  |
[*pg 1015]
BARGAINING IN THE SHADOW OF ADMINISTRATIVE PROCEDURE:
THE PUBLIC INTEREST IN RULEMAKING SETTLEMENT
INTRODUCTION
I. RULEMAKING SETTLEMENT
A. Settlement as a Type of Collaborative Governance
B. Problems with Rulemaking Settlement: A Principal-Agent Gap
II. RULEMAKING SETTLEMENT AGAINST THE BACKDROP OF PRESIDENTIAL TRANSITIONS
A. Settlement by Outgoing Administrations
B. Settlement by Incoming Administrations
III. CLOSING THE PRINCIPAL-AGENT GAP IN RULEMAKING SETTLEMENT
A. Monitoring Settlement Participation
B. Ex Ante Hard Look Review of the Merits of Settlements
CONCLUSION
FOOTNOTES
 |
INTRODUCTION
In recent years, a new account of administrative law, favoring private ordering over state-imposed solutions, has bolstered the acceptability of negotiated approaches to regulatory problems.1 Consistent with this account, administrative law has seen a growing trend toward flexible, consensual mechanisms for regulation,2 emphasizing less rigid, cooperative approaches over prolonged adversarial disputes. Procedural innovations, such as negotiated regulation (known less formally as "reg neg"), have proliferated as alternatives to more traditional administrative procedures, such as notice and comment rulemaking. Reformers' embrace of such solutions for their promise [*pg 1016] in promoting private consensus over public mandate has engendered much discussion in the literature.3
This Essay addresses the type of private ordering relating to the settlement of lawsuits challenging administrative rules or final agency actions, also known as "rulemaking settlements." Hardly a procedural innovation, the prospect of settlement is a traditional component of any strategy to influence agency decisionmaking. Rulemaking settlement is not only a private strategy, but also may serve as an important vehicle for an agency to implement its policy decisions. Despite its significance, the settlement of lawsuits is an understudied tool for implementing regulatory policy.4 Settlement is certainly more common than negotiated regulation,5 but unlike negotiated regulation it has received scant attention in the administrative law literature. One recent study describes rulemaking settlement as a "blind spot," mysteriously unaccounted for in both doctrinal and theoretical accounts of administrative law.6
In an effort to bring this blind spot into view, I describe rulemaking settlements in Part I and discuss how they differ from other collaborative governance mechanisms. Put simply, the basic process by which such settlements are implemented does not necessarily afford the same procedural protections as the routine agency rulemaking process or other collaborative approaches such as negotiated regulation. The Essay draws on a simple principal-agent structure to illustrate how administrative agencies face strong incentives to ignore the interests of important stakeholders -- principals who otherwise would be afforded procedural protections -- presenting a principal-agent gap in the rulemaking settlement context. This gap is widened by the threat of private parties' exit from settlement negotiations, [*pg 1017] coupled with high expected litigation costs that sometimes lead agency litigants to make concessions on appeal that they would not make in the context of the ordinary notice and comment process.
Part II addresses an example that raises an especially salient concern for administrative law doctrine and theory: rulemaking settlements against the backdrop of presidential transitions. The presidential transitions that follow elections often cause policy shifts, sometimes even when a new administration is of the same political party as an outgoing one. The prospect of an outgoing administration's settlement binding an incoming president's agency, or an incoming administration's settlement undermining one or more regulatory initiatives of an outgoing president, is hardly a new concern. However, the special legitimacy problem raised by settlements against this backdrop illustrates a need for serious reflection on the procedural protections in place during settlement -- if indeed there are any at all. Since private bargaining and agency concessions during settlement occur largely outside of the apparatus of administrative procedure, procedural protections designed to protect and promote the public interest may readily be skirted. In negotiating settlements, an agency will tend to afford large weight to some principals, including the executive branch and settling parties, but little or no weight to other key principals, including Congress.
Part III discusses some specific administrative law doctrines that, if respected by lawyers and enforced by courts, hold promise to illuminate the settlement process, taking it out of the obscure corner it occupies in administrative law circles and improving its accountability. I argue that the principal-agent gap resulting from settlement can best be addressed by focusing not only on the potential for exit by private litigants, resulting in the prospect of ongoing litigation for the agency, but on "voice" and "loyalty" as alternative mechanisms for modifying institutional behavior. Given the private bargaining behind a settlement or consent decree as well as the concomitant pressures for agency bias toward only those perspectives present at the bargaining table, I argue that some judicial role in protecting the public interest is appropriate. At the same time, judicial scrutiny of settlements must be balanced against preservation of private stakeholder participation in rulemaking settlements. I advocate expansive participation in settlement processes, a type of voice interest, but argue that this alone is not enough to ameliorate the principal-agent gap posed by rulemaking settlement. Specifically, attention to loyalty interests -- fidelity to statutory criteria and the procedural mechanisms designed [*pg 1018] to protect the public interest -- necessitates an ex ante hard look review of a rulemaking settlement at the time that an appeal is dismissed, or a consent decree or stay is approved. This approach, I argue, will help to close the principal-agent gap in rulemaking settlement and encourage better participation in settlement by private stakeholders, while also allowing stakeholders ample opportunity to enforce settlement terms.
| |
I. RULEMAKING SETTLEMENT
Agencies routinely enter into settlements limiting the scope of their regulatory discretion. To a degree, rulemaking settlement fits into a broader trend toward collaborative governance in the administrative state. However, the analogy between rulemaking settlement and other collaborative governance techniques, such as negotiated regulation, is limited. At the fore are concerns about the potential for abuse of settlements by private stakeholders and the more general implications for administrative legitimacy. Because settlement is most controversial when it reflects a compromise on substantive issues of interpretation or policy, this Essay focuses its discussion on these types of settlements rather than on less controversial -- but perhaps more common -- settlements in which the government merely agrees to a schedule or process for rule consideration.7 Conceptually, rulemaking settlement may be more likely than negotiated regulation to present a principal-agent gap because agency negotiators, in an appellate posture, face incentives for skirting important stakeholder perspectives -- including those that may be most concerned with safeguarding the public interest.
