at the Department of Justice
November 15, 2007
There have been a variety of reports in recent months suggesting that merger enforcement at the Department of Justice Antitrust Division is less vigorous than it once was, and less vigorous than it should be. I am tempted to respond by adapting the famous quotation attributed to Mark Twain: "The reports of [the] death [of our merger enforcement program] are greatly exaggerated."(1) In the process of assuring that I was quoting Twain accurately, I realized that his true remarks are even more apt. The popular version of this quotation traces to a handwritten note penned by Samuel Clemens in 1897:
James Ross Clemens, a cousin of mine was seriously ill two or three weeks ago in London but is well now. The report of my illness grew out of his illness, the report of my death was an exaggeration.(2)
Like those who were confused about Clemens' death 110 years ago, those who misperceive the demise of the Division's merger enforcement program have not been paying close attention.
Merger enforcement is one of the Division's highest priorities, second only to our aggressive program of rooting out and prosecuting hard core criminal cartel behavior. We recognize that most mergers are procompetitive, or at least competitively neutral, and we will not interfere with those transactions. But some proposed transactions would be anticompetitive, creating or enhancing market power that would disrupt the competitive processes and harm consumers, and we will not hesitate to enforce the law to protect competition and consumers. No doubt many more harmful transactions would be proposed were there no regime of antitrust merger review.
Our merger enforcement program aims to identify quickly and efficiently that subset of transactions likely to be harmful, and allow the others to proceed with minimal delay and regulatory burden. When we do see the potential for competitive harm, however, we take very seriously our responsibility to protect competition and consumers. First and foremost, that means applying the well-established framework of the Horizontal Merger Guidelines and bringing to bear the very best possible investigative and analytical skills to determine whether, in our view, the transaction is likely to cause substantial harm to competition. For most of the transactions we investigate every year, our legal and economic staffs very quickly conclude that such harm is not likely, and recommend that we close our investigation promptly, often without the need for a Second Request. Even when we conclude we should issue a Second Request, it does not foreordain a six month march toward full compliance. Very often we do so because we need some additional time to complete our review of a dispositive issue perhaps market definition, perhaps the likelihood of entry, perhaps some other topic or need access via formal process to a subset of company documents to confirm a conclusion that the transaction is unlikely to be harmful.
Importantly, only a small minority of our investigations lead our staff to conclude that a proposed transaction is likely to cause substantial harm. When staff does reach that conclusion, however, we are cognizant of the fact that we do not possess any authority to command that parties abandon or restructure their transaction. We are law enforcers, not sector regulators. If we believe that a transaction would violate Section 7, we must persuade a district court that we are correct and invoke the equitable power of the court to enjoin the transaction or require an appropriate remedy. Pleading and proving a Section 7 violation in court poses challenges, as some recent outcomes in litigated merger cases underscore.
I'll not opine in this forum as to the reasons for those results. Suffice it to say that winning in court is something we must do if parties choose to go forward with their transaction over our objections, and unlike the days of Von's Grocery, when the "Government always [won]"(3) victory is not guaranteed. Do we take into account whether we could prevail in court if we were put to that test? Absolutely. It would be irresponsible to bring cases that we had no reasonable expectation of winning. But let me be very clear: we are not at all deterred by the Division's loss in Oracle, or the FTC's recent losses in Arch Coal, Whole Foods, and Western Refining. Those outcomes reinforce the need of which we were already aware to pay attention to our litigation preparedness, and specifically how we would persuade a court of the correctness of the conclusion about the transaction's likely effects that we have already reached. We will continue to be vigilant in enforcing the law against transactions that would violate the antitrust laws and harm competition and consumers.
Look at our recent merger enforcement record. Since June of last year, the Division challenged or caused the restructuring of 18 transactions, and filed 15 complaints alleging violations of Section 7 of the Clayton Act. Those challenges have covered virtually the entire waterfront of potential theories of competitive harm.
