Washington University Law Review Volume 83 | Issue 6 2005 Remarks by Professor Jody Freeman to Japanese American Law Society Jody Freeman Follow this and additional works at: http://digitalcommons.law.wustl.edu/lawreview Part of the Administrative Law Commons, and the Comparative and Foreign Law Commons Recommended Citation Jody Freeman, Remarks by Professor Jody Freeman to Japanese American Law Society, 83 Wash. U. L. Q. 1859 (2005). Available at: http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5 This work is brought to you free of charge by the Wash U Law Repository. To explore the repository, click here (http://digitalcommons.law.wustl.edu/). For more information, contact repository@wulaw.wustl.edu REMARKS BY PROFESSOR JODY FREEMAN* TO JAPANESE AMERICAN LAW SOCIETY September 12, 2004 Tokyo First, let me thank you very much for inviting me to participate in this symposium along with my esteemed colleagues Professors Levin and Lubbers. In particular, I must thank Professor Takehisa Nakagawa, our host, and Professor Setuo Hiyama, who first contacted me to invite me on Professor Nakagawa’s behalf. They have both worked very hard in arranging this trip and in making us feel so welcome. I met Professor Hiyama many years ago when he was visiting the United States. We became friends and have kept in touch, and I am delighted to see him again. And I am very pleased to get to know Professor Nakagawa, whose work on Japanese administrative law I have read with admiration. I would like today to talk about some theoretical and practical developments in American administrative law, specifically the proposed move toward a system of “collaborative governance” in which public agencies and private parties participate jointly and cooperatively in the regulatory and administrative process. I should note, as Professor Levin talked about law reform and the role of the American Bar Association (ABA) and the Administrative Conference of the United States (ACUS), that much of collaborative governance evolved from the bottom up—from agencies themselves, or stakeholders, or academics working through ACUS who were interested in introducing negotiation and dispute resolution into agency processes. ACUS, working with the ABA, then turned some of these collaborative experiments into law, but others are still not legislatively authorized. Most theories of administrative law in the United States seek to provide an account of why administrative agencies are legitimate in our constitutional democracy, because within the American constitutional order, their place is precarious. Let me explain. There is nothing in the American Constitution that specifically establishes a separate administrative state, certainly not one as elaborate and powerful as we now * Professor of Law, Harvard University. I am indebted to Professors Takehisa Nakagawa and Setuo Hiyama for inviting me to speak to the Japanese American Law Society, and to Professors Levin and Lubbers for being such delightful co-panelists. I am grateful to Taimie Bryant for helpful advice. These published remarks are a lightly edited version of the remarks I delivered in Tokyo. All errors are mine. 1859 Wash U Law Repository 1860 WASHINGTON UNIVERSITY LAW QUARTERLY [VOL. 83:1859 have, reaching into every sector of the economy and every facet of daily life. As many of you know, our Constitution divides power among three branches of government in an effort to provide checks and balances against concentrated power and also in an effort to prevent hasty action.1 The Constitution provides that Congress makes the law, that the executive implements that laws, and that the courts interpret the laws.2 Executive power in the United States, that is, the power to implement the laws, is given by the Constitution to a unitary executive, meaning executive power resides in a single president. Yet administrative agencies, both so-called “independent” ones that are somewhat insulated from direct presidential control, and so-called “executive” ones, which directly answer to the President (and which must, therefore, perform the tricky balancing act of executing the President’s policies while also faithfully implementing statutes passed by Congress and delegated to the agencies for implementation), have flourished in the United States. And these agencies frequently integrate the very functions of rulemaking, implementation and adjudication (law-making, law execution and law interpretation) that are separated out in the Constitution to provide checks and balances. Agencies are not answerable directly to the electorate, because they are headed by political appointees, and yet they wield enormous power. This is the fundamental problem of American administrative law. My impression is that things are quite different in Japan, where the administrative state’s legitimacy and constitutional foundation are not questioned in this way. For a long time, administrative law scholars have wrestled with this problem of legitimacy and debated ways in which administrative power might be curbed and constrained. Not surprisingly for lawyers and legal scholars, judicial review has emerged as the crucial instrument for constraining administrative power. As a result, administrative law in the United States is largely devoted to studying