $11,774 $18,847 $11,744

2007January 3 6 3 $4,981 $4,670 $4,981February 6 5 6 $43,015 $4,703 $43,015March 4 7 6 $3,016 $19,076 $5,330April 4 7 6 $35,063 $32,729 $35,119May 4 7 4 $14,475 $58,014 $14,475June 5 5 6 $20,633 $4,064 $21,261July 8 4 9 $11,651 $26,693 $11,976August 0 1 $59

TOTAL 48 93 57 $232,890 $271,051 $240,700

privately losing the deal for a strategic buyer, potentially due to buy-side agencycosts. 4 Some combination of these explanations-in which P/E firms value pre-signing exclusivity and strategic buyers value post-signing exclusivity-might ex-plain the vastly disproportionate use of go-shops by P/E buyers.

Digging deeper into the go-shop subsample, Figure 1 documents the number ofbidders contacted before signing.

84. See Coates & Subramanian, supra note 33, at 307.


Figure 1

Number of Buyers Solicited Pre-Signing in Go-Shop Deals (n = 48)

>7 Bidders

5-7 Bidders8%

1 Bidder60%

2-4 Bidders17%

The data in Figure 1 reveals an important difference between go-shops: 60%(29 out of 48) are "pure" go-shops, where the target negotiates with a single bid-der during the pre-signing period. The remaining 40% of go-shops (19 out of 48)are what I term "add-on" go-shops, where the go-shop period was in addition toat least some degree of pre-signing shopping.8" Because pure go-shops seem con-ceptually different from add-on go-shops, I examine the two kinds of transactionsseparately in most of the analyses that follow.

Table 2 provides summary statistics on deal characteristics according to thedeal process that was used: traditional no-shop, add-on go-shop, or pure go-shop.Table 2 shows that go-shop deals are larger on average than no-shop deals, andthis difference is significant at 90% confidence. This finding is loosely consistentwith the theory recently articulated by the Delaware Court of Chancery that ex-plicit shopping is less important for larger deals because the market is more likelyto be aware that such companies are for sale.86 Put differently, practitioners canafford to be more flexible in shopping processes in large deals because potential

85. See WElL GOTSHAL, GOING PRIVATE TRA sAcrIONS, supra note 81, at 10 ("Surprisingly, one-halfof the transactions with a go-shop had some form of pre-signing market check."). But see Morton &Houtman, supra note 37, at 1 n.1 ("Nearly all of the thirty (30) transactions analyzed in the surveyinvolved targets that did not engage in a market canvass before entering into a merger agreement witha private equity buyer.").

86. See In re NetSmart Techs., Inc. S'holders Litig., 924 A.2d 171, 197 (Del. Ch. 2007) (suggestingthat an implicit market check may be viable in large-cap deals but not in small-cap deals).


Table 2Summary Statistics

This table provide summary statistics by deal process. The sample includes all buyouts ofU.S. companies larger than $50 million in value, announced between January 2006 andAugust 2007. "Pure" go-shop deals are defined as deals in which the target negotiated exclu-sively with a single bidder, then conducted a market canvass during a post-announcement"go-shop" period. "Add-on" go-shop deals are defined as deals in which the target con-ducted a market canvass pre-announcement but also engaged in a post-announcementmarket canvass pursuant to a go-shop clause. "No-shop" deals are defined as all other dealsand involve the traditional pre-announcement market canvass followed by the inability ofthe target to shop further after the deal is announced.

Add-on PureNo-Shop Go-Shop Go-Shop All Deals

Mean (Median) (n = 93) (n = 19) (n = 29) (n = 141)

Deal characteristics:

Deal Size ($MM)

Delaware incorporationBuyout group includes

managementInitiated by buy-side?

Pre-signing solicitationprocess:# of potential buyers

contacted# signing confidentiality

agreements# making bids

Post-signing solicitationprocess:# of potential buyers

contacted# signing confidentiality

agreements# making bids

Go-shop structuring:Go-shop initiated by seller?Bifurcated termination fee?Breakup during go-shop

period (% of equity value)Breakup fee after go-shop

period (% of equity value)Right to match third-party

bidder?Length of go-shop period (days)

$2,914.1($910.3)49.5%

$5,835.0($3,608.6)

73.7%

12.9% 15.8%

53.3% 68.4%

$4,207.8($1,479.9)

57.7%

24.1%93.1%

$3,574.1($1,193.2)

54.3%

15.6%

63.8%

31.6 (15) 15.9 (5.5) 1.0 (1) 22.4 (7)

16.1 (8.5) 7.8(4)3.9 (3) 2.7 (2)

1.0(1) 11.7(5)1.0(1) 3.1 (2)

0.0 (0) 33.0 (27) 39.6 (40) 12.3 (0)

0.07 (0) 1.5 (0) 3.2 (2) 0.9 (0)0.08 (0) 0.05 (0) 0.17 (0) 0.1 (0)

72.2%

84.2%1.65%

(1.54%)2.82%

(2.91%)

55.6%

88.9%1.45%

(1.41%)

2.62%(3.02%)

62.2%

87.0%1.53%

(1.41%)2.70%

(3.01%)

78.9% 59.3% 67.4%33.7 (30) 41.2 (45) 38.4 (40)


buyers (guided by highly motivated investment bankers) are likely to be informedthat the company is in play

Table 2 also shows that approximately half of all targets in the no-shop sampleare incorporated in Delaware, in proportion to the overall fraction of U.S. publiccompanies that are incorporated in that state."7 Though the difference is not statisti-cally significant, it is interesting to note that Delaware targets are overrepresentedin the go-shop sample. This finding might be due simply to the fact that Delawarecompanies are larger than non-Delaware companies, on average,88 so the differentincorporation mix simply reflects the size differential noted above. More interest-ingly, the overrepresentation of Delaware targets in the go-shop sample might reflecta greater willingness among practitioners to innovate with deal process in that state,given the sophistication of the courts that would be assessing such innovation.