I'll mention first the lawsuit we filed in May 2007 against the Daily Gazette Company and MediaNews Group challenging Daily Gazette's acquisition in 2004 of all the assets of the only other daily newspaper in Charleston, West Virginia.(4) Litigation in that case is pending demonstrating that that we are willing and prepared to litigate when the parties do not consent to adequate relief.(5) The Charleston case raises some interesting issues given its context: many of the operational aspects of the two papers including the pricing of advertising and subscriptions had been combined since 1958 in a joint operating arrangement, but the owners of those papers retained the ability and incentive to compete with one another to attract readers to their own newspaper, rather than merely enhancing the value of the venture. The acquisition extinguished that competition, and would have swiftly led to the closure of the acquired newspaper had our investigation not intervened.(6) Defendants cite Dagher and the antitrust immunity granted by the Newspaper Preservation Act in support of their motion to dismiss, which we have opposed.(7)
In many other cases we have obtained relief by settling our Section 7 claims via consent decree. Consent decrees are an "'integral part of antitrust enforcement.'"(8) They provide an efficient way to resolve our Section 7 claims, avoiding anticompetitive harm while allowing beneficial aspects of transactions to proceed, and also avoiding the many risks and burdens of adversarial litigation, including the possibility of losing the suit.(9) In still other cases our investigation led parties to fix their transaction in a manner that avoided anticompetitive harm.(10)
Some of the challenges resolved by decree addressed unilateral effects. We challenged several transactions because the firms in the market were differentiated from the perspective of consumers and we concluded that the transaction would combine the number one and two options for a substantial number of consumers. In both Cemex/Rinker(11) and Vulcan/Florida Rock,(12) announced this week, we concluded the transactions would have anticompetitive unilateral effects in several markets because, although there would remain two or three competitors post-merger, the firms were differentiated in geographic space due to high transportation costs relative to product value. In ATT/Dobson, the differentiation was in the nature of wireless telephone service offered by market participants. In the geographic markets of concern, the wireless networks wholly or partly owned by the parties "provide[d] greater depth and breadth of coverage than their PCS-based competitors" and thus, for a large set of customers, likely were "closer substitutes for each other than the other mobile wireless telecommunications services in these markets provided by firms who own only PCS spectrum."(13)
We challenged two other transactions that raised unilateral effects concerns involving undifferentiated products. In both Abitibi/Bowater(14) involving newsprint and Exelon/PSE&G(15) involving wholesale electricity generation in the PJM transmission area we concluded that the transaction would give the merged firm the incentive and ability to raise price by withholding capacity, regardless of the responses of other firms, because those other firms would lack sufficient capacity to defeat a price increase.(16)
We have also actively investigated potentially anticompetitive coordinated effects. At least three of our challenges in the past 18 months resulted from our conclusion that the transaction would increase the likelihood that the remaining firms in the market would successfully coordinate, tacitly or explicitly, to raise prices. In Mittal/Arcelor,(17) we concluded that the transaction would result in two firms the merged firm and a third party accounting for more than 80 percent of all sales of tin mill products in the eastern United States. Coordination between those firms would be profitable regardless of the response of the remaining fringe, and would be more likely to succeed given that the firms would no longer need to worry about disruption by Arcelor's Dofasco mill, located in southern Ontario. In both Cemex/Rinker and Vulcan/Florida Rock, we concluded that the transactions would have anticompetitive coordinated effects in certain markets, and in at least one such market coordinated effects were our principle concern.(18) In Vulcan/Florida Rock, for example, we alleged that the merger would reduce the number of firms competing for sales of coarse aggregate in West Atlanta from four to three, making tacit or explicit coordination more likely.(19)
In addition to challenges based on each of the categories of potential effects addressed in the Horizontal Merger Guidelines, this past year we challenged a vertical transaction that threatened harm to potential future competition. In Monsanto/D&PL,(20) the parties' proposed transaction contemplated that Monsanto would sell its Stoneville seed company, largely removing the horizontal overlap in cottonseed sales. Our investigation thus focused on the vertical aspects of the transaction: whether Monsanto's acquisition of D&PL would harm nascent competition in markets for transgenic cottonseed traits in the Southeast and South Central United States. Monsanto was the first to develop successful traits, and almost all cotton grown in these regions uses one or both of Monsanto's traits "Roundup Ready," which makes cotton tolerant to glyphosate herbicide, allowing such herbicide to be sprayed over the top of cotton containing the trait, killing weeds but not the cotton plants; and "Bollgard," which makes cotton plants resistant to many insects. We concluded that Monsanto's acquisition of D&PL would thwart or delay efforts by rival trait developers to bring competing traits to market, by disrupting ongoing development work or depriving those rivals access to the cottonseed material (germplasm) with a proven track record in the Southeast and South Central United States, where traits were most highly valued by farmers. Although we concluded that a divested Stoneville would provide a good platform for such trait development efforts, we determined that more needed to be done to avoid competitive harm, most significantly: (1) divestiture to the buyer of the Stoneville assets of rights to significant amounts of D&PL's most promising germplasm; (2) divestiture to Syngenta of the most promising D&PL germplasm containing the trait that Syngenta was developing in partnership with D&PL; and (3) modification of terms in Monsanto's licenses with third party seed companies that provided incentives to use only Monsanto traits to the exclusion of traits developed by others.