the doctrines of judicial review of agency action and the theories of deference under which courts defer to agency judgments and interpretations. Yet the theory of collaborative governance that has emerged in recent years, and of which I am a proponent, is not so focused on courts, nor even on the fundamental problem of the administrative state’s legitimacy. It assumes that agencies will be constrained by judicial review, as well as by congressional oversight (which is very active in the United States) and 1. See U.S. CONST. arts. I, II, III. 2. Id. http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5 2005] LESSONS FOR JAPAN 1861 executive management through the Office of the President (principally, but not exclusively the office of Management and Budget). The problem from the perspective of collaborative governance is not excessive administrative discretion but poor strategies for good policy-making. As the system works now, regulation is not especially well-informed, not very timely, not very good at adapting to new information or changed conditions, not very comprehensive or long-term in its perspective, and highly adversarial. Regulation is perceived as an “Us versus Them” enterprise in which there are winners and losers. The government is pitted against regulated entities that tend to resist regulation and a variety of interest groups with their own agendas join most disputes on one side or the other. The regulatory process is widely viewed as a zero-sum game in which everyone guards their information jealously and in which positions are staked out and defended. Collaborative governance, at least as a theoretical ideal, is focused on reducing the conflict, time and expense of the regulatory process when possible, but even more importantly on generating sound, implementable, well-informed, and adaptable policy, with as much ongoing participation among affected parties as possible. The best known example of this in the United States is regulatory negotiation, a process by which an agency with rulemaking authority convenes a group of stakeholders, and other parties with an interest in the rule, for the purpose of negotiating its content. This is in contrast to a traditional rulemaking process called, “notice and comment” rulemaking, where the agency proposes a rule it wishes to promulgate; all interested parties file comments; the agency considers those comments and builds a record; and the agency then publishes the final rule which is frequently challenged in court. I realize that rulemaking is not as common in Japan but for many if not most American agencies, it is the dominant method of regulation. Many things that in Japan would be informally decided would go through the notice and comment process in the United States. Typically, prior to the agency’s Notice of Proposed Rulemaking, interested parties informally contact the agency to argue for their interests. Then, during the comment period itself, the parties that are the object of the regulation file voluminous amounts of material opposing the rule. Only a small percentage of this material tends to be constructive and helpful to the regulatory enterprise.3 Much of it is defensive or unnecessary or 3. See Jody Freeman, Collaborative Governance in the Administrative State, 45 UCLA L. REV. 1, 12–13 (1997) [hereinafter Freeman, Collaborative Governance]. Wash U Law Repository 1862 WASHINGTON UNIVERSITY LAW QUARTERLY [VOL. 83:1859 designed to slow and overwhelm the agency. Similarly, interest groups that support the regulation file voluminous material in favor of it, usually pressing the agency to go further than it is willing to go. But the parties do not engage directly with each other, and the process is not well designed to generate information that might actually improve the rule or result in a viable alternative. Instead, the purpose of comment is often to delay and obstruct the agency’s progress. The parties outside the agency have no responsibility for, or “ownership” of, the regulatory process.4 A negotiated rulemaking, by contrast, incorporates the parties into the rulemaking process on the front end.5 Typically, a negotiation will include not only the directly affected regulated industry but also, for example, representatives from consumer, labor, and environmental groups whose members might be affected by the outcome and who would be in a position to challenge the rule in litigation. In successful regulatory negotiations, the agency establishes parameters, the parties negotiate and reach consensus, and all participants, including the agency, sign an agreement not to litigate if the agency implements the agreed upon rule. At its most ambitious, this collaborative approach seeks to reform not only rulemaking itself but also implementation and enforcement. Though it has not become part of the typical negotiated rulemaking process, one could imagine an approach in which the same parties that negotiated the rule would, as a group, continue to oversee its implementation, addressing difficulties as they arise.6 As we all know issues arise during the implementation of a policy that no one anticipated. It could be useful, therefore, if the negotiating committee were ongoing, so that it could serve as a forum for resolving these matters. In theory, trust ought to improve regulatory relationships. Once players have become familiar with each other and have built trust, implementation and enforcement should operate more smoothly. We do not have good data on how regulatory negotiations affect implementation and enforcement, so the positive potential of negotiated rulemaking remains somewhat speculative. But as an approach, it represents a departure from the traditional adversarial process. Of course even when it uses negotiated rulemaking, the government still retains the right to take a more aggressive approach in the end. Unlike in Japan, many American agencies have their own enforcement power, and can impose administrative penalties directly. But in a collaborative system, agencies would first ask how they might 4. Id. 5. Id. at 34. 