This document is available in two formats: this web page (for browsing content), and PDF (comparable to original document formatting). To view the PDF you will need Acrobat Reader, which may be downloaded from the Adobe site. For an official signed copy, please contact the Antitrust Documents Group.

U.S. Department of Justice Seal DEPARTMENT OF JUSTICE

Toward An International Dialogue on the Institutional Side of Antitrust

Deputy Assistant Attorney General
Antitrust Division
U.S. Department of Justice

Remarks as Prepared for
NYU Annual Survey of American Law Conference

New York, New York

February 19, 2010

For the last twenty years, discussions on international antitrust convergence and cooperation have focused on potential divergences in substantive doctrine.  Those discussions have taken place at important multilateral forums (notably, the Organization for Economic Cooperation and Development and the International Competition Network), have covered an array of different substantive doctrines, have substantially bridged the gaps between different jurisdictions, and have promoted greater levels of understanding.  During this era, leaders of different antitrust jurisdictions have also engaged in important bilateral discussions, cooperated on individual matters, and joined academic discussions around the world, all of which has helped to move the antitrust world toward a common analytical framework.  When reflecting on the progress made on this front, it is nothing short of remarkable.

To appreciate the paradigm change in the globalized world of antitrust, consider the case of the U.S. Horizontal Merger Guidelines.  When Assistant Attorney General Jim Rill and Federal Trade Commission Chair Janet Steiger oversaw the revision of the merger guidelines in the early 1990s, they spent little time thinking about how this enterprise related to jurisdictions outside the United States.  Today, that could hardly be less true.  In our ongoing effort to reevaluate the 1992 Horizontal Merger Guidelines, one of our first thoughts was to ask how this effort related to, could be informed by, and would affect other jurisdictions around the world.  To that end, we are committed to learning from more recent updates (including a current exercise in the UK and a 2004 revision by the European Commission).  We are also committed to hearing from developing competition authorities insofar as they often look to our merger guidelines for guidance.  To the extent that our guidelines are less than clear, feedback from developing authorities will help us evaluate how our guidelines can be made more transparent, more practicable, and easier to understand

The topic I want to discuss today is what as I see as an under-explored dimension of antitrust law and policy:  the institutional side of the equation.  To my mind, the dialogue that competition authorities around the world need to have over the next twenty years needs to continue addressing not only substantive doctrinal issues, but also ones of institutional design, administrative practice, and legal culture.  To that end, we all need to learn from one another’s experience, aspire to improve the institutional effectiveness of our competition regimes, and provide effective counsel to developing competition authorities that have the opportunity to learn from our experiences—both what traditions serve us well and what mistakes we have made along the way.

In my talk today, I will begin by discussing the perspective of new institutional economics on why institutional design matters.  After so doing, I will reflect on some of the challenges of perfecting a commitment to transparency and procedural fairness.  Finally, I will explain the Department’s approach to technical cooperation, drawing on some of the points discussed below with respect to procedural fairness.

I. Why Institutional Design Matters

The field of new institutional economics is not, at least in its “new” incarnation, much older than the concept of international antitrust.  One of the leaders of this movement, Nobel Prize-winner Oliver Williamson, is a great friend to the Antitrust Division, having served in the Division in the position roughly akin to our current Deputy Assistant Attorney General for Economics.(1)  As an academic, Williamson’s work has focused largely on how private institutions develop to address marketplace dilemmas—such as how two companies that must deal with one another (say, a power generation company and an electric power distribution company) can develop mechanisms for minimizing transactions costs in negotiating bilateral contracts.  In thinking through this sort of challenge, Williamson has offered an array of observations, insights, and suggestions, including the colorful metaphor that firms that relate to one another in this fashion (which is sometimes called a “bilateral monopoly”) can use the approach of “trading hostages” to ensure that the two firms deal fairly with one another.(2)  The teachings of new institutional economics also provides a helpful framework for evaluating regulatory responses that have developed to oversee such relationships, including ones between local governments and cable TV providers, independent power producers and electric utilities, and franchisors and their franchisees.