The vast majority of our merger challenges prevent the consummation of proposed transactions that would have harmed competition. This is a one of the chief benefits of the HSR Act's system of pre-notification. Even where the Act does not apply we have intervened to undo the adverse effects of transactions that have already been consummated. I have already mentioned the pending litigation involving Charleston newspapers. We also challenged Amsted's acquisition of the FM Industries (FMI) unit of Progress Rail Services, which created a monopoly in the market for end-of-car cushioning units used by certain types of railcars.(21) In Amsted/FMI, we persisted even though the defendant had already "dismantled FMI by firing its employees and disposing of virtually all FMI plant equipment through an auction."(22) A creative remedy was necessary to restore competition, so we required that Amsted sell rights to its "market-specific intellectual assets," including intangible assets and a royalty-free license to casting patterns, which together with readily available manufacturing equipment, would allow a new entrant to produce new end-of-car cushioning units in competition with Amsted.(23)
Finally, we have undertaken and are currently undertaking serious investigations of transactions involving partial acquisitions. The transactions we have seen involve many permutations of minority ownership overlap. In ATT/Dobson, two wireless telephone markets involved overlaps between Dobson and businesses in which ATT held minority interests that were small in percentage terms, but gave ATT "significant rights under each relevant partnership agreement to control core business decisions, obtain critical confidential competitive information, and share in profits at a rate significantly greater than the equity ownership share upon a sale of the partnership."(24) As a result, we concluded that ATT would "likely have the ability and incentive to coordinate the activities of the wholly-owned Dobson wireless business and the business in which it has a minority stake."(25) In the Hearst/MediaNews matter, Hearst was proposing to acquire a new class of shares of MediaNews, which operates newspapers around the country. Hearst and MediaNews were each other's principal daily newspaper competitors in the San Francisco Bay Area, and for that reason structured Hearst's equity investment as a "tracking stock" that the parties asserted was intended to give Hearst approximately a 30 percent stake only in MediaNews' newspaper operations outside the Bay Area.(26) We examined this arrangement closely, with a particular focus on whether Hearst's investment would give Hearst influence over MediaNews' business decisions relating to the Bay Area, provide Hearst with access to competitively sensitive information relating to the Bay Area, or create incentives for Hearst to disengage from aggressive competition against MNG in the Bay Area. As a result of our investigation, the parties made several revisions to their proposed relationship to avoid potential adverse effects on competition between the parties in the Bay Area.(27)
The Division's record of vigorous yet sound merger enforcement should speak for itself. Lest there be any doubt about our resolve to challenge those transactions that are likely to cause competitive harm, allow me to elaborate on a few points.
First, observers of our merger enforcement program should not infer from the fact that the last merger case that went to litigated decision Oracle(28) ended in a loss that the Division is reluctant to challenge mergers that we conclude likely would harm competition and consumers. Merging parties should not misjudge our will. If we conclude that a transaction would violate Section 7, we will be prepared to litigate to give force to that conclusion. As we have said now many times, if the facts in Oracle were presented again, we would litigate the case again.
To set the record straight, the Division has in fact litigated merger cases since losing in Oracle. We are currently in litigation in West Virginia over the Charleston newspapers matter discussed above,(29) and we were preparing to go to trial against the acquisition by Dairy Farmers of America, Inc., of a controlling interest in Southern Belle Dairy Company, LLC, when the parties settled on the courthouse steps.(30)
In addition, the Division would have litigated many more merger cases had the parties not settled our claims by offering sufficient relief to eliminate our competitive concerns protecting competition and consumers by providing all the relief we could have hoped to obtain through adversarial litigation. As I have said, each of our consent decrees reflects an efficient resolution of a complaint alleging a violation of Section 7. The relief obtained in these decrees should not be viewed as reflecting any less vigorous or effective merger enforcement than a dozen or more adversarial hearings over government requests for preliminary or permanent injunctions. The latter, while perhaps garnering greater publicity, would only have served to increase the burdens on us, the courts, and the parties, while simultaneously increasing the risk that our goals of protecting competition and consumers might have been less well served.
