Federal Trade Commission v. , Inc. — Copyright 2004 Dechert LLP. All rights reserved. Materials have been abridged from laws, court decisions, and administrative rulings and should not be considered as legal opinions on specific facts or as a Asubstitute View for legal counsel. From the Dugout

by Paul Friedman and Gorav Jindal Dechert LLP

The Threshold, Vol. III, No. 1 Fall 2007

This information or any portion thereof may not be copied or disseminated in any form or by any means or downloaded or stored in an electronic database or retrieval system without the express written consent of the www.dechert.com American Bar Association. Fall 2007 The Threshold Vol. VIII, No. 1

Federal Trade Commission v. Whole Foods Market, Inc.–A View From the Dugout [top]

Paul Friedman1

Partner, Dechert LLP

Gorav Jindal

Associate, Dechert LLP

The last issue of The Threshold likened the of Columbia Circuit denied the FTC’s emer- Federal Trade Commission’s (“FTC”) suit to gency motion for an injunction pending ap- block the acquisition of Wild Oats Market, peal. The parties consummated their transac- Inc. (“Wild Oats”) by Whole Foods Market, tion shortly thereafter. The FTC’s appeal from Inc. (“Whole Foods Market”) to one of three the district court’s order remains pending. baseball pitches swung at by the FTC’s Bureau Whole Foods Market has moved to dismiss the of Competition. As this article goes to publi- appeal as moot. The FTC opposes dismissal cation, the battle for this year’s World Series of its appeal. The FTC’s administrative com- championship has recently concluded so we plaint remains pending as well, although the will stick with the baseball metaphor. Some administrative case was stayed in August by loyal fans (mostly FTC alumni) sized up the order of the Commission. FTC’s odds of getting a hit and concluded that the FTC was going to hit a home run. Pat- Our perspective is unique. We got called up terned after FTC v. Staples, Inc., the complaint late, joining a talented team at the top of the teemed with references to “monopoly” mar- seventh inning as Whole Foods Market re- kets and provocative allegations about the tained our firm about one week after the FTC transaction’s effects. Yet, the parties dug in, issued its complaint. The purpose of this arti- contesting the FTC’s view of and cle is not to discuss in depth the judge’s deci- the competitive realities. The battle lines were sion, which as noted remains on appeal, but drawn and the lineups were set for trial. rather to provide a hopefully valuable and in- teresting litigant’s perspective on a very sig- The FTC did not prevail. The district court nificant and widely-reported preliminary in- denied the FTC’s request for a preliminary in- junction proceeding. junction, and both the district court and the United States Court of Appeals for the District “The Prelude to the Game”– The Transaction and The

1 Investigation The authors represented Whole Foods Market, Inc. during the preliminary injunction hearing discussed in this article, and also are counsel to it in the subsequent Whole Foods Market first opened its doors in proceedings. 1980 in Austin, Texas as a store focusing on healthy foods. Over time, Whole Foods Mar- ket expanded the range of products offered to