The institutional design of public institutions is often relegated to a second order consideration and takes a backseat to the analysis of substantive policy issues.  In the case of telecommunications regulation, for example, U.S. discussions around whether to mandate some form of “network neutrality regulation” have consumed far more energy than the relevant institutional questions about how any such regime would be managed.  In particular, few commentators or policymakers have focused on the question of what institutional strategy, structure, or set of processes—co-regulation, self-regulation, command-and-control regulation, or adjudication—to use and how any such regime would operate in practice.(3)  To be fair, I should note that this form of analysis is more prevalent in Europe than the U.S.(4) 

From my standpoint, the impact of institutional issues as an influence on the ultimate success of an agency is grossly underappreciated.<(5)  To be sure, particular positions on substantive policy issues are often more accessible and their analysis often promises higher short term payoffs than evaluating improvements in institutional structure and process.  In the long term, however, the investment in institutional effectiveness and improved processes will pay great dividends.  As FTC Commissioner Bill Kovacic put it (quoting Fred Hilmer), “[w]hat are you doing today to make sure that your successors will prosper five or ten years hence?”(6)  Consequently, any focus on agency effectiveness needs to ask how an agency is doing its work and not merely what work it purports to be doing.  Indeed, even the best-crafted statutory or regulatory regime will fail if the institutional structure, processes, and culture undermine the ability to implement the regime’s goals effectively.

Going forward, competition authorities and academics around the world have an opportunity to engage in a discussion focused on comparative institutional competence and comparative institutional practice.  To that end, the emerging traditions among different authorities and different procedural mechanisms provide an opportunity for benchmarking and a challenge in terms of thinking about institutional change.  Specifically, we have the opportunity to evaluate alternative practices and processes, the impact of institutional structure on culture and habits of mind, and how different institutional strategies can lead to different substantive outcomes.  The challenge is that, even when it is clear that there are better institutional approaches out there, institutional change is hard. 

II. Institutional Design and Transparency

For purpose of my talk today, I’d like to focus on one critical dimension of institutional design and administrative procedure—transparency.  By transparency, I mean making an agency’s operations visible to the public.  The how, of course, is multidimensional—the substantive standards employed by the relevant agency, the particular procedures it uses, and its willingness to talk openly about the legal and economic theories as well as underlying factual concerns in particular cases.  Let me discuss each of these points in turn.

  1. Transparency About Substantive Standards

In the U.S., antitrust law is a common law enterprise.  It is thus incumbent upon antitrust enforcers and the courts to give meaning to the broad directives of the Sherman and Clayton Acts.  Judging which mergers may “substantially lessen competition or tend toward monopoly” is not, for example, a task free from ambiguity.  Indeed, over the course of the last fifty years, courts have differed on how to interpret this standard from the Clayton Act.(7)  More recently, with our adoption of a pre-merger notification regime under the Hart Scott Rodino Act in 1976,(8) litigation over merger cases has become less frequent and the Supreme Court has essentially provided no guidance on the meaning of the Clayton Act during that time.

When judicial guidance is less forthcoming, the need for transparency as to the standards used by the antitrust agencies becomes even more important.  To that end, the Horizontal Merger Guidelines have provided an important means of explaining the substantive standards used by the antitrust agencies.  Both the 1984 version and the 1992 revision (which was adopted by the DOJ and the FTC) provide considerable insight into the analytical framework used by the agencies. 

Over the last 18 years, experience with the guidelines and new economic learning have arguably nudged the agencies away from the standards set forth in the 1992 guidelines.  To that end, in 2006 the agencies released commentaries on the Merger Guidelines, making clear that certain of the suggested approaches—such as the stepwise analysis of market definition, measurement of market shares, evaluation of entry, efficiencies, and other possible defenses—do not necessarily operate in practice along the lines specified by the guidelines.(9)  Similarly, the figure of an HHI of 1800 as indicative of the danger zone for mergers is one that many suggest is inaccurate as a matter of agency practice or judicial guidance.  To evaluate whether such areas are not well stated by the guidelines, the DOJ and FTC have embarked upon an effort to evaluate whether the guidelines should be updated and, if so, how.(10)