Counting up the number of merger challenges litigated or otherwise that the Division has recently undertaken as compared to past years cannot provide a reliable indication of the vigor of our merger enforcement. Were such numerical comparisons offered as evidence in a merger case, they would be laughed out of court.
Start with the fact that, in any period, at most a relatively small number of transactions out of the vast universe of all merger proposals will raise competitive concerns. Drawing conclusions from shifts in small numbers is perilous under any circumstances.(31) That is especially so when the Division has no control over what transactions are proposed to us. We look at each and every transaction the same way, applying our analytical framework to an intensive evaluation of the facts. The number of times we conclude a challenge is appropriate reflects the number of problematic transactions that are proposed, nothing more, nothing less. Correlation in these circumstances is a far cry from causation. To drive this point home, if I told you that the Nuclear Regulatory Commission did not turn down a single application to build a nuclear power plant in the United States between 1977 (just before the Three Mile Island incident) and 2007, would you conclude that regulatory policy was too lax?(32)Equally important, efforts to glean enforcement policy, or vigor, from apparent differences in the number of challenges over time would founder on other important facts. In thinking about why we might have challenged fewer mergers, it is sensible first to ask whether there is any reason to expect that fewer problematic mergers are being proposed. I submit there are two reasons to think so, having nothing to do with changing views about what makes a transaction "problematic." First, over the past several decades we (and our counterparts at the Federal Trade Commission) have done a good job of educating the business community (and their antitrust counselors) about the kinds of transactions that are likely to raise antitrust concerns. I would point not just to the Horizontal Merger Guidelines and the Commentary on those Guidelines issued in 2006,(33) but the considerable further details about how we have analyzed particular transactions that we provide in Competitive Impact Statements (when we challenge a transaction), Closing Statements (when we don't), and the many discussions of our investigations in speeches by Division officials. As a result of these efforts to foster greater transparency, parties are better equipped than ever to steer clear of transactions that would guarantee a challenge. Not surprisingly, few blatantly anticompetitive mergers are proposed, and where transactions do raise clear concerns the parties typically come forward quickly with a fix that eliminates potential competitive harm.
Second, wholly independent of parties' understanding of the antitrust constraints on their deal-making, there may be reasons to think that the mix of transactions presented for review has evolved over time to pose fewer competition concerns.(34) For example, many of the transactions we see involve companies that operate in the services and technology sectors, where it might be thought that on average transportation costs are lower and change more rapid than in some of the more traditional "industrial" sectors of the economy. With the broader economy constantly undergoing evolutionary change, there may be many reasons why one might reasonably expect to see a smaller percentage of transactions challenged even if antitrust policy remained perfectly constant.
As a result, I submit that looking at the number of transactions we challenge without controlling for the characteristics of the transactions presented for our review provides no useful insights into the vigor or effectiveness of our enforcement program. The fact is we challenge the transactions that are likely to cause competitive harm and do not challenge those that are not.
Our internal merger review process, and in particular the involvement of the "Front Office" (i.e., the Office of Operations, the pertinent Deputy Assistant Attorneys General, and ultimately the Assistant Attorney General), is designed to make our merger investigations more efficient and effective. To be sure, recommendations to challenge must go through a series of steps before the Division will file a complaint in court. But this has always been true, and it does not create any bias against challenging problematic transactions. To the contrary, we encourage our legal and economic staffs to investigate potentially anticompetitive transactions vigorously. We ask them to dig out the relevant facts and apply sound economic and legal analysis in making a determination whether the transaction is likely to cause harm to competition and consumers. When our staffs conclude that a transaction should be challenged, their recommendation is reviewed by the Front Office but that review is not designed to interpose a more stringent test than our staffs impose on themselves. Indeed, the Front Office is involved at every stage of these investigations, so that in virtually every case there is agreement about what course to pursue. When our staff's investigation demonstrates that a transaction is likely to cause substantial harm to competition and consumers, and that the Division likely could persuade a district court that it should exercise its remedial powers to prevent a violation of law, we will not hesitate to challenge the transaction.