3 Fall 2007 The Threshold Vol. VIII, No. 1

attract a broader universe of customers and to the court entered a temporary restraining order increase sales. Its newest stores, at 50,000– until the court ruled on the motion for a pre- 60,000 square feet, are comparable in size to liminary injunction. other . Before the acquisition, Whole Foods Market operated 194 supermar- “The Schedule and the Rules kets in the United States. Wild Oats, another of the Game” chain with a historical focus on natural, healthy foods, operated 115 stores in The FTC and the defendants had largely the United States in 2007 under three banners: agreed on a schedule for discovery, briefing Wild Oats Marketplace, Henry’s Farmers and trial, as well as the rules under which the Market, and Sun Harvest. Wild Oats’ stores preliminary injunction hearing would proceed, were smaller, averaging about 24,000 square before we came on board. Both the schedule feet. and the rules significantly affected litigation strategy. In February 2007, Whole Foods Market agreed to buy the stock of Wild Oats at a per share The agreed-upon Case Management Order price of $18.50, or $565 million. The FTC (“CMO”) called for a preliminary injunction issued a Second Request on March 13; there hearing in less than two months. The schedule were 26 documents and data specifications, leading up to the hearing was ambitious but, as including many of the standard requests in the it turned out for both sides, manageable.3 To Model Second Request for transactions. assist the court, the parties provided expert re- The FTC did not, however, request any pricing 2 ports and witness declarations to the court si- data. In response, Whole Foods Market and multaneous with the exchange among the par- Wild Oats produced a total of about 16.5 mil- ties. This enabled the court to review the evi- lion pages of documents and over 2.4 giga- dence in advance of the hearing and to prepare bytes of data. The FTC also conducted inves- its own questions for the witnesses and the tigational hearings of five employees each of lawyers at the hearing. Whole Foods Market and Wild Oats, the for- mer CEO of Wild Oats, and the CEO of Trader The hearing itself was streamlined. The only Joe’s. testimony presented live was the cross- examination and redirect examination of ex- On June 6, 2007, after conclusion of the FTC’s pert witnesses. The format, therefore, placed a investigation, the Commissioners voted 5-0 to file suit in the District of Columbia District Court seeking a temporary restraining order and preliminary injunction to prevent Whole 3 The CMO required the parties to exchange final Foods Market from acquiring any stock or as- lists of fact witnesses, including a brief summary of their anticipated testimony, on June 27, 2007. Fact dis- sets of Wild Oats. With the parties’ consent, covery closed nine days later, on July 6. Initial expert reports were due on July 9, rebuttal expert reports were due on July 13, and expert depositions were to be taken 2 Specification 5(e) of the FTC’s Model by July 18. Declarations of fact witnesses were due on Second Request for transactions involving July 13. Opening briefs on the motion for preliminary retail industries requests pricing data for the top injunction were due on July 20, reply briefs were due by 2000 UPCs by annual dollar sales. See July 25, a two-day hearing was scheduled for July 31- http://www.ftc.gov/os/2004/04/040428modelrequest. August 1, and proposed findings of fact and conclusions pdf. of law were due by August 3.

4 Fall 2007 The Threshold Vol. VIII, No. 1

significant premium on the clarity of declara- durable competitive rivalry between Whole tions, the selection of company documents, Foods Market and Wild Oats; and (2) econo- and the selection of deposition testimony to metric analyses conducted by the FTC’s eco- highlight to the court. nomic expert, Dr. Kevin Murphy, predicted anticompetitive effects if the merger were con- “The FTC’s Lineup”–The summated. Case Against the Transaction The FTC’s case featured statements by Whole Foods Market CEO John Mackey, which have The FTC’s case depended on its narrow view been featured in press reports across the coun- of the relevant product market. In a departure try. For example, in one e-mail, Mackey from prior merger enforcement involving food wrote: retailing,4 the FTC alleged a product market consisting of “premium and natural organic By buying [Wild Oats] we will greatly supermarkets” (“PNOS”). The FTC resisted enhance our comps over the next few defining, until compelled by the court, the pre- years and will avoid nasty price wars in cise characteristics of participants in the mar- Portland (both Oregon and Maine), ket it alleged. Eventually, the FTC listed ten Boulder, Nashville, and several other qualitative attributes that a market participant 5 cities which will harm our gross mar- “generally,” but not always, would possess. gins and profitability. Furthermore, we According to the FTC, only four firms in the eliminate forever the possibility of entire United States met its definition of , Super Value, or using PNOS: Whole Foods Market, Wild Oats, their brand equity to launch a compet- Earth Fare (a Southeastern regional supermar- ing national natural/organic food chain ket operator), and New Seasons (an Oregon- to rival us . . . . Eliminating them based operator with less than ten stores in the means eliminating this threat forever, Pacific Northwest). The FTC relied on two or almost forever.6 main lines of evidence to support its position, arguing that: (1) company documents, primar- The FTC also featured statements by Wild ily statements by Whole Foods Market’s cur- Oats’ ex-CEO as evidence of what the FTC rent CEO and statements by Wild Oats’ former described as “unique” competition between the CEO, showed the existence of a unique and two firms.