In deciding to review the guidelines, one essential rationale for our decision was the possibility of promoting a greater degree of transparency.  In particular, we are interested in ensuring that the guidelines actually reflect the realities of agency practice.  Some have suggested, however, that this is not important and that we should abstain from any alterations because, even if current practice differs from that set forth in the guidelines, “everyone” knows how the agencies really operate and any changes to the guidelines could be disruptive.  Our skepticism about that argument relates to the suggestion that “everyone” knows about the potential for a divergence of the stated standards and actual practice.  Both as the international deputy and as someone who recently came back to Washington D.C. from Colorado, I am personally confident that not everyone knows that the guidelines may be misleading.  Notably, countries around the world look to the guidelines for guidance and presume that they reflect the state of the art, not merely a fixed point in early 1990s American antitrust practice that time has left behind.  Similarly, practitioners not among the antitrust cognoscenti are very likely to assume that the guidelines are what they say they are.  Thus, to the extent that an update enables all sorts of communities that consult the guidelines to recognize areas where they have been misleading, I view that as a very healthy result.

I should add that the guidelines are only a first step for communicating our merger review standards in a transparent fashion.  In particular, we provide valuable guidance through our regular practice of issuing competitive impact statements as part of any settlement and, taking a page from the European Commission, we issue closing statements on increasing occasions.  And when we take a case to court, we must outline our core rationale, provide evidence to back up our theory of the case, and test our views through the crucible of a contested proceeding.

  1. Transparency About Process and Best Practices

One concern about being more transparent about agency process is that it restricts the options available to the agency.  Consider, for example, whether to provide parties with the right to confront the decision-maker and the right to offer a settlement before final decision.  On the one hand, making such a commitment might lead to a slower process in certain cases where the agency has reason to believe that this step is not time well spent.  On the other hand, pre-committing to such a tradition can provide an important safeguard to facilitate fair treatment of the affected parties and to help to ensure that the decision-maker is confronted with all relevant facts before making a judgment with important legal consequences.

At the United States Department of Justice, the commitment to all parties that they will have an opportunity to meet with the Assistant Attorney General before she decides to file an action in court is viewed as a quasi-religious tradition.  Such meetings, which I have attended during my earlier and current time at the Department, are taken very seriously, prepared for with care, and almost universally valuable.  It is thus easy for me to understand why the ICN has adopted a similar measure as a best practice.(11)  To my mind, the rationale for such meetings is equally powerful in the civil non-merger context as well.(12) 

Our commitment to providing parties with a chance to meet before filing suit is valuable not merely because it allows for an informal give-and-take between the Assistant Attorney General and the parties, but also because it requires the staff and the AAG to think carefully about all relevant issues before bringing a case.  Notably, if a key form of evidence (say, data showing a price increase) or a theory of competitive harm (say, raising rivals’ costs) underlies the case, the AAG will be in a position to evaluate that information closely and observe for herself the parties’ response to the issue.  Indeed, one “second order” effect of the commitment to providing parties with such a meeting is that the Department will sometimes use “red team-blue team” mock arguments that test the arguments made both by the Department and the opposing party.  This exercise is further internalized by the practice of ensuring that all documents developed by the Department reflect an awareness of the perspectives of the opposing party and an intellectually honest response to them.

A final reason why the tradition of an AAG meeting with the parties is important is because it enshrines the notion that the AAG decision-making process is separate and apart from that of the Department staff.  Notably, the staff traditionally develops its own recommendation, which is communicated to the parties and the AAG in advance of a meeting between the parties and the AAGIn this sense, the AAG (and her “front office”) constitutes a separate layer of review—a fresh set of eyes, as it were—that can evaluate the merits of a case (or a proposed settlement) independent of the judgment of those who develop the case.  Indeed, the AAG, both from career staff—both lawyers and economists—as well as front office personnel, benefits from an array of perspectives and often receives differing recommendations on how to proceed in a given case.  This structure provides a healthy second look at a case, brings a valuable (and broader) perspective to the evaluation of the merits of individual cases, and is a good example of the new institutional economics lesson that “organizations can be structured to optimize the benefits and costs of expert decision-making.”(13)  It also, like judicial oversight, ensures the evaluation of the evidence in a detached and more critical fashion.  After all, the lawyers who develop the case may well assume the posture of an advocate for a particular position whereas a more detached person—i.e., one who has invested far less time in it—can evaluate the evidence in a manner closer to that of a judge weighing the evidence in an impartial manner.