The effectiveness and efficiency of our review processes is not always completely transparent to the outside world. Because the outcome of our merger review does not hinge on HHI calculations or any other objective or readily observable benchmark, the time and effort we will need to conduct an investigation of any particular transaction is not perfectly predictable when a transaction is proposed. But in every case we tailor the investigation to match the facts presented. The Division sometimes is able to close an investigation of transactions in markets that are or appear moderately or highly concentrated without the need for intensive document or data review, based on our own industry expertise or facts that are readily obtainable early in our investigation. In other situations the Division must seek documents or data via Second Request or CIDs before we can conclude that competitive harm is unlikely. And many times when we conclude that a Second Request should be issued, we are nevertheless able to focus our further investigation on a subset of dispositive issues and thereby conclude in short order that the transaction is unlikely to cause competitive harm.
We are quite effective at tailoring our investigations to the circumstances. One measure of our success is the frequency with which we are able to close our investigation quickly despite having concluded that a Second Request should be issued. In 60 percent of such investigations over the past year we were able to close our investigation less than 90 days following the expiration of the initial waiting period.(35) We think this kind of flexibility and efficiency is a good thing, and it certainly should not be mistaken for an unfocused churning of our merger investigations.
I can't escape the subject of the Division's merger enforcement program without commenting on Whirlpool/Maytag. There are two distinct issues here.
Let me start by observing that, even we were to assume without any factual foundation, I might add that not challenging Whirlpool/Maytag was a mistake, it alone would not demonstrate any systematic bias in the Division's enforcement decisions. At most it would demonstrate that our predictions in that one case turned out to be wrong. An error in one case might simply mean that we misjudged some critical fact, or that circumstances unfolded in an unpredictable way.(36)
The second issue is whether the Division was correct in declining to challenge the merger. On this point, the Division previously has explained the reasons for its decision, and I will not revisit them here, except to note that those reasons had nothing to do with any bias against challenging transactions that we think are likely to harm competition.(37)
It is particularly unproductive to debate the Division's decision in Whirlpool/Maytag in a factual vacuum. That decision was based on an extraordinary amount of non-public information on market conditions that informed the Division's in-depth analysis. It would be vastly oversimplistic to assume that the merger should have been challenged solely because the market for washers and dryers was highly concentrated and the merger would have significantly increased concentration. There never was any question that the transaction would increase concentration in a concentrated market, but I should not have to remind anyone that merger analysis has come quite far from the days when concentration figures alone were a good predictor of anticompetitive harm. No sensible proponent of antitrust enforcement could seriously favor a return to the days when merger analysis stopped there, and mergers were routinely blocked just because they would increase concentration in a highly concentrated market.
The case law long ago established that a structural analysis of concentration is only the starting point for merger review.(38) The Horizontal Merger Guidelines are similarly clear. (39) Our recent experience analyzing proposed mergers confirms that, while relevant and sometimes significant, a transaction's impact on concentration does not irrefutably establish its competitive merits.
The traditional market structure-performance paradigm stems in large part from the expectation that tacit (or explicit) coordination is more likely in markets that are more highly concentrated. That concern is at its zenith when products are homogeneous, pricing is transparent, and the pace and character of transactions minimizes incentives to cheat and allows swift punishment for deviations. In many of the cases we review involving markets that are arguably highly concentrated, however, conditions for coordination are far less straightforward. Products (or services) are often highly customized, pricing and other terms are correspondingly complex and confidential, and winner-take-all long-term contracts with large customers often create incentives to cheat and reduce opportunities for effective retaliation.
This is not to say we unthinkingly reject concerns about potential coordination, or assume that the kinds of factors I have outlined always make coordination unlikely. We do not. Especially where concentration is high, or where evidence indicates that coordination schemes have been attempted in the past, we consider carefully whether a proposed merger is likely to make coordination more likely. Our challenges in Mittal/Arcelor and Vulcan/Florida Rock should demonstrate that we take coordination concerns seriously.(40) For example, we ask whether all the firms in the market would need to join in coordination in order to make it profitable for the largest firms to implement a tacit agreement.(41) We explore potential means of coordination that avoid the complexities of transaction- or customer-specific pricing, such as potential mechanisms involving allocations of customers or market share. And we ask whether the parties likely could alter the way in which goods or services are transacted post-merger so as to make coordination more feasible.