6. See id. at 72. http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5 2005] LESSONS FOR JAPAN 1863 facilitate and encourage greater compliance. Instead of assuming an oppositional stance from the outset, unilateral enforcement would be a last resort.7 Let me offer another example of a collaborative experiment besides regulatory negotiation, this time in the implementation process. The example is drawn from environmental law, a field in which collaborative experiments have flourished.8 Typically, firms that pollute the environment must apply for and obtain permits issued by the environmental agencies responsible for implementing environmental laws. These permits contain limits on, among other tings, the amount of air and water pollution the firms can emit. Usually the limits are performance standards that require firms to install pollution control technology. Traditionally, applying for a permit is a highly adversarial process in which the applicant submits documents and information, and the agency unilaterally decides the terms of the permit, after interpreting the relevant regulations. Typically, there is a back and forth between the applicant and the agency, but it is not a “collaborative” process in which they share information voluntarily and assist each other as if they had a common goal. Traditional permitting can be very costly and time consuming. For example, in a typical case, every process change within a firm that increases pollution would trigger a separate and lengthy permit application. For industries that make process changes frequently, like the computer chip industry, applying for a permit for every change can slow the firm’s ability to get new products to market (in this case computer chips, which turn over, I believe, something like every 18 months).9 In the 1990s, the Environmental Protection Agency adopted a program called “Project XL,” which was an experimental approach to issuing permits using a more collaborative model. Under Project XL, an agency negotiates permits with individual firms or industry sectors and includes other stakeholders (like local community and environmental groups) in the process as well.10 The goal of the negotiation is to ensure that the firm achieves greater environmental protection than it would achieved under 7. Id. 8. For an example of such a collaborative experiment, see id. at 57–61 (describing the Berry Corporation’s experience). 9. See id. at 61–66 (describing Intel Corporation’s experience with collaborative environmental regulation). 10. “XL” stands for “Excellence in Leadership.” See Regulatory Reinvention (XL) Pilot Projects, 60 Fed. Reg. 27,282 (May 23, 1995); see also Freeman, Collaborative Governance, supra note 3, at 55. Wash U Law Repository 1864 WASHINGTON UNIVERSITY LAW QUARTERLY [VOL. 83:1859 the traditional regulations and the normal permit process. The idea behind XL was that if the agency shows some regulatory flexibility the firm will produce better environmental performance than it would otherwise be legally obligated to achieve. Using this more collaborative approach, the bilateral nature of traditional permitting process converts into a multilateral process involving more stakeholders.11 In a typical XL project, the EPA might grant pre-approval of process changes in a single streamlined permit obviating the need for multiple permits for each change. The agency might authorize the firm to do things they otherwise could not do under the regulations. For example, the firm might be allowed to trade reductions of one kind of pollutant for increases in another, or reduce pollution in one medium (say water pollution) in exchange for increases in another medium (say air pollution).12 The agency is prepared to grant this flexibility only where staff believe there is a net benefit to the environment. Interestingly, XL negotiations do not produce legally enforceable agreements, but instead culminate in a detailed, written document known as a “Final Project Agreement” in which the agency and the firm, together with other stakeholders, make mutual commitments about what they will do.13 The permit looks more like a contract. Let me offer a third and final example. A collaborative approach can also be useful in managing natural resources like watersheds, forests and public lands, where conflicts frequently arise because a variety of stakeholders have a claim on the resource. The best known example of collaborative resource management is something called “habitat conservation planning.”14 This is a negotiated environmental planning process involving multiple parties, aimed at allowing development and resource extraction to occur, while at the same time protecting endangered species and ensuring the viability of fragile ecosystems. To explain the process requires a bit of background. Under an American law called the Endangered Species Act, it is prohibited to “take” (meaning to kill or harm) a species that is officially listed by the agency as “endangered” (to be endangered means the species is at risk of 11. See Lawrence E. Susskind & Joshua Secunda, The Risks and the Advantages of Agency Discretion: Evidence from EPA’s Project XL, 17 UCLA J. ENVTL. L. & POL’Y 67, 84–85 (1998/99). 