On management involvement, MBOs are overrepresented in the go-shop sam-ple, though this difference is not statistically significant. Table 2 also shows impor-tant differences in who initiates the transaction across the various deal processes.No-shop deals are initiated on the buy-side (though not necessarily by the even-tual buyer) in 53% of deals and by the seller in the remaining 47% of deals.89 Thefraction initiated on the buy-side increases to 68% in the add-on go-shop dealsand to 93% in the pure go-shop deals. These findings are consistent with thepractitioner playbook on pure go-shops, in which a buyer approaches a potentialseller and insists on exclusivity in exchange for a go-shop period after the deal isannounced. I return to the implications of this finding in the target shareholderreturns analysis in Part IV.C below.

Moving to the pre-signing solicitation process, Table 2 shows large but perhapsunsurprising differences across no-shops and go-shops. Examining first the tra-ditional no-shop process, Table 2 shows that 31.6 buyers are contacted on aver-age,9" of which 16.1 signed confidentiality agreements and 3.9 made bids. Amongthe 93 deals in the no-shop subsample, only one deal (Blackstone's purchase ofHilton in July 2007) involved exclusive one-on-one negotiations during the pre-signing process. However, this deal does not fit the mold of a traditional "implicitmarket check" process because of the large breakup fee ($560 million, or 2.8%of the deal value) in the deal, the absence of any post-signing solicitation rightbeyond the standard "fiduciary out," and the lack of any invitation in the pressrelease for competing proposals. Hilton's proxy statement indicates that Hiltondid not engage in a pre-signing market check because Blackstone threatened towalk away if Hilton canvassed the market, and the deterioration of the credit

87. See Guhan Subramanian, The Influence of Antitakeover Statutes on Incorporation Choice: Evidenceon the "Race" Debate and Antitakeover Overreaching, 150 U. PA. L. REv. 1795, 1804 (2002).

88. See Guhan Subramanian, The Disappearing Delaware Effect, 20J.L. ECON. & ORG. 32, 37 (2004)-89. For reasons that are described in more detail below, buyouts initiated by management are clas-

sified as being initiated on the buy-side. See infra Part IVC.90. This finding is roughly consistent with the only other study I am aware of that examines the

pre-signing solicitation process. See Audra L. Boone & J. Harold Mulherin, How Are Firms Sold?, 62J. FIN. 847 (2007) (examining a sample of 202 acquisitions for cash between 1998 and 2003 andreporting twenty-one buyers contacted, on average).


markets during the negotiation made it unlikely that other bidders would havecome forward if such a canvass had been attempted.9' Therefore, there are no ex-amples of an implicit market check in the no-shop sample; rather, the overall pictureis one of robust competition among buyers during the pre-signing process. This(non)finding is consistent with practitioner impressions that, despite the judicialvalidation by the Delaware courts, an implicit market check is more vulnerable tochallenge than the traditional canvassing of the marketplace.92

Among add-on go-shops, Table 2 shows that 15.9 potential buyers were con-tacted on average, of whom 7.8 signed confidentiality agreements and 2.7 madebids. This picture of the add-on deal process is significantly different from theno-shop deal process, at each step along the way to a signed deal. It suggests thatthe add-on go-shop process may be something of a "halfway house" between thestandard no-shop process and the pure go-shop process, in which sellers trade offsome amount of pre-signing shopping in exchange for a signed deal; or, alterna-tively, buyers are able to short-circuit some amount of pre-signing shopping withthe offer of a go-shop. Finally, the pure go-shop column is definitional: one buyeris contacted (or contacts the seller); this buyer signs a confidentiality agreement,conducts due diligence, and the deal is announced.

Perhaps more interesting is the post-signing solicitation process. Table 2 showsthat in the no-shop process, the seller does not contact any further bidders (as isrequired by the merger agreement), but an unsolicited bidder appears approxi-mately 8% of the time. This finding is consistent with prior work finding a 3-7%jump rate in the typical market canvass/no-shop process.93 In contrast, in go-shopdeals the seller's bankers contact a very large number of potential buyers: 33.0on average in the add-on go-shop case and 39.6 in the pure go-shop case. Whenpre-signing and post-signing canvassing are considered together, go-shop dealsactually solicit more potential buyers than the no-shop deals: 48.9 are solicitedin add-on go-shops (= 15.9 + 33.0) and 40.6 are solicited in the pure go-shops(1.0 + 39.6), compared to 31.6 in the no-shops. This finding cuts against per-ceptions that the go-shop process is illusory.9 4 However, a striking finding is thenumber of potential buyers who sign confidentiality agreements during the post-signing go-shop process: 1.5 in add-on go-shops and 3.2 in pure go-shops, farfewer than the 16.1 buyers who sign confidentiality agreements in no-shop deals.After examining this non-public information, the go-shop successfully generatesanother offer 5% of the time in an add-on go-shop and 17% of the time in a purego-shop, consistent with the intuition that the go-shop should yield a higher-value buyer more often when there has been no pre-signing shopping.