Batting cleanup for the FTC was its economic

4 expert, Dr. Kevin Murphy, the George J. Stig- A review of the FTC’s prior supermarket enforce- ler Distinguished Service Professor of Eco- ment matters reveals that in all eighteen of such matters since the adoption of the 1992 FTC Horizontal Merger nomics at the University of Chicago. Guidelines, until this merger, the FTC had concluded Dr. Murphy testified live at the preliminary that the relevant product market was at least as broad as injunction hearing in district court. the retail sale of food and grocery items in supermar- Dr. Murphy purported to show that Wild Oats kets. was Whole Foods Market’s closest competitor 5 Curiously, however, the FTC offered no principled and that the acquisition of Wild Oats by Whole means of determining whether a store met these criteria, admitting that the stores in the alleged product market would not necessarily possess every attribute or even possess them to the same degree. 6 FTC Complaint, June 5, 2007, Introduction.

5 Fall 2007 The Threshold Vol. VIII, No. 1

Foods Market would be “unambiguously anti- testimony of the only industry expert in the competitive.” case, overwhelmingly established significant, meaningful, intense and intensifying, effective “The Heart of the Defendants’ competition between Whole Foods Market and Lineup”–Our Case a broad range of other supermarkets. Whole Foods Market’s senior operating man- The defendants’ case focused on market reali- agers, including its two co-presidents, and the ties as shown by contemporaneous, ordinary regional presidents, provided critical evidence course of business records, supported by ex- that supported a broader product market than pert testimony. Defendants contended that this that proposed by the FTC. Each Whole Foods evidence showed that major supermarket op- Market and Wild Oats witness declaration erators had recognized and responded to the cited to dozens of ordinary course of business growth in consumer demand for natural and documents to support the statements each organic foods by positioning themselves to made. In total, defendants incorporated about protect existing market share and expand sales 70% of the 770 documentary exhibits into 14 of natural and organic products. At the same party sworn statements. In this way, the decla- time, Whole Foods Market had expanded its rations provided vivid, first-hand accounts of stores to reach consumers that may not have the actual competition Whole Foods Market historically shopped at Whole Foods Market. faced every day, describing why Whole Foods Defendants argued that while Whole Foods Market monitored the prices of, and adjusted Market may have started with a mission to sell its prices in responses to, supermarkets such as primarily natural and organic foods, its stores Kroger or Safeway or Stop ‘n Shop.7 And no longer reflected such a singular focus. while Whole Foods Market did compete with Rather, a typical Whole Foods Market store Wild Oats, defendants contended that it was today carries about 30,000 stock keeping units, not uniquely close competition as character- including a wide variety of conventional prod- ized by the FTC. The evidence showed that ucts. Many of Whole Foods Market’s prod- the prices of Wild Oats were typically higher ucts are not organic, including more than half than those of Whole Foods Market, that Whole of its produce and a significant percentage of its prepared foods, baked goods, and specialty items. Thus, while Whole Foods Market and Wild Oats continued to market themselves as a 7 While at first blush it may have appeared that the healthy-foods store, each faced ever-growing district court agreed with the FTC’s argument that the parties’ sworn statements were self-serving statements competition from “conventional” supermarkets entitled to little weight, a careful reading of the court’s in the sale of natural, organic, fresh, healthy opinion shows that the parties’ strategy resonated with products, and that competition was projected the court. The court agreed with the FTC only to the to intensify as supermarkets continued to mod- extent that they were in conflict with contemporaneous ify their formats to capture ever larger vol- business documents. See FTC v. Whole Foods Market, Inc., 502 F. Supp. 2d 1 (D.D.C. 2007), at 4, n. 4. Since umes of this growing segment. the declarations drew on contemporaneous documents, however, defendants argued that there were no conflicts. Defendants contended that the FTC’s proposed The decision, therefore, drew heavily on these sworn product market failed because it was simply declarations and the company documents on which they counterfactual. Defendants argued that the were based, as Judge Friedman cited extensively to the great weight of the evidence, including com- defendants’ declarations on key issues throughout the opinion. pany documents, market studies, econometric evidence, the views of third parties, and the