12. See Jody Freeman, The Contracting State, 28 FLA. ST. U.L. REV. 155, 193–94 (2000) [hereinafter Freeman, The Contracting State]. 13. Id. 14. See generally J.B. Ruhl, How to Kill Endangered Species, Legally: The Nuts and Bolts of Endangered Species Act “HCP” Permits for Real Estate Development, 5 ENVTL. LAW. 345 (1999) [hereinafter Ruhl, “HCP” Permits]. http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5 2005] LESSONS FOR JAPAN 1865 extinction).15 Many activities, both public and private—from building dams to cutting timber, to developing residential housing—would not be able to go forward if they destroyed essential habitat for such species. This would amount to a prohibited “take.” However, the same law also provides for a special permit process under which parties may “take” endangered species, as long as they negotiate an agreement under which the effects on the endangered species might be mitigated. The process by which this is done is called a “Habitat Conservation Plan” (HCP)16 As with negotiated rulemaking, here, a government agency (the U.S. Fish and Wildlife Service) convenes all the relevant stakeholders, including, for example, the permit applicant, the state and federal agencies in charge of fish and wildlife, local community groups and environmental groups, and local governments. And together, they negotiate a comprehensive plan.17 Unlike negotiated rulemaking, however, the product is a plan for managing a complex area over time, rather than a rule an agency will promulgate. The HCP ideally represents an attempt to take a more holistic and preventive approach to activities that are likely to have an adverse environmental effect. The advantage of the process is that it might go beyond what the Endangered Species Act would normally accomplish.18 Under this law, one simply cannot do anything to harm a species that is listed as endangered. Yet this protection only applies once that species is literally on the brink of extinction. It is an “all or nothing” approach. Once listed, the species is fully protected, but before that it is not. The HCP process allows participants to be more proactive and comprehensive. It enables them to protect other species that are not yet on the brink of extinction and to protect aspects of the environment that are not directly related to the species. One can see the similarity here with Project XL in that the flexibility of negotiation gives the agency an opportunity to perhaps regulate beyond its traditional reach—to regulate activities and issues that it normally does not regulate, in exchange for giving flexibility in other areas. These Habitat Conservation plans can provide for setting aside land, mitigating environmental harms, and taking a wide variety of measures for ecosystem protection that are very costly. They can cover from one acre to hundreds of thousands of acres and last anywhere from 5 to 100 years. 15. 16. 17. 18. See Endangered Species Act 16 U.S.C. §§ 1531–44 (2000). See generally Ruhl, “HCP” Permits, supra note 14. Freeman, The Contracting State, supra note 12, at 195. Id. Wash U Law Repository 1866 WASHINGTON UNIVERSITY LAW QUARTERLY [VOL. 83:1859 This is far more than the statute’s simple prohibition on “taking” a species could ever accomplish. And the success of HCPs relies extensively on expertise that lies outside of government.19 For instance, some environmental groups with experience in land conservation play key roles as managers and monitors of these plans over the long term. Indeed, this is a premise of collaborative governance: that knowledge, information, and expertise resides outside of government. So private actors may play new roles in a collaborative approach. So too might public agencies play new roles. The HCP process, just like Project XL and regulatory negotiation, illustrates this.20 For example, the agency in charge of the HCP process has tried to give permit applicants an incentive to negotiate plans by issuing a policy stating that once a Habitat Conservation Plan is finalized, the agency will not add more burdens later on. If the applicant abides by the terms of the agreement (for example, setting aside so much land for species, spending so much money mitigating the environmental impact of the development, and monitoring the health of the ecosystem), the applicant has fulfilled its responsibility. This seems similar to what one would expect in a standard private contract between two private parties. The purpose of the “no surprise” policy is to give the permit applicant some certainty about the “deal.” But in this case, it is