A. Settlement as a Type of Collaborative Governance
Settlement is hardly a novel concept in the administrative process. For several years, regulators have embraced settlements and other alternative dispute resolution techniques as fundamental components of the administrative process. Unlike traditional litigation, alternative dispute resolution provides an opportunity to deflect the adversarial nature of many administrative proceedings, while also empowering private stakeholders. Such techniques are touted -- often [*pg 1019] as "innovations" -- for their flexibility, efficiency, and legitimacy over more traditional administrative procedures.8
For example, it is now commonplace for an agency to offer the opportunity for mediation and settlement to parties to an adjudication, such as an enforcement proceeding.9 Such techniques can assist an agency in securing compliance with its regulations while also offering private parties the opportunity to avoid protracted litigation. Settlement raises little potential for controversy where it is bilateral and between two private parties, as may be the case in a license contest proceeding.10 In a sense, the government is not a party to such proceedings, although an agency may play an important role as referee in ensuring that the settlement outcome between private parties reflects the public interest. The greater the public interest issues raised in an agency proceeding, the more controversial settlement may become.11
But most settlements involving administrative process are not extragovernmental. If, for example, an agency brings an enforcement action against a private party, the dispute between the agency and the private party is often bilateral, but the nature of the interests at stake differs in a fundamental way from a bilateral settlement between private negotiators. The position of a regulated party in challenging the [*pg 1020] agency's action and entering into negotiated solutions may be motivated by private interests, but the agency is constrained (and hopefully motivated) by the public interest, including institutional and procedural checks designed to ensure that this public interest is not sacrificed. Matters become substantially more complex when private stakeholders with divergent interests must reach consensus and negotiated solutions to complex, multipolar private conflicts in addition to accommodating the public interest.
This last scenario is perhaps most common in the context of agency rulemaking. The Administrative Procedure Act (APA)12 sets out the basic framework for agency decisionmaking, including the notice and comment rulemaking process that agencies use to make law and policy. In the typical notice and comment procedure, an agency will publish a notice of proposed rulemaking, allow a period for public comment, consider comments submitted to it, and then approve and publish a final rule.13 Judicial review of the agency's process and regulation is separate from, but inextricably linked to, this process. Courts review agency regulations adopted through the rulemaking process to determine whether they are "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with the law."14 Since the 1970s, more expansive definitions of standing have increased the number, and broadened the range, of parties allowed to participate meaningfully in the judicial appeals process.15 Doctrinal developments also have created more opportunities to challenge rules, not just in as-applied enforcement proceedings but also prior to [*pg 1021] their enforcement.16 The unintended result, some suggest, is a system of "adversarial legalism,"17 characterized by the breathtaking frequency and length of judicial appeals of agency rules.18 Some see this appeals process as dysfunctional, ossifying the rulemaking process and detracting from sound policy formulation through rulemaking.19
Negotiated regulation, a consensus-based procedure in which stakeholders negotiate the substance of a proposed rule, is one procedure designed to address such problems.20 In the typical negotiated regulation, private stakeholders, along with an administrative agency, negotiate the substance of a proposed rule, which the agency then subjects to a more traditional notice and comment rulemaking process. Many commentators praise negotiated regulation for its potential advantages over traditional notice and comment rulemaking in promoting the ideals of collaborative governance of the administrative state.21 In implementing negotiated regulation, however, courts and commentators alike struggle to create adequate incentives for private stakeholder participation, on the one hand, while still promoting the public interest, on the other.22
Negotiated regulation holds out many advantages over the traditional notice and comment process for adopting rules. Since the negotiated regulation process commences prior to the initiation of official notice and comment procedures, parties have an early opportunity to lessen the adversarial tone that dominates many large rulemaking proceedings. If stakeholders can agree on a negotiated regulation prior to agency adoption of the regulation, appeals of the final regulation will be less likely. Negotiated regulation has received much discussion in the literature, with many commentators praising the technique,23 and others raising concerns with the strong potential for interest group domination of the process.24 At its worst, negotiated regulation can subvert the public interest to the extent that it accommodates the interests of private stakeholders in ways that either ignore important perspectives or conflict with regulatory goals.25
But negotiated regulation is hardly the only -- and certainly not the predominant -- negotiation technique utilized by administrative lawyers. It is far more common for regulations and final agency actions to be challenged on appeal before courts, where opportunities for rulemaking settlement are far more frequent than -- but certainly may overlap with -- traditional negotiated regulation.26 In fact, even in [*pg 1023] those cases in which negotiated regulation has been agreed to, parties may later attempt to challenge an agency's regulations in court.
Settlement of appeals is a common way for an agency to dispose of an ongoing adversarial dispute.27 As a category, rulemaking settlement includes pure bilateral settlements, in which private entities and an agency agree on a course of action but courts play little or no enforcement role, as well as consent decrees, in which a court's judicial imprimatur has preemption and enforcement effects far beyond purely consensual contractual settlement.28 Like negotiated regulation, settlement of appellate litigation has much to commend it, including the promotion of flexibility, efficiency in litigation and agency enforcement, expediency on the part of both government and private litigants, and legitimacy -- to the extent settlement reflects the shared consensus of stakeholders.