To my mind, the effort to narrow the relevant issues of disagreement by the parties subject to a Justice Department investigation and the investigating staff attorneys is one of the healthiest byproducts of transparency.  In requesting documents, for example, an openness about the theories of harm provides a basis for identifying what sort of limitations are appropriate.  By contrast, an unwillingness or inability to communicate meaningfully on such issues leaves the investigated parties guessing about the Department’s areas of competitive concern and thus less able to be responsive, both in providing documents and in developing a “white paper” addressing those concerns.  In my experience, discussions based on an open exchange of ideas are far more effective than ones where parties are unable to understand the relevant issues and develop an articulated basis for their concerns.(14)

One final area where procedural transparency can be quite important is the openness to discussing a possible settlement.  Parties are free to suggest potential settlements with us at appropriate points in our process and, given our tradition of being open about the relevant theories of harm, parties should be in a position where they can be responsive to our competitive concerns.  By contrast, to the extent that one “hides the ball”—for strategic purposes or to gain an advantage in litigation—that makes it far more difficult to reach an effective settlement.  In our experience, it is important to be open to considering settlements because, where firms are willing to settle competitive concerns, addressing such issues cooperatively can solve problems more quickly and enable us to focus our attention on issues that are not so easily resolved.

  1. Open Engagement on the Relevant Facts as well as the Legal and Economic Theories

I am well aware that many countries around the world rely on administrative systems that enable the administrative agency to impose a liability and order a remedy without trying a case to an independent judge.  This model has its relative virtues and vices vis a vis the judicial-centric system that we use and it is not my intention here to endorse one model or another one.  I do think, however, that certain features can be adopted even in an administrative context to promote greater transparency and better outcomes.  The principles discussed above—meetings with the parties, the use of red team-blue team exercises, openness to white papers that respond to articulated concerns, openness to discussing settlement, issuance of closing statements and competitive impact statements—can all be incorporated into either an administrative or judicial institutional structure.  And, in our experience, these steps all improve the efficiency and effectiveness of how a competition agency can operate.

It bears mention that the opportunity for engagement around issues of competitive concern is becoming easier on account of advances in communications technology.  Redmond, Washington is quite a distance from Brussels, for example, but the recent discussions in the European Commission’s recent “Browser” investigation of Microsoft followed just the sort of model outlined above.  As Microsoft General Counsel Brad Smith related:

And so we’ve had extensive discussions really over the last couple of years. We’ve been able to use those discussions to understand better what the Commission’s objectives and concerns have been.  We’ve been able to work to clarify the issues.  We’ve been able to work creatively to take additional steps. We’ve spent a lot of time talking with each other.  Just over the last four or five months, we had almost 20 videoconferences between Redmond and Brussels.  So, in effect we’ve spent a lot of time in what has felt like the same room, even if it wasn’t always the same room on a literal basis.(15)

It is also worth noting that this form of open and ongoing engagement—which is only possible on account of a transparent process—produced what both Microsoft and the European Commission hailed as a productive resolution.  As Smith explained, “I think out of that we’ve been able to reach the point where we are today, where we have something that I think works for the industry, it works for competition law, and we think we can apply it in a way that our engineers can implement.”(16)  Similarly, the European Commission welcomed Microsoft’s proposal, explaining that “it has the potential to give European consumers real choice over how they access and use the internet.”(17)  For our part, we welcomed the settlement, “commend[ing] the efforts of the European Commission and Microsoft Corporation, which have announced that they have reached a comprehensive settlement resolving their disputes under European competition law.”(18)

One important point adheres in all the principles outlined above:  parties should be invited to present their perspective on facts and the law at all parts of the process.  But this openness to dialogue only works in an environment of mutual respect and decorum.  Just as enforcers should not keep parties guessing about what their concerns are, private parties should not engage in disrespectful or obstructive advocacy.  Private parties that engage in bullying tactics and are not interested in an honest interchange of ideas chill the environment for reasonable discussion.  Thankfully, I have found such experiences rare during my time at the Department, but I acknowledge that such experiences are not unheard of.