government giving an assurance to a private party, and it seems to limit government’s traditional regulatory power in the future. This raises important questions about how much a government both can and should bind its own regulatory power through agreement. It raises the even more fundamental question about the extent to which the relationships between agencies and private parties can or should mirror those in the private sector. This brings us, of course to the most difficult question of all, which is, what is the appropriate role for government in regulation? The central idea behind collaborative governance, as a normative theory, is to convert a top-down, hierarchical approach to governance—in which an agency dictates requirements, and private parties resist compliance—into a more horizontal, multi-lateral approach in which public and private parties engage in pro-active problem-solving, working together to design and implement a regulatory regime.21 19. See Jody Freeman, The Private Role in Public Governance, 75 N.Y.U. L. REV. 543, 553–54 (2000). 20. See Freeman, The Contracting State, supra note 12, at 208. 21. Freeman, Collaborative Governance, supra note 3, at 27. http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5 2005] LESSONS FOR JAPAN 1867 Naturally, of course, not everything in these negotiations can be up for grabs: the agency must still set parameters within which consensus will be achieved. The agency maintains its role as the representative of the larger public interest and remains constrained by the limits of relevant statutes. In other words, there are still legal, and of course political, constraints on agency flexibility. Yet the agency takes on new roles in collaborative governance as a convenor of affected parties, a facilitator of problemsolving, and as a partner throughout the regulatory process as agreements are implemented.22 Moving toward collaboration would present a new set of challenges for administrative agencies in the American context. It highlights the need to develop a different self-conception—a different “identity” if you will— among agency officials.23 Collaboration is something for which agency staff are not especially well-trained and toward which many of them are not favorably disposed. Many regulatory agencies in the United States have developed internal cultures that are quite adversarial. They are used to being regulators, not facilitators, conveners, and partners. They are not typically trained, for example, in dispute resolution. This is understandable of course; regulators are trained to set standards, and enforcement staff to punish violators. Still, the lack of exposure to alternative approaches is striking. Collaborative initiatives take time to develop. They depend on the cultivation of relationships and the development of norms, along with training in techniques like negotiation, something that I suspect is very much a part of the Japanese administrative process and second nature in Japanese society. But this is relatively foreign to the American process. Americans are, I dare say, notoriously impatient (I say this as a Canadian having lived in the U.S. most of my adult life). I say this only half facetiously—American culture is “muscular” compared to others. negotiation and accommodation are not dominant reactions to conflict, I have found. But more important than this, perhaps, are the political imperatives that shape agency behavior. Political appointees who serve only a short time want to claim quick and clear successes. Collaborative initiatives are complicated, time consuming and not always successful, at least not in the short-term. All of this is to say that because of incentives peculiar to the American system, and because of unique aspects of 22. Id. at 31–33. 23. See id. at 13–14. Wash U Law Repository 1868 WASHINGTON UNIVERSITY LAW QUARTERLY [VOL. 83:1859 American culture, there are many obstacles to collaborative governance within American administrative agencies themselves.24 As a theoretical and normative idea, collaborative governance represents an alternative to the traditional adversarialism that characterizes American regulation, implementation and enforcement. As a practical matter, adopting a collaborative approach may improve both the quality and legitimacy of the regulatory process. To assess whether the theory works in practice, however, more experimentation is needed along with monitoring and empirical evaluation. This is developing, but it is going slowly. For example, Professor Langbein and I have written an article assessing the empirical performance of regulatory negotiation.25 Our assessment was based on a two-part study that she conducted with another academic, Neil Kerwin. In the first stage of the study, participants in a set of negotiated rulemakings were interviewed extensively to learn what they thought of the process.26 In the second stage, Kerwin and Langbein evaluated a set of negotiated rules against comparable rules produced in the traditional notice-and-comment process.27 The study shows that the participants in negotiated rulemaking are more satisfied with negotiated rules; that they think the negotiated rules will be easier to implement; that they learned much more than they had in the past in the traditional adversarial