Whether or not anticompetitive coordination is likely, we will also analyze potential unilateral effects. In this realm, the role of concentration is far less well developed. Some of our recent challenges based on unilateral effects concerns have involved relatively unconcentrated markets, as in Exelon/PSE&G, where the post-merger HHI was in the range of 2,200.(42) In others such as ATT/Dobson the merging firms accounted for a very high percentage of all sales in the market. Likewise, the transactions we did not challenge fell across the full spectrum of concentration: from those that had little effect on concentration in unconcentrated markets to those that may have appeared to increase it substantially in relatively concentrated markets.
In the latter cases especially it may be tempting for outside observers to perceive that a challenge is likely, and to be surprised when one is not forthcoming. In fact, however, our analysis is far more sophisticated, and our decisions are uniformly based on our in-depth assessment of the pertinent facts. Allow me to illustrate by addressing just two of the issues that have proven dispositive in recent investigations:
Are the merging firms really the first and second choices for any distinct group of customers? When coordination is unlikely and products are substantially differentiated, the most plausible theory of unilateral competitive harm typically will be that the merger would combine the first and second choices of a significant group of customers. Sometimes initial indications and even 4(c) documents give credence to such a hypothesis. One of the first steps in such an investigation will be to probe the facts bearing on this issue. Sometimes interviews of a cross-section of both firms' customer base are enough to persuade us that customers view other firms as offering options that are sufficiently close to rule out any substantial competitive harm. Sometimes we must wade into the data on bidding patterns. When those data show that customers disproportionately choose firms other than the merging parties as was the case in Blackboard/WebCT(43) that can be strong evidence that anticompetitive harm is unlikely.
Note that a critical factor in this analysis is the question whether the merged firm could feasibly identify which customers lack effective options so as to exploit the enhanced market power it would obtain vis-à-vis that group of customers without experiencing an unprofitable flight of customers that have additional options. Sometimes company documents make plain that each customer's option-set is well understood. In other cases, however, the facts drive the opposite conclusion, leading us to conclude that anticompetitive harm is unlikely because the merged firm would not risk losing important customers whose array of options they do not well understand.
Will entry or repositioning by other incumbents preclude a sustained price increase by the merged firm? When it appears likely that the merging firms are in fact the first and second choices of a substantial group of customers, the question often arises: couldn't other firms fare equally well at satisfying these customers? Within the Guidelines framework, this question raises issues of entry and expansion, or in the vernacular of firms offering differentiated products, "repositioning."
These issues are very often the salient factual questions in our merger investigations, and they can be hard ones to assess. We do not merely assume that entry or repositioning will be "timely, likely and sufficient" to prevent a significant price increase, especially where the facts show that the merged firms have sustained high shares of the market (or relevant customer group) for some time. Especially in rapidly-evolving technology markets, it is not uncommon for the merging parties to point to a number of firms that seem poised to attract a substantial number of customers if the merged firms were to raise price. Two scenarios are fairly common: (a) firms already in the market that are asserted to have plans to introduce products that will capture a significant portion of the merged firms' base of customers, and (b) firms that offer products different from those of the merging parties, but which a significant number of customers assertedly would find attractive as technology and customer demand evolves. Both of these scenarios, if borne out by the facts, potentially could prevent a proposed merger from having anticompetitive effects.
But we do not merely assume that the nascent competitors identified by the parties would in fact emerge sufficiently soon and with sufficient scale to preclude anticompetitive harm. Rather, we must investigate the facts, which often requires that we not only interview customers and the supposed competitors, and seek documents from the parties reflecting their views of future competitive threats, but also in appropriate cases issue CIDs to the competitors to probe their internal assessment of their prospects.
I have in many ways only scratched the surface of our merger review. But it should be clear that merger enforcement is alive and well at the Department of Justice. We devote substantial resources to analyzing each and every transaction presented for our review, plus many other non-reportable transactions that we identify on our own initiative. In deciding whether those transactions likely would harm competition, we are driven by the facts and sound economic analysis. We will not hesitate to challenge transactions that we conclude are likely to cause harm to competition and consumers, as a fair review of our record will confirm. Parties who underestimate our resolve do so at their peril.