For example, in 1987, the Occupational Safety and Health Administration (OSHA) published its final rule regulating workplace formaldehyde exposure.29 OSHA set its long-term formaldehyde exposure limit at 1 part per million (ppm).30 Unions appealed, arguing that long-term formaldehyde exposure levels should be set at 0.5 ppm rather than 1 ppm.31 On appeal, the District of Columbia Circuit remanded the rule because it concluded that OSHA needed to elabo- [*pg 1024] rate more clearly its scientific reasoning.32 When OSHA stalled on remand, industry and union groups independently began negotiations. The private negotiations resulted in a compromise, setting the exposure limit at 0.75 ppm, along with some modifications to other requirements in the rule.33 The joint union-industry proposal was submitted to OSHA, subjected to OSHA review, and reintroduced into the notice and comment process. After a brief and uneventful notice and comment process, the rule became final in June 1992.34 The District of Columbia Circuit's reversal of the original formaldehyde rule had sent a signal that created strong incentives for the parties to agree to a more stringent standard than that originally proposed by OSHA. As one study describes it, "the very real threat that a more stringent (0.5 ppm) standard would be set by traditional rulemaking made possible the negotiation of a less stringent (0.75 ppm) standard."35 To the extent that OSHA fully utilized a notice and comment process to implement the terms of its settlement, the formaldehyde settlement is relatively uncontroversial.36
In sum, both private stakeholders and agencies value settlement to the extent it assists them in reducing or avoiding litigation costs. Additionally, as OSHA's formaldehyde settlement illustrates, stakeholders see settlement as an important strategy for committing an agency to a course of action, often involving a substantive policy choice. Not surprisingly, the legality of rulemaking settlements is sometimes called into question.37 Once an agency enters into an oth- [*pg 1025] erwise valid settlement, modification of the terms of the settlement through interpretation by that agency can give rise to considerable confusion and undermine private stakeholder incentives for participating in settlements with the government.38 Similarly, if settlements are not capable of enforcement, private stakeholders will face little incentive to participate. On the other hand, always allowing settlement to bind agencies can thwart flexibility and undermine the public interest values promoted by administrative process.
B. Problems with Rulemaking Settlement: A Principal-Agent Gap
Although rulemaking settlements may overlap with collaborative governance mechanisms such as negotiated regulation, they also differ in fundamental and important ways. In the course of briefly discussing the key differences between rulemaking settlement and negotiated regulation, I sketch a simple principal-agent account of administrative procedure to illustrate the particular deficiencies unique to rulemaking settlement as compared to other collaborative governance techniques, such as negotiated regulation. These differences illustrate how settlement may fall short of the goals of collaborative governance techniques. The principal-agent model posits an agent, here an administrative agency, as influenced by institutional principals, here Congress, the president, and private stakeholders, who are routinely key participants in the decisionmaking process under the APA. As developed below, the incentives an agency faces to settle an appeal may increase monitoring costs for these principals, thus contributing to a principal-agent gap.39
The differences between rulemaking settlement and other collaborative governance techniques can be traced in large part to the unique incentives that private litigants and agencies face when settling appeals of agency rules. My analysis of the structural architecture of rulemaking settlement draws on the common tripartite focus on "exit," "voice," and "loyalty" as ways to influence institutional behavior.40 In the settlement context, the threat of exit -- in the form of expected private stakeholder litigation and the ensuing costs to the agency -- is the predominant incentive influencing agency litigants. By threatening continued litigation against an agency, private stakeholders in the appellate posture may be in a position to extract concessions that an agency refused to make in the notice and comment process. This is the primary incentive that private stakeholder exit presents to agency litigants. The incentives that private exit poses for agency negotiators to settle appeals, coupled with high expected litigation costs to the agency as a consequence of such exit, contribute to the principal-agent gap in the appellate context. That is, private incentives for exit coupled with high expected litigation costs for the government may lead agency negotiators to make concessions in settlement that are not necessarily in the public interest, widening the disparity between the agency's actions and those institutions and participants to which the agency is accountable. After discussing the exit problems, I discuss how attention to voice and loyalty might help to minimize the principal-agent gap created by the threat of exit.
The principal-agent gap in the context of rulemaking settlement is more problematic than in other contexts. First, settlement of an appeal only becomes a possibility where traditional negotiated regulation has fallen short in the goal of defusing adversarial challenges at the front end of the agency regulatory process. Because the notice and comment process runs its course prior to appeal of a final regulation, rulemaking settlement does not provide the same benefit of discouraging litigious behavior. Rulemaking settlement becomes a useful [*pg 1027] strategy only once negotiated regulation has failed to prevent litigation. It should be noted, however, that rulemaking settlement in the context of appellate litigation also may jump-start a new negotiated regulation process before the agency. Although rulemaking settlement can provide a range of benefits, including providing a forum for negotiating outside the litigation process altogether,41 it does not provide the same forward-looking opportunity to avert adversarialism that negotiated regulation does. To the extent an adversarial mindset already has set in, and perhaps solidified, among parties to an appeal by the time that settlement negotiations commence, the prospect for strategic behavior on the part of participants is stronger in the context of rulemaking settlement than in other collaborative governance contexts. Such strategic behavior has consequences for how the agency and private parties participate in settlement negotiations.