process; and that they think the rules they produced were of superior quality to what the agency would have produced on its own.28 It may be, as others have argued, that negotiating rules is as costly as the traditional approach, especially if it is necessary to subsidize the less wealthy interest groups in order to ensure that they participate. And of course negotiations can take a long time, usually between a year and two years but sometimes longer. We believe, however, that time spent on the front end is likely to save time on implementation on the back end; at least, this is certainly worth looking into.29 We concede that negotiating rules will not necessarily eliminate the threat of litigation. There will still be litigation in some cases. There is always a risk that some parties will negotiate and then, if they do not get all they want, turn around and sue. Yet if there is to be a lawsuit, the fact that the challenged rule was 24. See generally id. at 67–82. 25. Jody Freeman & Laura I. Langbein, Regulatory Negotiation and the Legitimacy Benefit, 9 N.Y.U. ENVTL. L.J. 60 (2000). 26. Id. at 79. 27. Id. at 105–06. 28. Id. at 121–22. 29. Id. http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5 2005] LESSONS FOR JAPAN 1869 negotiated can narrow the areas of conflict, reducing the burden on both the parties and the reviewing court. Still, we believe that the negotiation process can reduce the likelihood of litigation. For example, signing an agreement that parties will not litigate once they agree to the consensus rule at least creates a norm against such behavior. Moreover, doing so would be risky: if parties do this, they will not be regarded as reliable negotiating parties in the future. So there is a strong possibility that those who wish to be repeat players (including corporations, industry trade associations, as well as environmental, labor, and consumer groups) will behave reliably. There may also be litigation, and understandably, if the agency departs from the consensus rule and promulgates something different. In those cases, we can expect court challenges. Still, in terms of the quality of the rule and its acceptability to the major affected parties, we feel that negotiated rulemaking is comparatively superior.30 I should hasten to add that there is a lively debate about this in the United States, with some scholars arguing that negotiated rules and other collaborative experiments like those I have discussed here, are not legitimate because they amount to “deals” struck among insiders. This is a criticism that I have heard about the Japanese system as well, and it is a reminder that we need to erect safeguards against the possibility of agency capture by the most powerful interests. The careful design of the process, to ensure that it includes a variety of representatives from different interest groups, is crucial. And the agency must understand and actively play its role on behalf of the larger public interest. The agency should not be a passive arbiter, acceding to any deal as long as parties agree. In other words, agreement itself should not be a sign of success. Rather, the question is, does the rule, permit, plan, or agreement being negotiated advance the policy articulated by Congress in the relevant statute and will it accomplish the regulatory goal? Judicial review can play an important role here, ensuring that the procedural safeguards governing the negotiation have been observed. I will return to this when I discuss the possibilities for collaborative governance in Japan. In addition to empirical work on negotiated rulemaking, there are also some empirical studies of Project XL. Some believe that Project XL is largely successful—that it speeds up the permit process for industries that need faster approval for process changes, and also results in a net gain for environmental protection. Yet others believe that XL amounts to private 30. Id. Wash U Law Repository 1870 WASHINGTON UNIVERSITY LAW QUARTERLY [VOL. 83:1859 “sweetheart” deals with corporations.31 Critics claim that the process is not truly balanced but instead dominated by industry, and they believe that regulatory “flexibility” simply means weaker regulation. A final objection is that Project XL is beyond the authority of the EPA because no statute explicitly authorizes agencies to pursue this approach.32 So while collaborative governance remains promising as a theory, and has been used with some success in practice, it is not without its critics. It is not clear yet just how much stakeholders and agencies wish to embrace it. And it is not clear how generalizable it will be beyond the contexts in which it is currently being tried. For example, it would be important to do research on collaborative efforts in financial regulation and other sectors to know if it can be successful more generally. As a cultural matter, collaborative governance seems much more Japanese than American. I say this because of its relatively greater reliance on informality and its relative lack of adversarialism. A collaborative approach seems consistent with the historical reliance in Japan on informal guidance (gyosei shido) and the principle of negotiating to arrive at “voluntariness.”33 By relatively greater reliance on informality I do not mean, however, to