In short, the overall assessment of the go-shop solicitation process is mixed:more potential buyers contacted in go-shop deals than in no-shop deals but sub-stantially fewer bidders signing confidentiality agreements. One straightforward

91. See Hilton Hotels Corp., Form DEFMI4A, at 25-28 (Aug. 8, 2007).92. See supra notes 74-76 and accompanying text.93. See Coates & Subramanian, supra note 33, at 371.94. See supra notes 74-76 and accompanying text.


explanation for these results is that the "bird in hand" that exists in a go-shop dealcreates a price floor, and potential buyers who are below this floor will declineto go forward by signing a confidentiality agreement. The puzzle is why theseprospective bidders should drop out before gaining access to confidential infor-mation; that is, before they would know with any certainty whether their offerwould be higher or lower than the outstanding offer. Far more understandablewould be an equal number signing confidentiality agreements in go-shop dealsbut fewer actual bids in the go-shop sample compared to the no-shop sample. Infact, the announced deal should send a signal to the marketplace that the com-pany is worth at least the announced deal price, and therefore more (not fewer)bidders should be interested in signing confidentiality agreements. The concernof course is that the announced deal in a go-shop process does more than create aprice floor; it also deters potential higher-value bidders from even conducting duediligence because of the non-level playing field created by the breakup fee andmatch right, in addition to other deal protection features.

The final section in Table 2 presents further details on the structuring of the add-on and pure go-shops. The go-shop term (as distinct from the deal itself) is initi-ated by the seller nearly three-quarters of the time in add-on go-shops, comparedto pure go-shops which are initiated by the seller only half the time. Althoughthis difference is not statistically significant, it is directionally consistent with theintuition that add-on go-shops are more likely to be a manifestation of seller bar-gaining power than pure go-shops. Table 2 further shows that virtually all go-shopdeals (87%) have a bifurcated termination fee, with an initial fee that is 57% of thefee that goes in effect after the go-shop period expires. Interestingly, the breakupfees in the pure go-shop context are not meaningfully smaller than the fees in theadd-on go-shop sample. One would expect smaller fees in the pure go-shop casebecause of the greater reliance on the post-signing solicitation process. Neverthe-less, the final two rows in Table 2 show areas where the pure go-shop process doesseem to have more viability than the add-on go-shop process: a significantly lowerincidence of match rights (59% versus 79%) and longer go-shop windows (41 daysversus 34 days). These statistics suggest that target boards (and perhaps buyerstoo) understand the importance of preserving a meaningful solicitation process ina pure go-shop context and structure the go-shop accordingly

B. Go-SHOP OUTCOMES

Table 3 continues the inquiry within the go-shop sample by documenting theinstances of "successful" go-shops in the sample, where a successful go-shop isdefined as one in which another bidder emerged during the go-shop period.Table 3 shows that a higher bidder appeared in six deals, far more frequently thanprior commentators have suggested,9" but consistent with practitioner reports

95. See, e.g., Memorandum from Mark A. Morton & Roxanne L. Houtman, Potter, Anderson &Corroon LLP, Go-Shops: Market Check Magic or Mirage? 10-11 (May 11, 2007); Friedman et al.,supra note 26 (comments of Louis D'Ambrosio, chief executive officer of Avaya, Inc., which wastaken private by Texas Pacific Group during the sample period using a go-shop) (from 1:22:02 ofvideo transcript) ("There are very very few situations in which something has happened during the

0 0z z

~0

-u

0

00

0

0

-u.~ 0)

0) -~0) 0)

0. 0)0.

.z ~

.0

0(0

a~X... (0

0.

0

-z,cc 71

(-)

0

u

(0

.9-(0

0)

00

(0

0 0 0z z z

u

u

0

0

$- 0

0z

(0

0)

0. 0~

(0

0

n0.9-bq


that go-shop windows provide a meaningful look for third parties.16 For example,Robert Friendman, chief legal officer of the Blackstone Group, reports:

We looked at every go-shop target... every single one in the last two and a half years.We had plenty of time to counter-bid, and in each case... we found the signed deal tobe off-the-charts too high and walked away But if we wanted to counter-bid we hadplenty of time to do it.... Go-shops are totally meaningful. There were deals like TXUand Kinder Morgan that we didn't know about until they were announced, althoughwe had heard rumors about them. And then we had an opportunity to participate,and we looked very hard at both and said 'no thanks.'... Both the strategic universeand the private equity universe would be reticent to come in during a classic no-shopprocess [after a signed deal is announced]. We just wouldn't do it. But when you puta 'For Sale' sign on the door, and say come get me, then people drop everything andlook because they are being invited in. 97

Three of the deal jumpers were strategic buyers (Community Health Systems,Advanced Medical Optics, and Brands Holdings), two were other financial buyers(Hellman & Friedman Capital Partners and Veritas Capital), and one was unspeci-fied in the proxy statement. The most striking feature of Table 3 is the last column:none of the successful go-shops involved management buyouts, or, conversely, allof the successful go-shops involved targets that did not have management involve-ment in the initial buyout group.9 This finding cannot be explained by higherpremiums in go-shop MBOs: in unreported analyses I find no meaningful differ-ences in premiums received or returns to target shareholders between go-shopMBOs and go-shop non-MBOs.

Interestingly, the reluctance to jump an MBO does not seem to apply to thepre-signing phase. Because I track the deals in my sample from the signing ofthe merger agreement, the sample does not capture the phenomenon in whichmanagement "tees up" the company for acquisition by a third party Prior stud-ies, however, report robust competition during the pre-signing phase in public-company MBOs.99 As discussed in Part II, post-signing competition with amanagement group may be more difficult than pre-signing competition because of

go-shop .... So I love the instrument, it's a wonderful idea,.., but what I question is why haven't moredeals been topped during that period of time?"). One potential reason for the misperception amongpractitioners might be that all of the successful go-shops occurred in 2007. Because practitioners'perceptions of the marketplace are formed largely through anecdotal experience rather than throughsystematic survey, practitioner perception is likely to lag reality

96. See supra notes 72-73 and accompanying text.97. Friedman et al., supra note 26 (from 1:14:54 of video transcript).98. In contemporaneous work Russel Denton similarly notes the absence of deal jumpers in go-

shop MBO deals. See J. Russel Denton, Note, Stacked Deck: Go-Shops and Auction Theory, 60 STAN. L.REV. (forthcoming 2008).