A second difference is that rulemaking settlement frequently involves a more limited number of parties than negotiated regulation. In rulemaking settlement, the stakeholders participating in the settlement are not necessarily inclusive of all those who might have been interested in a regulation from initiation of the notice and comment process, but instead is limited to parties present on appeal. Professor Cary Coglianese has noted, for example, that "individual members of the public file a considerable number of comments in the rulemaking process but are left out of the nonpublic litigation process."42 If the incentives for private exit are particularly strong, and expected litigation costs are high, agencies may find it advantageous to limit the parties to an appeal and enter into negotiations.
To be sure, negotiated regulation, which also limits participation, raises a similar concern,43 but particular caution may be warranted in touting the benefits of rulemaking settlement given the strong incentives for agency collusion in a limited settlement. Sometimes this deal is not implemented through the notice and comment procedures that [*pg 1028] were used in adopting the regulation that is the subject of litigation and settlement.
For example, consider the rulemaking settlement that grew from OSHA's now-infamous lockout/tagout rulemaking. In 1989, OSHA published a final regulation that required employers to place locks and tags, such as circuit breakers, on machines and power supplies to prevent industrial accidents.44 Six industry groups and two labor unions filed for judicial review.45 The parties could not agree to a settlement. However, one of the industry appellants, American Petroleum Institute, entered into a compromise with OSHA.46 Pursuant to this compromise, OSHA agreed to insert nonmandatory guidelines into its enforcement manual.47 OSHA also made an official "correction" to the rule, clarifying that "work permits or comparable means" could substitute for group lockouts in certain situations.48 By adopting guidelines outside of the notice and comment process -- as permitted under the interpretive rule or policy statement exceptions to notice and comment rulemaking,49 -- OSHA was able to make concessions to private stakeholders on appeal without undergoing the full notice and comment process.
Although the lockout/tagout settlement seems like a pragmatic effort on behalf of an agency to manage its scarce resources on appeal, the difference from traditional negotiated regulation should raise concerns. Union groups challenged the interpretation reached during the settlement, but the District of Columbia Circuit upheld the agency's new interpretation because it was "consistent with the record evidence and would have constituted a logical outgrowth of the proposed rules if originally promulgated as corrected."50 At a minimum, the lockout/tagout settlement is suspect to the extent that it excluded the concerns of third parties that were central to the acceptability of the agency's action under almost any account of agency legitimacy. The fact that the settlement agreement did not include the labor union heightens the potential for interest group capture of the [*pg 1029] agency decisionmaking process, rendering the agency's interpretation suspect. Although the agency's policy choice may well have been understandable given the threat of further litigation, the procedural deficiency in its implementation may raise an arbitrary and capriciousness concern.51
A third reason that rulemaking settlement raises heightened principal-agent gap problems is the secretive nature of the process. Settlement procedures, unlike negotiated regulations, are often closed to the public. Settlement often results from confidential mediation in which the parties may not only be few, but also free from public scrutiny.52 Professor Cary Coglianese observes that litigation and the prospect of settlement "offer[] interest groups and the agency an opportunity to do something they were not permitted to do in the notice-and-comment period: negotiate in secret."53 Rules governing ex parte contacts, for example, do not apply once a final rule has been issued.54 Confidential mediation is valued as a negotiation tool to the extent that it encourages all parties to air their concerns openly and honestly. At the same time, confidential settlement processes may lead agencies to adopt positions that later raise opposition from affected stakeholders. The results of a settlement may be rendered even more suspect if an agency makes concessions outside of the notice and comment process and fails to publish them in the Federal Register.55 Where the incentives for private exit are strong and where ex- [*pg 1030] pected litigation costs are high, agencies may even be the party to seek to ensure confidentiality in the settlement process.
The effects of confidentiality, coupled with the limited range of stakeholder participants, pose a monitoring cost problem in the context of rulemaking settlement. In an appellate posture, an agency and its lawyers have incentives to settle to avoid litigation costs and more quickly implement enforcement of a rule. Settlement may be preferable in such a context, but it is not always socially desirable simply because of litigation savings. Sometimes the stakeholders agreeing to a settlement may not accurately reflect the full complexities of the principal-agent relationship surrounding an agency's decisionmaking process. In particular, the interests of principals such as Congress or stakeholders not actively participating in the judicial appeal may be given short shrift. Compare this process with the ordinary negotiated regulation process, where these principals are more likely to be included in the agent's negotiations because notice and comment rulemaking has yet to run its course. In other words, a principal-agent gap may create a bias of sorts in the appeals settlement process that does not necessarily apply in other collaborative governance contexts. For example, ossification of the agency decisionmaking process may increase the costs of litigating agency action, making negotiated regulation or settlement more attractive to agencies and private parties. Unlike negotiated regulation, however, settlement may be problematic to the extent that it offers far less in the way of procedural protections and, potentially, less rigorous agency consideration of the action ultimately taken.
A fourth and final contrast between rulemaking settlement and negotiated regulation relates to the role of the courts. While negotiated regulation occurs prior to appeal, and thus operates largely outside the pale of judicial scrutiny, the involvement of appellate courts in rulemaking settlement has both adverse and beneficial implications. In the literature on civil settlements, commentators treat settlements with skepticism on the grounds that they impede the generation of precedents on important matters of public policy,56 or that they enable settling parties to act strategically to cause the creation of favorable precedents while avoiding the generation of unfavorable [*pg 1031] ones.57 To the extent that courts play a significant role in creating precedents that protect the public interest, settlement of rule appeals would seem to present the same concerns about stifling case law development and manipulation of precedent that plague settlements more broadly.