6 Fall 2007 The Threshold Vol. VIII, No. 1

Foods Market had developed no practices, vigorous competition between those retailers policies, or strategies directed specifically at and Whole Foods Market. And most impor- Wild Oats, and that Whole Foods Market did tantly, the FTC elected not to cross-examine not view Wild Oats as a competitive con- Dr. Stanton on his analysis of industry trends straint. in response to changes in consumer demand, and offered no industry expert of its own. In addition to sworn testimony from company officials, Whole Foods Market and Wild Oats The FTC’s specific response to Dr. Stanton offered testimony from an industry expert, was that his views should be discounted be- Dr. John Stanton, and two sworn statements cause they were not informed by ordinary from third party manufacturers to drive home course of business documents from the defen- three main points about the industry. First, dants or third parties. But the court disagreed, these witnesses stated that Whole Foods Mar- noting that the “state of the industry itself is an ket possesses all of the attributes of a super- important factor in a case like this,” and that market. It offers the same range and depth of the FTC’s failure to offer its own expert or to product offerings as other supermarkets. Its cross-examine Dr. Stanton at trial was a strate- focus on natural, organic, fresh, and gourmet gic choice made by the FTC.8 foods merely serves as its point of differentia- tion. Second, the witnesses supported the The parties presented economic evidence from proposition that differentiation is not the basis two-time FTC Bureau of Economics Director, for defining the relevant product as the FTC Dr. David Scheffman. In expert reports, depo- had alleged; it is the means by which super- sitions, and live testimony, Dr. Scheffman of- markets compete with each other. The evi- fered three main opinions, the first two of dence showed that supermarkets differentiate which addressed the FTC’s alleged product themselves from one another to encourage a market while the last addressed whether the consumer to drive past the closest supermarket transaction would produce anticompetitive ef- if it is not theirs. Finally, defendants presented fects. First, Dr. Scheffman conducted a criti- abundant evidence that the growth in con- cal loss analysis examining the likely response sumer demand for natural, organic, and fresh by consumers if the prices of groceries in foods that had been sweeping across the coun- PNOS rose relative to the prices of groceries in 9 try over the past five years had received con- other supermarkets. siderable attention from food retailers. Su- permarkets were investing significantly to Using either a 1% or 5% hypothetical price adapt by changing formats and adding natural, increase, Dr. Scheffman determined that the organic, fresh, and gourmet foods, which eroded the “uniqueness” that previously served as Whole Foods Market’s primary point of dif- 8 Id. at 13. Dr. Stanton’s report is cited fifty-five ferentiation. The evidence showed that those times in Judge Friedman’s opinion. investments were ongoing. 9 Critical loss analysis attempts to determine the per- centage of lost sales at which a hypothetical price in- While the FTC tried to rebut these points, De- crease is unprofitable. The break-even point is called fendants argued that it offered no evidence the critical loss. The analysis is conducted by compar- ing the magnitudes of actual loss, which is estimated from manufacturers to suggest a unique prod- lost volume at the hypothetical price increase, and criti- uct market of premium natural and organic cal loss. If the actual loss is less than critical loss, the supermarkets. Although there was some evi- price increase is profitable. Otherwise, it is not. dence from other food retailers, defendants contended that it all confirmed the existence of

7 Fall 2007 The Threshold Vol. VIII, No. 1

actual loss would exceed critical loss.10 This other grocery retailers, cannot be reconciled means a projected price increase would not be with the FTC’s views that competition was profitable. Dr. Scheffman determined that the primarily confined to PNOS. actual loss would be high because a large number of customers representing a substantial Finally, Dr. Scheffman analyzed Whole Foods volume of purchases are marginal. And he Market and Wild Oats prices. His cross- testified that it is these marginal consumers, sectional pricing analysis, which examined rather than core consumers, for which firms actual prices (i.e., the prices that are loaded compete because they are susceptible to be into the cash registers at each store), found no won or retained through better prices, im- variation in Whole Foods Market or Wild Oats proved service, higher quality or more diverse pricing based on the presence or absence of the product selection. other, an empirical observation that was incon- sistent with the FTC’s theory of the relevant Second, Dr. Scheffman analyzed the source of market but, as was later pointed out by the sales for new Whole Foods Market stores and court, consistent with the other credible evi- found that its sales were generated overwhelm- dence in the case. ingly from other supermarkets regardless of the existence of a nearby PNOS. Under the “Pitching to the FTC’s Heavy FTC’s theory of the market, in areas with Wild Hitters” Oats (or another PNOS), Whole Foods Mar- ket’s sales should have been taken predomi- Every great player is fiercely competitive. Ty nantly from Wild Oats. As analyzed by Dr. Cobb, arguably the greatest hitter in baseball Scheffman, however, the data showed that behind Babe Ruth, was known for his hatred Whole Foods Market sales came predomi- for his opponents. In this respect, Whole nantly from other supermarkets. He also testi- Foods Market CEO John Mackey is no differ- fied that in areas without PNOS, all of Whole ent. Foods Market new store sales came from other grocery retailers, which the FTC contended Mr. Mackey whole-heartedly believes in were not competitors in the relevant market. Whole Foods Market, its business model, and Dr. Scheffman concluded that these data, its people. He is literal and outspoken, and which showed that the main competitive inter- does not like his competitors. While the FTC action was between Whole Foods Market and heavily relied on statements of Mr. Mackey concerning Wild Oats, defendants pointed out to the court that in deposition testimony and 10 Since there was no evidence to determine the cross- documents, Mr. Mackey used similar tough elasticity of demand between PNOS and other super- talk to describe the full range of his competi- markets, Dr. Scheffman based his estimate of actual loss tors, including supermarkets that the FTC did on, among other things, information and data contained not regard as part of the market in which in forty-seven market studies, which revealed that gro- Whole Foods Market competed. cery shopping is highly price sensitive, the absence of a large, definable core group among PNOS consumers, that consumers can and do shift purchases between As defendants showed, Mr. Mackey repeatedly PNOS and other supermarkets, that most Whole Foods testified and wrote about competition with Market and Wild Oats consumers also shop at other many supermarkets and other food retailers as supermarkets, and that other supermarkets actively tar- the new competitive reality for Whole Foods get Whole Foods Market and Wild Oats consumers. Market. He testified that, while Whole Foods Market may have once flown under the radar