1. Deborah Feinstein has similarly concluded that "reports of the complete demise of federal merger enforcement have surely been exaggerated." Deborah L. Feinstein, Recent Trends in U.S. Merger Enforcement: Down But Not Out, Antitrust, Summer 2007, 74, 80.
2. Mark Twain Quotations--Death, http://www.twainquotes.com/Death.html (last visited Nov. 14, 2007).
3. United States v. Von's Grocery Co., 384 U.S. 270, 301 (1966) (Stewart, J., dissenting).
4. See Complaint, United States v. Daily Gazette Co., No. 2:07-0329 (S.D. W. Va. May 22, 2007), available at http://www.usdoj.gov/atr/cases/f223400/223469.pdf.
5. See infra pp. 12-13.
6. See Complaint, supra note 4, para. 23.
7. See Memorandum of the United States in Opposition to Defendants' Motion to Dismiss, Daily Gazette, No. 2:07-0329 (S.D. W. Va. Oct. 5, 2007).
8. United States v. Gillette Co., 406 F. Supp. 713, 715 (D. Mass. 1975) (citation omitted).
9. Recognizing these benefits, Senator Tunney observed in 1973 that courts reviewing the Division's settlements pursuant to what became known as the Tunney Act should not "engage in extended proceedings which might have the effect of vitiating the benefits of prompt and less costly settlement through the consent decree process." 119 Cong. Rec. 24,598 (1973); see also, e.g., McDermott, Inc. v. AmClyde, 511 U.S. 202, 215 (1994) (noting that "public policy wisely encourages settlements").
10. Antitrust Div., U.S. Dep't of Justice, Antitrust Division Policy Guide to Merger Remedies 26 (2004), available at http://www.usdoj.gov/atr/ public/guidelines/205108.pdf ("A fix-it-first remedy is a structural remedy that the parties implement and the Division accepts before a merger is consummated. A fix-it-first remedy eliminates the Division's antitrust concerns and therefore the need to file a case.") (footnote omitted).
11. See Competitive Impact Statement, United States v. Cemex, S.A.B. de C.V., No. 1:07-cv-00640 (D.D.C. May 23, 2007), available at http://www.usdoj.gov/atr/cases/f223500/223543.pdf.
12. See Competitive Impact Statement, United States v. Vulcan Materials Co., No. 1:07-cv-02044-EGS (D.D.C. Nov. 13, 2007).
13. Competitive Impact Statement at 11, United States v. AT&T Inc., No. 1:07-cv-01952 (D.D.C. Oct. 30, 2007), available at http://www.usdoj.gov/atr/cases/f227300/227309.pdf.
14. See Competitive Impact Statement, United States v. Abitibi-Consolidated Inc., No. 1:07-cv-01912 (D.D.C. Oct. 23, 2007), available at http://www.usdoj.gov/atr/cases/f227100/227108.pdf. The author was recused from the Division's investigation in this matter.
15. See Competitive Impact Statement, United States v. Exelon Corp., No. 1:06-cv-01138 (D.D.C. Aug 10, 2006), available at http://www.usdoj.gov/atr/cases/f217700/217717.pdf.
16. See id. at 10; Competitive Impact Statement, supra note 14, at 6; see also Elizabeth Armington et al., The Year in Review: Economics at the Antitrust Division, 2005-2006, 29 Rev. Indus. Org. 305, 315-23 (2006).
17. See Competitive Impact Statement, United States v. Mittal Steel Co., No. 1:06-cv-01360-ESH (D.D.C. Aug. 1, 2006), available at http://www.usdoj.gov/atr/cases/f217400/217491.pdf.
18. As noted above, we also concluded that the transactions would have anticompetitive unilateral effects in several markets. See supra notes 11, 12, and accompanying text.
19. Complaint paras. 38, 43, Vulcan Materials, No. 1:07-cv-02044-EGS (D.D.C. Nov. 13, 2007).
20. See Competitive Impact Statement, United States v. Monsanto Co., No. 1:07-cv-00992-RMV (D.D.C. May 31, 2007), available at http://www.usdoj.gov/atr/cases/f223600/223682.pdf.
21. See Competitive Impact Statement, United States v. Amsted Indus., Inc., No. 1:07-cv-00710 (D.D.C. Apr. 18, 2007), available at http://www.usdoj.gov/atr/cases/ f222700/222730.pdf. The author was recused from the Division's investigation in this matter.