Of more potential benefit, however, is the role that courts have an opportunity to play in evaluating rulemaking settlements by approving dismissal of a case, approving consent decrees, or staying an adopted rule pending implementation of a settlement.58 Whether judicial involvement in a rulemaking settlement promotes or impairs the public interest will depend on the terms of the settlement, as well as the standards the courts apply in approving it. On the one hand, courts' willingness to approve settlements based on the consensus of a narrow range of stakeholders provides private stakeholders the opportunity to hold an agency hostage to a consensus that runs counter to the public interest. On the other hand -- notwithstanding the District of Columbia Circuit's deferential approach in the OSHA lockout/tagout rulemaking settlement -- judicial involvement holds promise to police the public interest aspects of rulemaking settlements. To serve this function, courts must fashion an appropriate set of evaluative standards for approving settlements. In addition, once settlement is reached, courts continue to play a monitoring role: if an agency reneges on the terms of a settlement, other parties to the agreement might seek judicial enforcement. In contrast to their role in rulemaking settlement, courts serve little or no role in monitoring negotiated regulation, and no role whatsoever in enforcing agency departures from the terms of negotiated regulation.59
Equating rulemaking settlement with negotiated regulation -- a process itself fraught with problems60 -- masks unique problems posed by the settlement of rule appeals. In its ideal form, negotiated regulation may be a measuring stick against which rulemaking settlements [*pg 1032] can be evaluated, and perhaps even an ideal toward which settlement should strive. It cannot be assumed, however, that rulemaking settlements automatically adhere to the collaborative governance ideals proponents of negotiated regulation embrace simply because they reflect private ordering. In particular, rulemaking settlement raises a potential principal-agent gap for an agency's policy choices that negotiated regulation does not necessarily raise. The gap is due in part to the more strategic mindset of private litigants, in part to strong incentives for exit coupled with litigation cost consequences for the agency, and in part to the increased role of the courts and the potential for the manipulation of precedent. The involvement of courts in the settlement process may hold promise as a way of mitigating this gap, increasing the accountability of the settlement process and its regulatory results. After exploring some particular examples of settlements run amok from the perspective of sound administrative procedure in the next Part, I return to this theme in Part III, exploring how courts can balance exit against voice and loyalty interests to hold agency litigants accountable.
| |
II. RULEMAKING SETTLEMENT AGAINST THE BACKDROP OF PRESIDENTIAL TRANSITIONS
When rulemaking settlement works well, it conforms to the procedures afforded by negotiated regulation, effectuating a form of collaborative governance between private interests and the agency. This is most likely where a broad range of stakeholders participates in the settlement and an agency affords new notice and comment procedures to legitimate the results of a settlement. OSHA's formaldehyde rule is a good example.61Although OSHA's compromise on appeal departed from the specific terms of its proposed rule, OSHA's position in the settlement was within the scope of its initial position in the proposed rulemaking -- that the difference in the attendant risk associated with a change from 0.5 ppm to 1 ppm was not "significant."62 Thus, the agency's agreement to adopt a 0.75 ppm standard easily could be said to be a logical outgrowth of the initially proposed rule, not subject to the provision of new notice and comment.63
Rulemaking settlement raises potential problems, however, to the extent that crucial stakeholders are excluded from the negotiation, as was the case with OSHA's lockout/tagout settlement.64 Moreover, rulemaking settlement may jeopardize the public interest if an agency departs substantially from the position it endorsed during previous notice and comment procedures, or to the extent that the settlement reflects an unscrutinized shift in the agency's position. A salient illustration of this problem with settlement arises during, but is not limited to, presidential transitions -- events that often reflect major substantive policy shifts, even when the presidency stays in the same political party. Rulemaking settlement presents two interesting substantive problems that can arise during presidential transitions: an outgoing administration may use settlement to commit a new administration to a policy course, or an incoming administration may use settlement to undermine regulatory actions adopted by the outgoing administration. Presidential transitions thus illustrate the larger problem with using a settlement to effectuate an agency's shift to a new substantive policy position.