99. See Sarah W Peck, The Influence of Professional Investors on the Failure of Management BuyoutAttempts, 40 J. FIN. ECON. 267, 273 (1996) (reporting a 23% jump rate for a sample of 111 public-company MBOs between 1984 and 1987); John C. Coates IV, An Empirical Assessment of MBO Bids:Techniques, Outcomes, and Delaware Corporate Law 15 (Working Paper, Oct. 18, 2005) (on file with TheBusiness Lawyer) (reporting a 40% jump rate for a sample of public-company MBOs between 1988and 1999).


the presence of a breakup fee, a match right, or the "ticking clock" for conductingdue diligence and making an offer.

C. TARGET SHAREHOLDER RETURNS

I now examine returns to target shareholders by the deal process used. Foreach deal, I calculate cumulative buy-and-hold abnormal returns ("CARs"),net of the S&P Composite Index,0 ° from thirty days prior to deal announce-ment through thirty days after announcement. In unreported analyses, I findno meaningful difference in pre-signing run-up between the go-shop and no-shop samples over longer event windows (beginning sixty days prior to dealannouncement),'' contrary to practitioner impressions that go-shop deals areless likely to be leaked and therefore less likely to generate run-up. One possibleexplanation is that the pre-deal run-up in target companies in general is notdue to the leakage of inside information but rather due to analyst assessment ofpublic information that the company is likely to be a takeover target. If correct,this conclusion would suggest that the perceived benefit of go-shop processes asa way of reducing pre-deal run-up is small at best.

On the back-end, the very end of the time frame under analysis (July/Aug.2007) is problematic because of rapidly tightening credit, which caused "tur-moil" in many transactions announced during that period.10 2 As shown in Table

1, go-shop deals have become more frequent over time, reaching approximately

50% of announced transactions by the end of the sample period. Therefore, the

credit crisis has had a disproportionate influence on go-shop deals compared to

no-shop deals. In order to minimize this effect and focus instead on the question

of value extracted by the target board of directors, I end the announcement win-

dow at thirty days after deal announcement. Although not all go-shop periodshave expired by that time, conversations with practitioners suggest that virtually

all of the shopping is completed within thirty days after the announcement of

the deal.If go-shop clauses are an effective tool for identifying the highest-value buyer

and extracting full value from it, then returns to target shareholders should be

higher (or at least not lower) in go-shop deals than in the traditional no-shop

route. If instead go-shop deals deter potentially higher-value bidders, then target

shareholder returns should be lower in the go-shop sample than in the no-shopsample. Figure 2 shows cumulative daily abnormal returns for the full sample ac-

cording to the deal process used.Figure 2 shows no meaningful difference in target shareholder returns between

no-shop deals and add-on go-shop deals but approximately 5% higher abnormal

100. Other indices, such as the Datastream Total Market Index, yield similar results.101. See, e.g., Freescale Semiconductor Inc., Form DEFM14A, at 23 (Oct. 19, 2006) (noting "in-

creased possibility of a leak (and potential detrimental effects of a leak) by seeking and holding discus-sions with multiple third parties").

102. See Friedman et al., supra note 26, at 11 (Morgan Stanley/JPMorgan slide titled "NotableTransactions in Turmoil") (on file with The Business Lawyer).


Figure 2

Cumulative Abnormal Returns for Target Shareholders by Deal Process

1.25-

-No Shop

1.2 - Add-on Go-Shop - - - - - - --- Pure Go-Shop --- s -

1.05

1

0.95

0.9 ............ ...

C.) ~ ? 04 0 4 - - -- 4 0 4 C

Days to Deal Announcement

returns in the pure go-shop sample. This difference in portfolio CARs betweenthe pure go-shops and all other deals is statistically significant at 90% confi-dence (t-statistic = 1.69). The evidence presented in Figure 2 cuts against theconventional wisdom that pure go-shops are particularly suspect (from a targetshareholder perspective) because they involve no pre-signing market check. 10 3 Tothe contrary the evidence presented in Figure 2 suggests that there may be some-thing about pure go-shops (but not add-on go-shops) that allows target boards toextract more from a potential acquirer, despite the absence of any competition inthe pre-signing process.

The results presented in Figure 2 are of course problematic because of potentialselection effects between the go-shop and no-shop samples. The summary statis-tics presented in Tables 1 and 2 identify three important sources of potential bias.First, Table 1 shows that go-shop deals occur, on average, later in the sample pe-riod than no-shop deals. If overall market conditions and market pricing changedduring the twenty-month window of analysis, then this timing difference couldskew the calculation of target shareholder returns between the go-shop and no-shop samples. Second, Table 2 shows that go-shop deals are larger on averagethan no-shop deals. This effect would serve to bias the relative returns in go-shopdeals downward, because deal premiums are generally lower in larger deals.

Third, Table 2 shows that go-shop deals are far more likely to be initiated onthe buy-side than on the sell-side. It is possible (perhaps even plausible) that dealsinitiated on the buy-side reflect undervalued companies in the marketplace, while

103. See supra notes 74-76 and accompanying text.


deals initiated on the sell-side reflect overvalued companies in the marketplace."'If correct, this pattern would suggest higher potential returns in deals initiated onthe buy-side because the baseline price (i.e., market price) is lower than intrinsicvalue. One mitigating factor is the negotiation process itself. Because all the deals inthe sample are negotiated acquisitions, in which the buyer signs a confidential-ity agreement in order to gain access to the target's books and records, the usualinformational asymmetry between seller and buyer is minimized (and perhapseliminated) through the due diligence process. In negotiation terms the seller willresist offers that do not reflect a full (non-public) value for the company, while thebuyer will resist offers that reflect a (public) overvaluation of the company Nev-ertheless, to the extent that the negotiation process does not fully overcome theinitial information asymmetry between seller and buyer, the hypothesized effectwould increase the perceived relative returns in go-shop deals because go-shopdeals are far more likely to be initiated on the buy-side.