suggest a lack of lawfulness or legal authority for government action. I mean instead two things. First, that a negotiated or problem-solving approach requires a government agency to have some flexibility and discretion, and also to be more adaptable when it plays the unfamiliar role of negotiating partner than it might otherwise be when simply dictating regulation. This is quite different from a hierarchical approach in which government simply promulgates rules after passively receiving input from parties in a process that everyone regards as a zerosum game. Second, collaborative governance is somewhat informal in that, frequently, it relies on agreements that are more akin to contract than to traditional regulation. For example, while negotiated rulemaking ultimately produces a promulgated “rule,” the underlying basis for that rule is really a rather informal agreement struck between the parties. Similarly, the negotiated approach to permitting in Project XL relies on a “final project agreement” (FPA) which looks more like a contract than a traditional permit. The FPA contains the mutual commitments made by the 31. See Rena I. Steinzor, Reinventing Environmental Regulation: The Dangerous Journey from Command to Self-Control, 22 HARV. ENVTL. L. REV. 103, 141 (1998). 32. See generally Bradford Mank, The Environmental Protection Agency’s Project XL and Other Regulatory Reform Initiatives: The Need for Legislative Authorization, 25 ECOLOGY L.Q. 1 (1998). 33. See CARL F. GOODMAN, THE RULE OF LAW IN JAPAN: A COMPARATIVE ANALYSIS (2003). http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5 2005] LESSONS FOR JAPAN 1871 government agency, the regulated firm, and other interested parties such as local community and environmental groups. Only after the FPA is signed does the agency issue a traditional permit to the company. This is also true of habitat conservation plans, which proceed as structured contract negotiations even though they ultimately result in a permit. The negotiation occurs among a host of actors: the developer, federal and state agencies, as well as environmental and local community groups. To be more accurate, however, such an “agreement-based” approach to regulation seems to be something of a hybrid between the Japanese reliance on “guidance” and “voluntariness,” and the traditional American system of more formalized procedures.34 Indeed, the best way to describe many collaborative experiments is as a mix of formal and informal agency action. In some instances, collaborative governance is highly formalized. For example, remember, the habitat conservation planning process is authorized and structured by the Endangered Species Act. And regulatory negotiation which was at first experimental, was subsequently authorized and structured by Congress in a statute called the Negotiated Rulemaking Act.35 Indeed, this statutory reform was proposed by ACUS and supported by the ABA. This is part of what Professor Levin has called “the era of adaptation in administrative law reform.” Professor Lubbers will explain more about this law, so I will not cover it in detail. Suffice it to say that it requires all regulatory negotiations conducted by federal agencies to be open to the public; that minutes of meetings be kept; that consensus rules still be circulated for public comment; and that any resulting rule promulgated by the agency, regardless of the consensus reached, can still be challenged in court and subjected to legal review. Importantly, courts are not to grant consensus rules any greater deference than they do traditional rules. So there are many safeguards that accompany the collaborative approaches that have been formally authorized. Indeed, for some, there are too many. That is, formalizing collaboration and spelling out detailed procedures that must be observed is thought by some to interfere with the creativity and flexibility that a negotiated approach ideally affords. Of course, the purpose of the procedures is to make the process transparent and therefore accountable. This raises a potential trade-off between accountability to the public and a focus on coming to creative agreements 34. Id. 35. Pub. L. No. 104-320 (1996). Wash U Law Repository 1872 WASHINGTON UNIVERSITY LAW QUARTERLY [VOL. 83:1859 that will be, among other things, implementable. Greater formality can add expense and delay as well. In any event, to say that collaborative governance is relatively informal is merely to say that the negotiation process itself is relatively flexible and meant to promote “thinking out of the box,” and also that the final agreements among the parties are not, unless they are incorporated into a rule or permit, independently legally enforceable. They are more like informal “voluntary” agreements, with the threat of traditional regulation lurking in the background. At the same time, it is also true that some experiments in collaboration are not authorized by statute. These are closer still to the informality of the Japanese administrative process because they rely on an agency’s assumption that it has the power to engage in a more collaborative process as part of its broadly conceived enforcement discretion. Project XL falls into this category. The EPA traditionally has considerable discretion to determine the