A. Settlement by Outgoing Administrations
To administrative law scholars, the controversies posed by settlements during presidential transitions are not new. Much to the chagrin of new presidents, outgoing administrations occasionally have settled cases in ways that limit the discretion of the incoming administration. For example, in the closing months of the Carter administration, the Environmental Protection Agency (EPA) settled several cases with environmental groups that had challenged the EPA for failure to enforce the Federal Water Pollution Control Act (now known as the Clean Water Act), charging the EPA with failure to meet statutory deadlines and for consideration of improper criteria.65 Over the objections of three intervenors in the litigation, the EPA agreed to a settlement amounting to "a detailed, comprehensive regulatory program for implementation by EPA of the toxic pollutant control and pretreatment objectives" of the Clean Water Act.66 The EPA settlement did not dictate the substantive terms of the regulation; it did, however, specify the general strategy the EPA would pur- [*pg 1034] sue in regulating toxins.67 In the consent decree, the EPA also targeted particular industries and specific pollutants for investigation and possible regulation, specifying deadlines by which regulations would be issued and compliance demanded.68
After remanding the settlement for a determination that the EPA had not improperly curtailed its regulatory discretion, the District of Columbia Circuit upheld the settlement as it had been challenged and affirmed by a district court.69 On remand, the district court held that the EPA had discretion to make such commitments with respect to its regulatory programs,70 and the District of Columbia Circuit agreed on appeal again.71 Although the consent decree required the agency to apply criteria and standards not found in the Clean Water Act, and also required the agency to undertake programs not required by the statute, the District of Columbia Circuit panel upheld the agreement and did not find this settlement to be an impermissible interference with the agency's discretion.72
In a vigorous dissent, Judge Wilkey warned of the "evil[s] of consent decree[s]," including what he believed to be their "potential to freeze the regulatory processes of representative democracy."73 Judge Wilkey reasoned that the court did not have the jurisdiction to impose in a consent decree terms that it could not include in a judgment [*pg 1035] if the case were to go to trial.74 Academic commentary echoed these concerns.75 Attorney General Meese gave his imprimatur to Judge Wilkey's view in his 1986 "Department Policy Regarding Consent Decrees and Settlement Agreements," which prohibits a government agency from entering into a consent decree "that divests a [government official] of discretion" or "that converts into a mandatory duty the otherwise discretionary authority [of an official] to revise, amend, or promulgate regulations."76
Despite the policy announced by Attorney General Meese, agencies have continued to enter into settlements that limit their discretion.77 For example, some pesticide manufacturers have urged [*pg 1036] Christine Todd Whitman, President George W. Bush's EPA administrator, to reject the terms of a January 19, 2001, EPA settlement with the Natural Resources Defense Council.78 The settlement, agreed to in the last days of the Clinton administration, stipulated that the EPA must reevaluate the risks and set tolerances for eleven chemicals that are considered particularly harmful to infants and children, as well as decide the cumulative risks of four families of chemicals.79
Civil litigation jurisprudence would advise treating rulemaking settlements as contracts, deferring to the voluntary nature of the settlement agreement along with an agency's judgment in entering into the settlement. Although this approach raises little concern in the context of voluntary, bilateral agreements, it also has been extended to the enforcement of consent decrees. As the Supreme Court stated in United States v. Armour & Co.:80
Naturally, the agreement reached normally embodies a compromise; in exchange for the savings of cost and elimination of risk, the parties each give up something they might have won had they proceeded with the litigation. Thus the decree itself cannot be said to have a purpose; rather the parties have purposes, generally opposed to each other, and the resultant decree embodies as much of those opposing purposes as the respective parties have the bargaining power and skill to achieve. For these reasons, the scope of a consent decree must be discerned within its four corners, and not by reference to what might satisfy the purposes of one of the parties to it.81
In contexts in which Congress has given private party standing to challenge the agency's actions, the cases acknowledge that agencies need some discretion to manage their litigation resources through settlement. So long as an agency commits itself to a course of action within the scope of its delegated lawmaking powers, this is permissible, and courts often defer to the agency's interpretation of the settlement under Chevron.82
Deference to an agency's interpretation of a settlement's terms subsequent to settlement approval has many advantages. One way to conceptualize such choices in the context of settlement is to understand them as any other agency policy choice. Instead of responding to scientific or other underlying studies, as is normally the case in rulemaking,83 the agency is responding to policy concerns that might encompass the threat of further litigation and the ensuing costs. The agency will be best positioned to make resource decisions regarding strategy during litigation and, as the Court recognized in Armour, settlement reflects a strategic choice in the adversarial litigation context.84 Once parties have agreed to settlement, the Chevron deference standard is desirable from the perspective of promoting effective regulation, decreasing litigation costs, and enhancing the accountability of the agency decisionmaking process.85 For such reasons, the U.S. Court of Appeals for the Third Circuit has stated: "We generally defer to an agency's interpretation of agreements within the scope of the agency's expertise . . . and the case for deference is particularly strong when the agency has interpreted regulatory terms regarding which it must often apply its expertise."86
A court might extend such deference not only to an agency's interpretation of a previously approved settlement, but also to initial [*pg 1038] approval of the settlement.87 To the extent that agencies retain flexibility in the implementation of their settlement terms in subsequent rulemaking proceedings, such settlements will be unlikely to raise significant legitimacy concerns. However, agreeing to a specific regulatory standard in a settlement -- such as a pollution or tolerance level -- raises some obvious procedural problems under the APA. In many cases, including those discussed above, this is not a significant issue since the details of the standards will be established through notice and comment rulemaking following the settlement.
At the same time, to the extent that an agency is using settlement to precommit itself to a substantive policy choice, the notice and comment process may become little more than a charade. Where an agency does not implement the terms of a settlement through notice and comment but instead uses publication rules to implement the terms of a settlement in ways that are not binding, courts will not extend Chevron deference to the settlement.88 Moreover, if deference is too strong, the settlement agreement becomes meaningless, and private parties will have little incentive to settle with administrative agencies.89
Settlement by outgoing administrations illustrates the problem with deferring to settlement merely because it reflects a voluntary agreement among litigants, including the agency. Without standards for careful evaluation of rulemaking settlements, courts may unintentionally bind a new administration or may open the litigation process to strategic manipulation in the months preceding a presidential transition. Chevron, commonly understood to mean deference to the [*pg 1039] agency's law and policy choices, does not always mandate strong judicial deference to settlement in this context.
B. Settlement by Incoming Administrations
Settlement also can be problematic from a legitimacy perspective when used by an incoming administration to modify or repeal adopted regulations. To the extent that a new administration uses the settlement of lawsuits as a housecleaning tool, it may be an important mechanism for implementing new policy positions.