In order to control for these potential biases, I construct a propensity score foreach deal based on deal timing (natural log of the number of days since January 1,2006), deal size (natural log of deal value), and deal initiator (= 1 if initiated onthe buy-side). Because MBOs would be prone to the undervaluation story ratherthan the overvaluation story (because management is, on net, on the buy-sideof the transaction), MBOs are classified as initiating on the buy-side. For eachpure go-shop deal in the sample (n = 29), 1 identify the no-shop deal that has theclosest propensity score, using the absolute value of the difference in scores (i.e.,nearest-neighbor matching). Because add-on go-shops do not have meaningfullydifferent target shareholder returns than the no-shop deals and also are quite dif-ferent in terms of deal process from the no-shop sample, these deals are excludedfrom the analysis. I then construct portfolios of the pure go-shop deals and thematched-pair no-shop deals, using the same methodology as in Figure 2. Figure 3shows the cumulative abnormal buy-and-hold returns for these two portfolios.

Consistent with the overall findings from Figure 2, Figure 3 shows that averagetarget shareholder returns in go-shop deals are approximately 5% higher thantarget returns in no-shop deals, though this difference is no longer statistically sig-nificant (t-statistic = 1.40). The propensity score matching technique used in theFigure 3 analysis attempts to control for known differences in deal size, deal tim-ing, and deal initiation between the go-shop and no-shop samples. As with virtu-ally any analysis of this kind, there may be other, as yet undiscovered, sourcesof bias between the go-shop and no-shop samples. Nevertheless, the results pre-sented here at least call into question the weight of practitioner commentary thatpure go-shops allow the buyer to steal the company from the public shareholders"on the cheap.' l

To the contrary, the evidence presented here suggests that there may be some-thing about pure go-shops (but not add-on go-shops) that allows the target board

104. 1 thank Professor Bob Mnookin for this point. Note that this hypothesis is consistent with thesemi-strong version of the efficient market hypothesis.

105. See supra notes 74-76 and accompanying text.


Figure 3

Cumulative Abnormal Returns for Target Shareholders by Deal Process(matched pair sample)

1.25

1.2 •- 1 No Shop ] .. _

/ -- Pure Go-Shop] .. -

1. 15

1.05

0.95

0.9 ..... . .. "T' "T "T 'W

Days to Deal Announcement

to extract more from a potential acquirer, despite the absence of any competitionin the pre-signing process. The question then becomes from where, precisely, doesthis additional value come. The simple explanation offered by some prior commen-tators, that go-shops simply reflect sell-side bargaining power,10 6 is inconsistentwith the distinction reported here between add-on go-shops and pure go-shops. Amore subtle explanation appears from a close review of the proxy statements in thepure go-shop deals. On the sell-side, the proxy statements frequently documentthe seller's interest in exclusivity as a means of minimizing disruption, as well asthe benefits of the "bird in hand" that arises from a go-shop process. 07 On the buy-side, the proxy statements frequently document the initial bidder's willingness topay more in order to maintain exclusivity in the deal.'018 These sources of value

106. See supra notes 72-73 and accompanying text.107. See, e.g., Nuveen Investments, Form DEFM14A, at 27 (Aug. 14, 2007) ("The Disinterested

Directors and Goldman Sachs further discussed whether the attractiveness of [the offer from MadisonDearborn] was sufficiently high that the benefit that might be gained by conducting a pre-signingmarket check would not outweigh the potentially adverse consequences to our operations that mightresult as a result of undertaking such a process. Among other things, the Disinterested Directors con-sidered the uncertainty that could result from a pre-signing market check, which could cause financialadvisors and consultants to suspend recommending our products and services to their clients, affectthe stability of our relationships with members of our investment teams and impair our ability to retainkey employees should any of these individuals become aware of a market check."); CDW Corp., FormDEFM14A, at 14 (July 13, 2007) ("Our board decided not to authorize contact with any strategicbuyer [or any other buyer] until a post-signing 'go-shop' period because of concerns of furnishingproprietary information about our company to competitors and prospective competitors before it wasclear that an attractive price for our company could be achieved.").

108. See, e.g., Aeroflex Inc., Form DEFM14A, at 24 (June 22, 2007) ("After discussion, the spe-cial committee asked TWP [its bankers] to communicate to the initial bidding group that... Aeroflex


creation are consistent with a presentation made by senior M&A practitioners atthe Harvard Law School in November 2007. These practitioners list "establishfloor" and "leverage against bidder based on threat of competition" as benefits ofthe go-shop process compared to the traditional no-shop route. °9 The evidencepresented in Figure 3 suggests that the economic value of these benefits mighttranslate into approximately 5% higher returns for target shareholders.