  1. Procedural Fairness, Agency Effectiveness, and the Steps Going Forward

In short, it is our sincere belief that, by identifying the relevant theories of harm as early as possible in the process and communicating them in a straightforward fashion, antitrust agencies can be both more effective and more efficient.  As AAG Varney put it, “open and frequent dialogue between competition law enforcers and those under investigation not only helps ensure fairness to the parties but also facilitates more effective enforcement.”(19)  In light of this tradition, OECD Working Party 3 is now focused on a set of issues that fall under the heading of procedural fairness.  More broadly, we believe that this topic is one that fits well with the ICN’s focus on practical issues.  Over last few years, the ICN has begun work on institutional issues of this nature.  In 2009, the EC hosted a meeting among heads of competition agencies, focusing on sharing experiences on the practical issues involved in running an agency, such as strategic planning, prioritization, and communications strategies.  To advance our collective thinking on the valuable ideas that came out of the conference, the ICN has established a new working group on agency effectiveness.

We believe that these many forums provide the international community with an important opportunity to focus on procedural fairness.  As is well recognized, these efforts are important because procedural fairness promotes the kind of “transparent and predictable business environment” that attracts international investment and entrepreneurial activity.(20)  It is also, of course, the very thing that enables lawyers to engage in effective counseling and businesses to effectively self-regulate based on the anticipated concerns of antitrust authorities.  In a more opaque environment, by contrast, such private compliance with public goals is much harder to accomplish.

III. Technical Cooperation Agenda

Recognizing the importance of institution building and transparency underscores that work with emerging antitrust authorities is time well spent.  To that end, the DOJ and FTC recently held a workshop and authored a report that re-evaluated their approach to what historically was called “technical assistance,” developing some important guiding principles that I would like to discuss today.(21)

Two foundational principles bear particular mention.  First, engagement with emerging antitrust authorities is not merely an effort to help the host country.  Rather, as our use of the term “technical cooperation” underscores, the interaction in technical assistance is a two-way street.  Stated simply, a focus on best practices, basic investigative techniques, institutional strategies, and substantive principles requires us to look carefully at how we are operating in practice.  Moreover, the time spent with emerging antitrust authorities paves the way for continued cooperation after the formal technical assistance program has ended.  Second, effective technical assistance efforts cannot be “one-off” teaching efforts or ad hoc cooperation—they must be part of a long-term relationship.  To that end, our recent report suggested that such relationships should last as long as ten years, enabling the providers of support to learn about the relevant local conditions and establish a trusted working relationship.

To provide a greater level of strategic focus and direction for our technical cooperation efforts, the DOJ and FTC called for the development of memoranda of understanding (MOUs) with our foreign antitrust partners as a means of framing the nature and extent of that cooperation  In particular, such MOUs would, as our joint report put it, “establish a framework for the provision of technical assistance, aim to facilitate informal consultations on cases and policy matters, and include a commitment to hold periodic meetings among policy-level officials.”(22)  By putting the MOU into writing, the agencies can capture both their expectations for the cooperation and their commitment to it—as opposed to an ad hoc approach.  In principle, such MOUs should “facilitate ongoing communication between the agencies outside the context of a particular training event and encourage relationship-building at the staff level,” aim “to ensure that expectations are being met,” and, quite significantly, “stress the importance of developing the capacity of supporting institutions (e.g., universities, bar associations, and the judiciary).”  I am very glad to report that such an aspiration is not merely theoretical, as we have recently adopted such an arrangement with the Russian Federal Anti-Monopoly Service.(23) 

IV. Conclusion

The antitrust world is increasingly globalized and interconnected.  The need to promote convergence on substantive doctrines has received and will continue to receive considerable attention.  As that discussion goes forward, however, it is increasingly important that it be joined by a focus on institutional design and practice, particularly as to the promotion of transparency in the conduct of antitrust investigations.  In that respect, all our agencies can improve our operations.  Indeed, one of the healthy aspects of a multijurisdictional world is that sister agencies can challenge one another and model means of improving our institutional practices.  In our case, we will do our best to encourage this discussion and to improve our adherence to best practices in this important area.


1. See, e.g. Oliver E. Williamson, Vertical Merger Guidelines: Interpreting the 1982 Reforms, 71 Cal. L. Rev. 604 (1983); Oliver E. Williamson, Why Law, Economics, and Organization?, 1 Annual Rev. of
L. & Soc. Sci. 369 (2005).

2. See, e.g., Oliver E. Williamson, Credible Commitments: Using Hostages to Support Exchange, 73 Am. Econ. Rev. 519, 519-20 (1983).

3. For exceptions to this point see Philip J. Weiser, The Future of Internet Regulation, 43 U.C. Davis L. Rev. 529 (2009).