For instance, the transition between President Bill Clinton and President George W. Bush gave rise to charges of "midnight regulation."90 Settlement of challenges to midnight regulations may have been one way for the incoming administration to provide relief from hastily adopted and ill-conceived regulatory programs. At the same time, charges of an unprecedented proliferation of midnight regulations during the waning months of the Clinton administration were greatly exaggerated, leveled largely as a media strategy to discredit the policy approach of the outgoing administration. Midnight regulations often reflect the culmination of a lengthy rulemaking process, a process that is sometimes held up against the agency's wishes for political or budgetary reasons.91 To the extent that recently adopted rules face pending challenges in court, an incoming president's agency might enter into a settlement, enshrining it in a consent decree that promises to notice a repeal or modification to the rules. Such consent decrees also may promise to suspend enforcement of new rules [*pg 1040] pending the outcome of a new rulemaking proceeding, but if such suspension is indefinite a decree is probably subject to attack.
Consider the following example: before Congress's landmark vote to suspend OSHA's ergonomics rules,92 business lobbies, led by the United States Chamber of Commerce, already had filed suit against the rules on the somewhat spurious grounds that the government did not conduct sufficient consultation (despite eight years of public hearings) before issuing the rules.93 If President George W. Bush's administration had attempted to change these "midnight" rules through settlement with the challengers, it is likely that it would have had far more difficulty explaining the legitimacy of its action than arose in the context of congressional disapproval.
To be sure, an agency's discretion to settle is sometimes constrained by law. Courts have held, for instance, that an agency cannot indefinitely suspend rules by agreeing not to enforce them.94 In response to a rule challenge, an agency may enter into a consent decree to propose a new rule. It probably also can suspend applicability of a rule pending the outcome of the new rulemaking process. An altogether more difficult question, however, is whether an agency, as a part of a settlement, can commit itself to adopting or repealing a particular rule. If the appeal is understood as a part of the rulemaking process, an agency probably cannot make such a commitment, to the extent that it renders the notice and comment a charade. An agency can agree, however, to the substance of a proposed rule -- at least to the extent the substance of the rule is a matter of agency discretion -- although the agency's commitment probably is not enforceable by a court.95
So, looking to OSHA's ergonomics rules, had Congress not provided relief to the rules' challengers, the Department of Labor might [*pg 1041] have used settlement of the lawsuit as a means to re-notice the rules. Absent the lawsuit, however, the Department of Labor could have re-noticed the rules, proposing repeal, anyway. The lawsuit challenging the ergonomics rules would have provided the department a rationale for suspending enforcement of the rules pending the outcome of a new rulemaking to repeal or modify the rules. That new rulemaking, though, would have been subject to APA standards of review once final rules were adopted, and labor interests almost certainly would have sought to raise arbitrariness arguments on appeal if the result of the new rulemaking was repeal of the ergonomics rules.
Such settlement is not limited to hypotheticals. It has been used regularly by incoming administrations. One particularly high-profile settlement involved the Department of Interior's recent decision to prepare a supplemental environmental impact statement studying the impact of snowmobile and other winter-vehicle use in Yellowstone Park, made in the face of a pending appeal. This decision, approved by a federal judge in Wyoming, was seen as a first step in George W. Bush's reversal of the Clinton administration's ban on snowmobile use in Yellowstone Park.96 Environmental groups, concerned that objections to the data behind the snowmobile ban were driven not by process but by dissatisfaction with the outcome, raised considerable objection to the decision.97
In another example, commencing shortly after President George W. Bush's inauguration, the United States Department of Agriculture entered into a settlement with hog producers who had challenged an adopted (but not yet published) agency rule.98 The agency's settlement reversed a Clinton administration rule that would end the pork promotion program, a program known for advertising "the other white meat" with funds collected from a tax on independent hog farmers. The rule had been adopted following a referendum by the farmers to end the program.99 The referendum results announced [*pg 1042] January 11, 2001, showed that a majority of producers and importers who voted favored termination (15,951-14,396).100 Shortly after the agency's announcement, on January 12, 2001, the National Pork Producers Council, along with the Michigan Pork Producers Association, filed a lawsuit seeking an injunction against the agency in the United States District Court for the Western District of Michigan.101 The challenge alleged that the agency did not have the authority to call for a referendum and that the voting process was riddled with irregularities.102 Pursuant to a temporary restraining order, the agency was prohibited from publishing a final rule to terminate the Pork Promotion, Research, and Consumer Information Order.103 A subsequent order by the judge, drawing on a memorandum from Andrew Card, President Bush's chief of staff, allowed for publication of a final rule pending review by Bush administration officials but required that the effective date of the rule be delayed for at least sixty days.104 The agency followed by entering into a settlement with the challengers.105
The pork producers' settlement reflects a major shift in the agency's substantive policy position on the legitimacy of the referendum underlying the adopted rule. In agreeing to the settlement, the agency reversed its prior position, developed through notice and comment. There was no notice and comment for withdrawal of the rule ending the pork promotion program. At a minimum, if the agency relied on information of voting irregularities in the referendum, cases like Portland Cement Ass'n v. Ruckelshaus106 would seem to require the agency to subject its new methods for determining the legitimacy of a referendum to notice and comment.107
But, the agency may argue -- as incoming administrations often do in the case of an outgoing administration's settlements -- that this is an agency's new legal interpretation, subject to Chevron deference. If the basis for the settlement is a legal issue subject to Chevron deference, however, there is a strong argument that it should have been developed through notice and comment rulemaking -- especially since it reverses an agency interpretation that previously was adopted through notice and comment rulemaking. To be sure, one may argue that legal interpretations adopted in briefs on appeal in defense of agency interpretations may be subject to Chevron deference,108 but a shift in agency legal interpretation adopted in the course of litigation or through non-notice and comment procedures should not be used as a basis for sub rosa repeal or modification of a rule that has been adopted through notice and comment. Such shifting legal interpretations risk turning the notice and comment process into little more than a sham.