22. Id. at 1.
23. See id. at 10.
24. Competitive Impact Statement, supra note 13, at 9.
26. See Press Release, U.S. Dep't of Justice, Statement of the Department of Justice's Antitrust Division Regarding Its Investigation of Hearst Corporation's Proposed Acquisition of Tracking Stock in MediaNews Group Inc. (Oct. 25, 2007), available at http://www.usdoj.gov/atr/public/press_releases/2007/227168.pdf.
27. The parties' revised Shareholders Agreement covering Hearst's Class C stock in MediaNews was filed with the SEC and is available at http://www.medianewsgroup.com/ Financials/ Forms/2007/Form%208-K%20-%20Web%20Site.pdf.
28. United States v. Oracle Corp., 331 F. Supp. 2d. 2d 1098 (N.D. Cal. 2004).
29. See supra note 4 and accompanying text.
30. United States v. Dairy Farmers of Am., Inc., No. Civ. A. 6:03-206-KSF, 2007 WL 1200094 (E.D. Ky. Mar. 23, 2007). The complaint in the Dairy Farmers case was filed in 2003, but the Division was pursuing its claims to trial following a successful appeal and remand, both of which post-dated the district court's decision in Oracle.
31. See, e.g., ABA Section of Antitrust Law, Am. Bar Ass'n, Econometrics 15-16 (2005).
32. In fact no applications to build new plants were filed between 1977 and 2007. See, e.g., Companies File the First Nuclear Plant Application in 29 Years, EERE Network News (U.S. Dep't of Energy/Washington, D.C.), Sept. 26, 2007, http://www.eere.energy.gov/news/news_detail.cfm/news_id=11307.
33. U.S. Dep't of Justice & Fed. Trade Comm'n, Commentary on the Horizontal Merger Guidelines (2006), available at http://www.usdoj.gov/atr/public/guidelines/215247.htm.
34. Baker and Shapiro acknowledge that such trends could explain variations in enforcement statistics. See Jonathan B. Baker & Carl Shapiro, Reinvigorating Horizontal Merger Enforcement 13 (Oct. 2007) (unpublished working paper), available at http://faculty.haas.berkeley.edu/shapiro/mergerpolicy.pdf.
35. This calculation is based on transactions for which the initial HSR waiting period expired between June 1, 2006 and July 31, 2007, and includes both those in which a Second Request was issued and those that were "pulled and refilled" to extend the waiting period, whether or not a Second Request was ultimately issued.
36. I am indebted to Dennis Carlton for the insight that any retrospective assessment of merger policy as distinct from the correctness of any given enforcement decision must proceed by examining the universe of transactions that were investigated. As he explains, even if merger policy were exactly right, it would still be true that the government would make random errors in cases.
37. See Press Release, U.S. Dep't of Justice, Department of Justice Antitrust Division Statement on the Closing of Its Investigation of Whirlpool's Acquisition of Maytag (Mar. 29, 2006), available at http://www.usdoj.gov/atr/public/press_releases/2006/215326.pdf; see also Thomas O. Barnett, Assistant Attorney Gen., U.S. Dep't of Justice, Merger Review: A Quest for Efficiency, Remarks at the New York State Bar Association Antitrust Section Annual Meeting 10-11 (Jan. 25, 2007), available at http://www.usdoj.gov/atr/public/speeches/221173.pdf.
38. See, e.g., United States v. Gen. Dynamics Corp., 415 U.S. 486, 501-04 (1974).
39. U.S. Dep't of Justice & Federal Trade Comm'n, Horizontal Merger Guidelines § 1.52 (1997), available at http://www.usdoj.gov/atr/public/guidelines/hmg.pdf.
40. See supra notes 17-19 and accompanying text.
41. Such an analysis can entail a critical loss assessment, which asks whether the coordinating firms would lose sufficient volume to the non-colluding fringe in order to render the coordination unprofitable.
42. See Competitive Impact Statement, supra note 15, at 10; Barnett, supra note 37, at 11; David L. Meyer, Section 2 Standards and Consumer Welfare: Some Lessons from the World of Merger Enforcement, 2007 Colum. Bus. L. Rev. 371, 397.
43. Armington et al., supra note 16, at 307; Meyer, supra note 42, at 396-97.