8 Fall 2007 The Threshold Vol. VIII, No. 1

of larger supermarkets, as its business grew its gin on a shopping basket can vary with the threat to these supermarkets grew as well. product mix, and the failure to account for mix Larger supermarkets noticed Whole Foods can undermine the reliability of the analysis. Market’s growth and mimicked its ingredients “Partial shrinkage” can occur when merchan- for success. Mr. Mackey stated on many occa- dise is marked down for quick sale. Neither of sions that Whole Foods Market has never the defendants tracked “partial shrinkage.” Dr. faced more competition than it does today. Murphy’s economic models did not consider Defendants argued that, taken as a whole, the “partial shrinkage,” which can be most pro- full record of Mr. Mackey’s statements proved nounced in periods of competitive entry. De- that he is intensely competitive and keenly fendants contended that these deficiencies ren- aware of the threats facing his company from dered any conclusions from these analyses competitors across the spectrum of food retail- suspect. Dr. Murphy offered no direct evi- ing, and not just Wild Oats.11 dence that Wild Oats imposed a unique price constraint on Whole Foods Market. The evi- Defendants’ attack on Dr. Murphy, the FTC’s dence revealed that Wild Oats’ prices were, in expert, was more difficult because of his solid fact, higher than Whole Foods Market’s prices. credentials and the mountain of econometric Dr. Murphy attempted to use margins to show analyses on which he relied. The FTC de- that Whole Foods Market margins varied pended on three lines of Dr. Murphy’s analysis across stores depending on the presence of a to support its position about both market defi- Wild Oats store nearby, and made inferences nition and the transaction’s likely competitive about price from these results. But effects: (1) his analysis of the effect of Whole Dr. Murphy testified on cross examination that Foods Market’s entry on Wild Oats’ margins; his analysis produced a result that was not sta- (2) his cross-sectional analysis of margin data tistically significantly different from zero. De- to show that margins at Whole Foods Market’s fendants thus argued that Dr. Murphy’s cross- stores varied based on the presence or absence sectional analysis showed that Whole Foods of Wild Oats; and (3) his study of the effect of Market’s margins were not dependent on the Whole Foods Market entry on Wild Oats’ presence or absence of Wild Oats. prices. Dr. Murphy also sought to draw inferences Through cross examination, defendants sought about the effects of potential Wild Oats store to show that Dr. Murphy studied the wrong closings on Whole Foods Market’s prices by events and used the wrong measures. Specifi- studying the effect Whole Foods Market’s en- cally, defendants tried to establish, and believe try had on Wild Oats’ prices and margins. they did establish, that Dr. Murphy’s use of Dr. Murphy testified that one could predict the margin data rather than prices was fraught effect on Whole Foods Market’s prices of with problems because of the inability to con- closing Wild Oats stores from the effects trol for mix or “partial shrinkage.” The mar- Whole Foods Market had on Wild Oats. In essence, Dr. Murphy asked the court to accept that the exit of a high-priced firm (Wild Oats) 11 It was, therefore, only mildly surprising to find no would have the same effect as the entry of a reference to John Mackey’s statements in the district court opinion. In context, Mackey’s statements were lower-priced, higher quality firm (Whole consistent with the view that Whole Foods Market com- Foods Market). The court disagreed, noting petes with many supermarkets, one of which was Wild that it was “unwilling to accept the assumption Oats. that the effects on Wild Oats from Whole Foods Market’s entries provide a mirror from which predictions can reliably be made about