content of permits and to enforce them as the agency sees fit. Cast this way, negotiating an XL agreement is perfectly within the agency’s traditional authority. And courts are wary of interfering with an agency’s enforcement discretion. Similarly, by relative lack of adversarialism I do not mean to suggest that the parties in a collaborative process have no interests or positions. They would not willingly join a negotiated enterprise unless they had something to gain. Without a robust regulatory process in the background, they have no incentive to negotiate. Obviously, collaboration requires a shadow legal default against which the negotiating occurs. Nor do I mean to suggest that the traditional regulatory system is completely devoid of negotiation. The rulemaking process, the permit application process and the enforcement/settlement process involve a great deal of back-and-forth between the agency and the affected parties. Still, the difference between this and an explicitly collaborative approach is that the former reats the “parties” as outsiders to decisionmaking whereas the latter treats them as partners who might share in decisionmaking from the choice of regulatory standard through implementation. Without the right incentives, and without the potential for “win-win” solutions, collaboration would not work. In a way, all of this may sound familiar to you. It seems that collaboration already, to a significant extent, characterizes the Japanese approach to regulation. From what I have learned of the Japanese system, agencies rely heavily on negotiation, persuasion, and cajoling, all of which are aimed at arriving at an eventual consensus between regulator and regulated. So to suggest a normative model based on negotiation and http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5 2005] LESSONS FOR JAPAN 1873 consensus perhaps would not strike any of you as unusual. But it is quite controversial in the United States. Yet, it may be precisely because collaborative governance is so superficially compatible with Japanese administrative law and culture that it would not be suitable for transplantation from America, at least not without some very significant adaptation. If I understand the Japanese system correctly, administrative agencies tend to use administrative guidance as their primary regulatory tool. Professor Goodman’s estimate, in his article on the rule of law in Japan, was that 80% of administrative activity in Japan is through the use of guidance.36 And I know that Professor Nakagawa has written about how Japanese lawyers understand guidance. My impression is that guidance is best described as a process of negotiation and consultation, the aim of which is voluntary agreement. But of course because guidance is not the same as a “disposition,” it cannot be subject to judicial review in the Japanese system. In addition, if again I am correct, the Japanese system is characterized by two other features that are different from the United States: (1) very broad discretion to interpret and expand legislative mandates beyond specifically delegated “duties” to more amorphous and less easily cabined functions; and (2) virtually no serious fear of being reversed by courts because of the difficulty of access to courts for plaintiffs, and because principles of strong deference apply when courts do review agency decisions. Furthermore, there is no presumption of reviewability.37 Collaborative governance (both the ideal theory and the less than ideal practical experiments I have described), has developed in the United States in a particular context and against a background that is quite different from the situation in Japan. The American administrative system is characterized by a highly proceduralized notice and comment process for rulemaking; sophisticated interest groups (representing not only industry but also consumers, environment, and labor) who participate extensively in every stage of rulemaking; fairly low standing thresholds for accessing judicial review; a presumption of reviewability; and flexible standards of deference that allow, at least in some cases, quite vigorous judicial review of agency action. When collaborative governance emerged, it was against the background of this traditional regime becoming highly adversarial, slow, and burdensome, so much so that one scholar called the rulemaking 36. See GOODMAN, supra note 33. 37. Id. Wash U Law Repository 1874 WASHINGTON UNIVERSITY LAW QUARTERLY [VOL. 83:1859 process “ossified.”38 Collaborative governance was intended to be at least a partial solution to these problems. Collaborative governance may be less suitable in Japan, however, because it seems that Japanese administrative law is not afflicted with the same problems. Ironically, to make collaboration function effectively in Japan, it might be necessary to build some additional formality back into the system. Collaboration requires procedural safeguards if it is to work well. These include judicial review to oversee and check the process, and sophisticated interest group participation to ensure that negotiations do not become bilateral deals. 38. Thomas O. McGarity, Some Thoughts on “Deossifying” the Rulemaking Process, 41 DUKE L.J. 1385 (1992). http://digitalcommons.law.wustl.edu/lawreview/vol83/iss6/5