V DISCUSSION

At the highest level, the data indicates that not all go-shops are created equal,and courts should therefore continue their case-by-case assessment of go-shopsrather than issuing a categorical pronouncement on their validity as a matter ofDelaware corporate law Add-on go-shops should be presumptively legal, if thepre-deal market canvass resembles the market canvass in the traditional no-shopprocess, because the add-on shopping can only yield a higher price than what thetraditional process would have achieved. Pure go-shops are potentially more prob-lematic because of the absence of a pre-signing market check. The evidence pre-sented here indicates that a pure go-shop can be a valuable tool for extracting thehighest possible price in the sale of the company A close examination of the proxystatements in pure go-shop deals provides anecdotal evidence that might explainthis counterintuitive finding. In many instances, both seller and buyer will valueexclusivity in the negotiation process, thereby creating value that can be dividedbetween the parties."t 0 The potential "win-win" nature of the go-shop technology

was likely to engage in a pre-signing 'market check' of the proposed transaction price but might beprepared to forgo this if the proposed share per share price were to be revised to $16.50 or more.");Jameson Inns Inc., Form DEFM14A, at 18 (June 29, 2006) ("JMP Securities [investment bankers forthe seller] further indicated that there was a direct correlation between the purchase price the board ofdirectors would be willing to accept and our opportunity to solicit or receive a competing offer superiorto the terms of the merger agreement that may be agreed by the board and JER."); HUB International,Form DEFM14A, at 10 (May 4, 2007) ("[Rlepresentatives of Apax explained that they did not wantto participate in an auction process .... At the end of the meeting, representatives of Apax advised Mr.Hughes that, subject to further due diligence, they were prepared to present Hub a proposal within tenbusiness days. In addition, Apax requested that for the remainder of the week, Mr. Hughes not contactany other party regarding a potential transaction. Mr. Hughes agreed that he would not make any suchsolicitations with the view that this would improve Apax's proposal."); Nuveen Investments, Form DE-FM 14A, at 29 (Aug. 14, 2007) ("[The special committee] informed Madison Dearborn that the specialcommittee might be willing to consider an offer in the range of $62.00 to $65.00 per share subject toa concurrent market check, but if Madison Dearborn wished to negotiate a transaction on an exclusivebasis for a limited period of time, Madison Dearborn would have to increase its proposed purchaseprice to at least $65.00 per share. [Two days later], Madison Dearborn delivered a revised indication ofinterest to the special committee that included a proposed purchase price of $65.00 per share."); SpiritFinance, Form DEFM 14A, at 19 (June 1, 2007) ("IThe Company advised Macquarie that the [$14.161offer price would likely be viewed by its board of directors to be insufficient to grant Macquarie therequested exclusive negotiation period and other terms requested, and suggested that a price of $15.00per share would be more likely to be acceptable to the Company's board of directors. [Three days later]Macquarie submitted a revised draft MOU indicating.., a price of $14.50 per share.").

109. See Friedman et al., supra note 26, at 33.110. See ROBERT H. MNOOKIN, ScoT R. PEPPEr & ANDREW S. TULUMELLO, BEYOND WINNING: NEGOTIATING

TO CREATE VALUE IN DEALS AND DisputEs 16 (2000) (identifying "noncompetitive similarities" as a source,of value creation in negotiations).


suggests that it might be a "better mousetrap" than the traditional method for sell-ing companies that are in Revlon mode. If correct, we would expect to see furtherdissemination of the go-shop technology in the future. The trajectory of go-shopdissemination thus far is consistent with this prediction.

While the source of the value creation on the sell-side is intuitively plausible,the proposed buy-side benefits raise a puzzle: why would a PIE buyer pay morefor exclusivity when it can pay less, on average, by engaging in the traditional pre-signing auction process? One possibility is that PIE buyers overvalue exclusivity,perhaps not realizing that bidders in an open auction will rationally underinvestin due diligence and therefore shade their bids downward because of their uncer-tainty in the valuation of the target. Another possible explanation arises from thecompensation structure of P/E buyers. As described in Part II, the 2% managementfee creates incentives for P/E buyers to invest capital rather than return it to theirlimited partner investors. Table 2 shows that in the traditional no-shop process thesell-side advisors, on average, contact 32 buyers, of whom 16 sign confidentialityagreements. This means that a typical PIE buyer, with no source of private valuein the auction, has a 3% chance of winning at the outset, and only a 6% chance ofwinning after investing in due diligence. In contrast, a P/E buyer who can negotiateexclusively with a seller in a pure go-shop deal has an 86% chance of closing thedeal if management is not involved (with deal jumpers winning in only 3 out of 22pure go-shop non-MBOs) and a 100% chance of closing the deal if managementis involved in the buyout group-far better odds if one's primary goal is to closedeals. Put another way, PIE firms are forced to play a "volume game" because ofthe enormous pools of capital that they have to invest."' The empirical evidencepresented here indicates that the go-shop provision is a far more effective way ofplaying a volume game (i.e., deploying capital by closing deals) than the traditionalno-shop route. The cost of course arises in lower returns for funds, on average,which will emerge as these funds are liquidated over the next five to ten years.Thus the desirability of the go-shop clause from the buyer's perspective may be, atleast in part, a manifestation of agency problems between the GPs and LPs in pri-vate equity firms. This conclusion, if correct, would also explain why P/E buyersare more attracted to the go-shop technology than strategic buyers." 2

While the data suggests a positive overall assessment of go-shops from the per-spective of target shareholders, there is cause for concern in the subset of go-shops in which current management is part of the buyout group. The fact that nohigher bidder has emerged in an MBO go-shop to date (after nearly two years of

111. 1 thank David Levine, Harvard JD/MBA '09, for this point.112. A recent theoretical paper compares the expected outcome from an auction-like mechanism

(resembling in key respects the traditional no-talk process) and the expected outcome from a sequen-tial bidding mechanism (resembling in key respects the go-shop process). See Jeremy Bulow & PaulKlemperer, When Are Auctions Best? (Stanford Graduate School of Business Research Paper No. 1973,June 2007) (on file with The Business Lawyer). While the authors conclude from their analysis that"an auction is usually more profitable for a seller than a sequential process" (emphasis in original), themodel does not incorporate the buy-side agency issues suggested here that may cause bidders to paymore for exclusivity. Id. at 2.


experience with go-shops, in a frenzied deal environment) suggests that third par-ties may be wary of entering a bidding contest, or that bankers might not conductas thorough and energetic a search, when management has already picked its pre-ferred buyout partner. A management team with difficult-to-acquire firm-specificskills and knowledge can use its inherent advantage to buy the company fromthe public shareholders at a lower price by effectively committing to its favoredbuyout group and making clear its unwillingness to work with any other buyoutgroup that might emerge during the go-shop process.