9 Fall 2007 The Threshold Vol. VIII, No. 1

the effects on Whole Foods Market from Wild 12 “A Post-Game Comment on Oats’ future exits if this transaction occurs.” the FTC’s Lineup Against Finally, Dr. Murphy examined the effect of Whole Foods Market vs. Its Whole Foods Market entry on Wild Oats Lineup Against Staples, Inc.” prices. Although Dr. Murphy selected five entry events to study, he relied on the results 13 In FTC v. Staples, Inc., the FTC successfully from only two–West Hartford, CT and Fort proved a novel, narrow product market– Collins, CO. From these two events, namely, the sale of consumable office supplies Dr. Murphy concluded that Wild Oats’ prices through office superstores.14 Staples also pro- are lower following Whole Foods Market en- vided a blueprint for analyzing horizontal re- try. Defendants contended that the study was tail mergers. The differences between the evi- flawed, however, for three main reasons. dence the FTC offered in Staples compared to First, a finding that Wild Oats prices are lower that offered by the FTC in Whole Foods Mar- after Whole Foods Market entered West Hart- ket is striking, and may explain the different ford, CT was not a good predictor about results. whether Whole Foods Market would raise prices in West Hartford, CT or anywhere else. • Declarations. In Staples, the FTC had 36 Second, Dr. Murphy’s analysis failed to con- third party declarations while in the Whole trol for entry by other firms. In Fort Collins, Foods Market case, it had only one–a CO, for example, , a supermarket statement from a food retailer that was not owned by Kroger, entered within five miles of even a supermarket. Wild Oats at essentially the same time as Whole Foods Market entered. But Documents and entry analyses. In Staples, Dr. Murphy attributed the entire price effect to • the defendants’ documents, including Whole Foods Market’s entry. Third, Wild analysis of new market entry, showed that Oats, in fact, exited Fort Collins, and yet they focused primarily on competition Dr. Murphy did not study Whole Foods Mar- from other superstores. The defendants’ ket’s prices in Fort Collins after Wild Oats’ documents in Whole Foods Market, in- exit. cluding new market entry analyses, fo- cused primarily on other supermarkets, not PNOS.

• Pricing vs. margin data. In Staples, the FTC focused on pricing evidence while the FTC focused on margin evidence in Whole Foods Market.

12 FTC v. Whole Foods Market, Inc., 502 F. Supp. 2d at 21. 13 Dr. Murphy’s findings in West Hartford, CT, how- ever, were skewed because of the relative prices of each firm’s salad bars. Removing the salad bars from the pricing study essentially muted the pricing disparity. 14 970 F. Supp. 1066 (D.D.C. 1997).

10 Fall 2007 The Threshold Vol. VIII, No. 1

• Price effects. The FTC showed that Sta- While we believe the district court reached the ples priced 13% higher in one-firm mar- correct decision, and that the D.C. Circuit kets than in three-firm markets. But nei- properly denied the FTC’s emergency motion ther Whole Foods Market nor Wild Oats for a stay pending appeal, this litigation was prices were different in one-firm markets every bit a battle between two deep teams, as compared to markets in which other both championing different but vital principles PNOS existed. of competition and both committed to diamet- rically opposed views about the marketplace • Constraining effect of the acquired firm. facts. And while it is not clear whether the In Staples, the defendants were deep dis- FTC’s challenge to Whole Foods Market is counters, and the acquired firm was an ag- truly analogous to big-league baseball, it was, gressive competitor which helped drive without question, an interesting way to spend prices lower. Wild Oats, on the other the summer. hand, was a higher priced competitor, far from aggressive, and had no effect on Whole Foods Market pricing.

11