4. See Ofcom, Identifying Appropriate regulatory solutions: Principles for Analysing Self- and Co-regulation § 2.14 (2008), available at statement/statement.pdf; Ofcom, Initial Assessments of When to Adopt Self- or Co-Regulation
§ 2.17 (2008), available at

5. See William E. Kovacic, Rating the Competition Agencies: What Constitutes Good Performance, , 907 (2009) (“[G]ood agency process includes the establishment of effective internal quality control mechanisms, the adoption of transparency and accountability tools to increase public understanding of its activities, and a commitment to seek continuing improvements in its operations and in its substantive programs.”); see also Philip J. Weiser, Institutional Design, FCC Reform, and the Hidden Side of the Administrative State, 61 Admin. L. Rev. 675 (2009) (discussing importance of institutional structure, processes, and culture).

6. Kovacic, Rating the Competition Agencies, supra note 5, at 906.

7. Compare United States v. Von's Grocery Co., 384 U.S. 270 (1966) (holding that a merger between the third and sixth largest supermarkets in a geographic area was illegal) with United States v. Baker Hughes Inc., 908 F.2d 981, 990-92 (D.C. Cir. 1990) (Thomas, J.) (allowing merger and casting doubt on continued vitality of Von’s Grocery).

8. Hart-Scott-Rodino Antitrust Improvements Act of 1976, Pub. L. No. 94-435, 90 Stat. 1383 (1976).

9. U.S. Department of Justice & Federal Trade Commission, Commentary on the Horizontal Merger Guidelines (2006) available at; see also Carl Shapiro, Deputy Assistant Attorney General, Updating the Merger Guidelines: Issues for the Upcoming Workshops, Fall Forum, Antitrust Section, American Bar Association (Nov. 12, 2009), available at

10. U.S. Department of Justice, Dep’t of Justice and FTC To Hold Workshops Concerning Horizontal Merger Guidelines (Sept. 22, 2009); see also Christine A. Varney, Assistant Attorney General, Merger Guidelines Workshops, Third Annual Georgetown Law Global Antitrust Enforcement Symposium
(Sept. 22, 2009), available at; Carl Shapiro, Deputy Assistant Attorney General, Updating the Merger Guidelines: Issues for the Upcoming Workshops, Fall Forum, Antitrust Section, American Bar Association (Nov. 12, 2009), available at

11. See Int’l Competition Network, Recommended Practices for Merger Notification Procedures 19 (last visited Nov. 19, 2009), available at media/archive0611/mnprecpractices.pdf (“Prior to a final adverse enforcement decision on the merits, merging parties should be provided with sufficient and timely information on the facts and the competitive concerns that form the basis for the proposed adverse decision and should have a meaningful opportunity to respond to such concerns.”).

12. I should add here that the procedural issues differ in the criminal area for good reasons and my remarks today do not address that context.

13. Jeffrey J. Rachlinski & Cynthia R. Farina, Cognitive Psychology and Optimal Government Design, 87 Cornell L. Rev. 549, 561 (2002).

14. For a discussion of the failings of a model that devalues transparency and proceeds in an ad hoc fashion, see my criticism of the Federal Communications Commission in Weiser, supranote 5.

15. Brad Smith, General Counsel, Microsoft Corp., Microsoft on the European Commission Market Testing Announcement (Oct. 7, 2009) available at

16. Id.

17. Press Release, European Commission, Commission Market Tests Microsoft’s Proposal To Ensure Consumer Choice of Web Browsers, Welcomes Further Improvements in Field of Interoperability (Oct. 7, 2009) available at

18. Press Release, Assistant Attorney General for Antitrust, Christine Varney, Statement on European Commission Microsoft Settlement (December 16, 2009), available at

19. Christine A. Varney, Assistant Attorney General, Procedural Fairness, 13th Annual Competition Conference of the International Bar Association (Sept. 12, 2009), available at public/speeches/249974.htm.

20. International Competition Network, Competition Policy Implementation Working Group Subgroup 1, Agency Effectiveness Project 11 (2008), available at

21. U.S. Department of Justice & Federal Trade Commission, Charting the Future Course of International Technical Assistance (2009) available at

22. Id.

23. Memorandum of Understanding on Antitrust Cooperation, U.S-Russ. (Nov. 10, 2009), available at