One potential solution would be a bright-line rule that management may notnegotiate its employment contract with any buyout partner until the board haspicked the bidder that delivers the most value for its shareholders. Consistentwith this approach, the Delaware Court of Chancery recently struck down a pro-posed settlement of shareholder litigation due to the CEO's precommitment to adeal with the acquirer (though this case was not in the go-shop context).1 3 Whilethis solution would likely create a more viable post-signing go-shop process inMBOs, there are serious problems with such an approach. The first is one ofenforcement: a bright-line rule would prevent formal employment agreementsbetween the buyout group and managers, but it could not prevent implicit under-standings, or private conversations, that inevitably would result during the duediligence process. A second problem is one of effectively capturing the deterringactivity: while it is feasible to prohibit employment agreements between the buy-out group and the management team, it is not feasible (at least without radicalchange in our current corporate law doctrine of conflict transactions) to preventmanagement from being part of the buyout group (that is, prohibiting MBOs).Because of the strong efficiency reasons to permit MBOs, prohibiting MBOs inthe interest of encouraging third-party bidders during the go-shop process wouldseem to throw the baby out with the bathwater.

A third, and perhaps most important, problem arises on the buy-side. In manybuyouts the current managers are essential to the deal, and the P/E firm will quitenaturally be unwilling to proceed if the seller cannot "deliver" management (in theform of signed employment agreements) at the closing. Management too mightlook for alternatives outside the company if they are not guaranteed employmentin the continuing enterprise. The irony, then, of prohibiting employment agree-ments in pure go-shop MBOs is that P/E firms will pay a lower price becausemanagement has left or might leave-precisely what such a rule is intended toavoid by facilitating third-party bids.

The better approach, then, is to regulate pure go-shop MBO deals throughgreater scrutiny of the specific features of the go-shop. As a starting point courtsshould look for (and therefore boards should insist on) what are becoming com-monplace features in all go-shops, such as a longer (fifty to sixty day) windowto shop, the identification of a potential Superior Offer (rather than a signeddeal) during the go-shop period, and a bifurcated termination fee. The Delaware

113. See In re SS & C Techs., Inc., S'holders Litig., 911 A.2d 816, 820-21 (Del. Ch. 2006).


courts have already sent strong signals that these are important features to ensurea meaningful go-shop process.1 4 Practitioners also stress these elements in theirguidance to clients.15

In addition to these basic deal structuring points, courts should look for simpleinformation rights (i.e., keeping the initial bidder informed about higher bidsthat emerge), rather than the increasingly commonplace match right, in pure go-shop MBOs. Although Delaware courts have endorsed match rights in the buyoutcontext,' 16 basic economic theory and the strong weight of practitioner anecdotalevidence indicates that match rights deter bids. Particularly in the case of a purego-shop MBO, the Delaware courts should reverse course to consider a match rightas a strong negative factor in the standard Revlon analysis.

Even better, sell-side boards should push for a provision in the merger agree-ment that explicitly prevents the MBO group from participating in any post-signing auction." 7 While Delaware courts have expressed a strong distaste forcontractual precommitments that prevent a board from accepting a higher-valuebid that comes later,'1 8 this go-shop structuring would induce the MBO groupto put full value on the table in its initial bid. Delaware courts should formallyacknowledge what is well-known among practitioners, that a loyal sell-side boardwill not necessarily keep itself open to higher bidders up until the moment of

closing.In addition to not deterring third-parties through match rights, Delaware courts

should look favorably on inducement fees that are specified as part of the initialmerger agreement. For example, a target board might extract a concession fromthe MBO group to reimburse out-of-pocket expenses for any third-party buyerwho makes a bona fide superior proposal that is at least 5% higher than the dealprice, even if the third party does not win in the end. This deal term would be

114. See In re Topps Co. S'holders Litig., 926 A.2d 58, 84 (Del. Ch. 2007) (commenting favorablyon forty-day go-shop period and bifurcated termination fee); In re Lear Corp. S'holder Litig., 926 A.2d94, 119-20 (Del. Ch. 2007) (expressing skepticism about the go-shop because the whole deal had tobe signed up during go-shop period).

115. See, e.g., Sharon Geraghty, Canada: Emerging Trends in Deal Protection Techniques, ToRYs LLP,Nov. 16, 2006, http://www.mondaq.com/article.asp?articleid=44346 ("The key negotiated elements ofthese go-shop provisions are the duration of the solicitation period and whether a break fee will haveto be paid if the target is successful.").

116. See, e.g., In re Pennaco Energy, Inc. S'holder Litig., 787 A.2d 691, 707 (Del. Ch. 2001); In reToys "R" Us, Inc. S'holder Litig., 877 A.2d 975, 1015-21 (Del. Ch. 2005).

117. Kerzner International is the only deal of which I am aware that used a go-shop with this fea-ture. This deal is not included in my sample because Kerzner was incorporated in Bermuda.

118. See Omnicare, Inc. v. NCS Healthcare, Inc., 818 A.2d 914, 936-39 (Del. 2003). While it isof course not trivial to recommend a reversal of Delaware Supreme Court precedent, Omnicare wasa highly unusual three-two decision that is perceived to be weak precedent among practitioners,academics, and even other judges. Post-Omnicare decisions in the Delaware Court of Chancery havealready cut back on its scope. See, e.g., Orman v. Cullman, Civ. A. No. 18039, 2004 WL 2348395, at*5 (Del. Ch. Oct. 20, 2004) (noting that here, unlike in Omnicare, voting agreement only preventeddefendants from exercising their fiduciary duties as shareholders and not as officers or directors). Torenounce formally a doctrine that categorically prevents contractual precommitments in merger agree-ments would open the dealmaking space to permit innovations such as the one proposed in the text.Finally, the change in the membership of the Delaware Supreme Court since the Omnicare decisionsuggests that an explicit reversal is not implausible.


a radical innovation because in every case that I am aware of an inducement fee isnegotiated ex post, i.e., after a bidder appears. An ex ante inducement fee never-theless should be part of the target board's negotiation toolkit, particularly in thego-shop MBO context, because of the strong (and correct) perception of a non-level playing field between the MBO group and potential third-party bidders.

If the MBO group already holds a significant stake in the target, sell-side boardsshould push for a contractual commitment by management to sell in to any higheroffer that emerges during the go-shop period.' 9 The corollary to this point is thatsell-side boards should not allow the MBO group to lock up its shares with its buy-out partner because locked up inside shares would naturally deter third parties. 20

The weight given to these two deal features should increase as management's stakein the target increases. For example, consider a case where management holds30%, initiates a going-private transaction using a pure go-shop process, and com-mits irrevocably to voting its shares for the deal with its preferred buyout partner.It would not seem unreasonable to invalidate the go-shop solely on the basis ofthe locked up shares, even if all other aspects of the sale process are pristine.

Finally, courts should take a close look at the negotiation process to ensure thatsomething was extracted in exchange for exclusivity in a pure go-shop MBO-ideally, a higher price. This proposal mirrors prior policy recommendations thatJohn Coates and I have made that target boards should extract something in ex-change for deal protection in traditional arms-length M&A deals. 2 ' Even for atarget board that does not wish to engage in a pre-deal market canvass for fear ofdisrupting the business, giving up the right to canvass the marketplace prior tosigning has value and therefore should be paid for by the MBO group.

To summarize, Delaware courts should presume that an add-on go-shop pro-cess satisfies the target board's Revlon duties, unless the pre-signing process is sotruncated as to more resemble a pure go-shop than the traditional no-shop route.Within the category of pure go-shops, and particularly in the case of pure go-shopMBOs, courts should scrutinize the specifics of the go-shop structure to determinethe viability of the shopping process. Long go-shop windows, small and bifurcatedbreakup fees, information rights (rather than matching rights), ex ante inducementfees, and contractual commitments by the buyout group to sell into any higher offershould all weigh in favor of the target board satisfying its Revlon requirement. Byextension, target boards and their advisors should push for these deal terms in thecontext of a pure go-shop MBO. This insistence on specific process features wouldfollow the approach that the Delaware courts have taken with respect to other typesof conflict transactions.'22 In addition, the proposed distinction between go-shop

119. Cf. Lear, 926 A.2d at 97-98 (Icahn agreed to sell his 24% stake into any non-matched supe-rior offer, though this was not an MBO).

120. See, e.g., Everlast Worldwide Inc., Form DEFA14A (June 1, 2007) (noting that Seth Horowitz,president of Everlast, agreed to vote his 19.2% stake in favor of the Hidary transaction and against anycompeting transaction).

121. See Coates & Subramanian, supra note 33, at 381.122. See, e.g., In re Pure Res., Inc. S'holders Litig., 808 A.2d 421, 444 (Del. Ch. 2002) (requiring a

non-waivable majority-of-the-minority condition, a promise by the controlling shareholder to execute


MBOs and go-shop non-MBOs would track the well-developed distinction in cor-porate law between conflict transactions and non-conflict (arms-length) transac-tions. While conflict transactions are more suspect and therefore are subjected tohigher scrutiny (namely, entire fairness review) by courts, the Delaware courts havehistorically been willing to forego a substantive fairness inquiry if appropriate pro-cedural protections are in place.123 The formulation of go-shop doctrine proposedhere takes precisely this approach.

VI. CONCLUSION

This Article presents the first systematic empirical evidence on the effect ofgo-shop provisions, which have emerged as an important new dealmaking tech-nology during the private equity boom of 2005-07. Contrary to the weight ofpractitioner commentary to date, I find little reason for categorical skepticism ofgo-shops from the perspective of target shareholders. On average, go-shops yieldmore aggregate search, significant post-signing competition, and slightly higherreturns to target shareholders than traditional no-shop deals. A closer look at thedynamics of the negotiation process suggests that target boards are often able toextract a higher price from a potential acquirer in exchange for exclusivity duringthe pre-signing phase. P/E buyers are willing to pay for such exclusivity giventheir need to play a "volume game" in order to deploy the massive amounts of un-invested P/E capital that exist today The doctrinal implication of these findings isthat go-shop provisions, appropriately structured, can satisfy a target board's Rev-Ion duties. The transactional implication of these findings is that go-shops can bea "better mousetrap" in deal structuring-a "win-win" for both buyer and seller.

There is nevertheless a potential "dark side" to go-shop provisions in the con-text of MBOs. I find no post-signing competition in go-shop MBOs during mysample period, consistent with practitioner wisdom that MBOs give incumbentmanagers a significant advantage over other potential buyers. This (non)findingsuggests that target boards of directors and Delaware courts should pay close at-tention to the precise structure of go-shops in the MBO context. Specifically, longgo-shop windows, small and bifurcated breakup fees, information rights (ratherthan matching rights), ex ante inducement fees, and contractual commitmentsby the buyout group to sell into any higher offer should all weigh in favor of thetarget board satisfying its Revlon requirement.

a prompt short-form merger at the deal price if it obtained more than 90% of the shares, and no "re-tributive threats" by the controlling shareholder in order for a tender offer freeze-out to be deemednon-coercive by the court).

123. See ALLEN, KRAAKMAN & SUBPAMANIAN, supra note 50, at 321-22.

Download - 63 Bus Law 729

Top Related