| |
III. CLOSING THE PRINCIPAL-AGENT GAP IN RULEMAKING SETTLEMENT
Agency concession on substantive policy issues in the course of an appellate settlement raises serious legitimacy concerns, especially when it occurs in the context of presidential transitions. In the context of rulemaking settlements, there is a potential that negotiations will be limited to a small number of parties who are not carefully selected for inclusion. Since settlement can occur outside the realm of procedural protections afforded by the APA, an agency can readily make concessions that run contrary to the views of excluded stakeholders. To the extent that an agency uses settlement to commit itself to a substantive policy position, there is no guarantee that this position will be consistent with an agency position developed earlier under APA procedures. If courts merely rubber-stamp settlements, they will contribute to the principal-agent gap in monitoring the accountability of agency rulemaking.
Yet, under existing doctrine, the exact role that courts should play in evaluating settlements and their implementation is unclear, leaving much to judicial discretion. In terms of mood, courts can take a deferential approach and treat settlements as any other contract, or may decide to invoke more rigorous arbitrary and capricious review in evaluating the merits of settlements. A deferential approach places a strong emphasis on exit interests as a monitoring device. My argument is that, although protecting participatory rights of exit (in the form of a party's right to object to a settlement) is necessary, a focus on voice and loyalty concerns over exit interests also will serve an important function in enhancing the accountability of rulemaking settlement. Courts should not rely on exit incentives alone to protect the public interest during rulemaking settlement. They also must play a role in striking a balance between voice and loyalty as distinct ways of holding the agency's policy choices to account.109
In terms of timing, courts may evaluate settlements ex ante -- at the time of the initial settlement -- or ex post -- in reviewing final agency action implementing a settlement. As I argue in this Part, the [*pg 1045] potential for accountability problems can be minimized if courts carefully evaluate participation in settlement agreements where settlement involves an agency's commitment to pursuing a substantive policy position, respecting voice as well as exit in this context. Although necessary, monitoring settlement participation is hardly sufficient. To enhance accountability, courts also should pay attention to loyalty influences by applying APA review standards, including arbitrary and capricious review, to review the merits of settlements ex ante, at the approval stage.
A. Monitoring Settlement Participation
In ex post review of repeals or modifications to regulations implementing settlements, broad opportunities for intervention are commonplace, allowing for collateral attack of the final rule even by groups who were not necessarily parties to an initial settlement. In evaluating the various types of participation in rulemaking settlement, it is important to distinguish exit interests from voice interests. Exit interests represent the rights of parties to a proceeding, or those who are aware of a pending proceeding and able to intervene, not to agree to the settlement. By failing to agree to a settlement, a party effectively exits the consensual governance mechanism by continuing with litigation.
Of course, protecting exit interests is the most fundamental type of participation necessary to ensure the legitimacy of a settlement. Civil class action settlements provide a particularly useful analogy in understanding the value of exit interests in assisting courts in monitoring settlements. In the context of class action settlements, a class member who has been inadequately represented may attack the class judgment in subsequent litigation.110 Martin v. Wilks111 clarified the question, holding that res judicata does not require absent class members to object to inadequate representation in class proceedings.112 In that well-chronicled case, a group of white firefighters had sued Birmingham, Alabama, alleging that an affirmative action con- [*pg 1046] sent decree violated Title VII of the Civil Rights Act of 1964.113 The white firefighters had not intervened in the suit that had been resolved by a consent decree, so the defendants in Martin argued that they were bound by the terms of the consent decree.114 The Court held, however, that even white firefighters who were on notice of the opportunity to intervene in the earlier proceeding were not bound because they were not parties.115 According to the Court, the rules of civil procedure make "[j]oinder as a party, rather than knowledge of a lawsuit and an opportunity to intervene . . . the method by which potential parties are subjected to the jurisdiction of the court and bound by a judgment or decree."116 Similarly, the failure of a stakeholder to object to a settlement on appeal should not preclude its later ability to attack the settlement collaterally, as may occur ex post in appeal of a final rule implementing a settlement.117
Martin v. Wilks involved a type of exit interest -- an effort by potential intervenors to object to unacceptable terms of a settlement and to litigate outside of the context of the consensus reflected in the settlement. Just as in the context of civil class actions, those parties to an appeal who are on notice of a settlement must retain the ability to attack the settlement collaterally. Protecting participatory rights of exit -- in the form of a party's right to object to the terms of settlement -- is a necessary form of participation in promoting the accountability of rulemaking settlement. If a key stakeholder objects to the settlement's terms, that stakeholder can continue to litigate the agency's initial action as a party or can attack the settlement in a separate proceeding.
In the administrative law context, however, if a settlement has consequences beyond the immediate parties, such as influence on an [*pg 1047] agency interpretation or policy regarding the public interest, a focus on exit interests alone may not be sufficient to ensure institutional accountability. For example, as illustrated here, key stakeholders may lack party status, or may not know the terms of a settlement. Exit interests serve an important role in settlement of administrative rule appeals by creating a strong incentive for an agency to make concessions to certain litigants in the appellate context. At the same time, the prospect of exit from a settlement may encourage an agency to keep its settlement confidential, negotiate with only a small number of parties, or limit the scope of parties present in an appeal. In this sense, exit interests may widen the principal-agent gap in the settlement of appeals involving administrative agencies. In the context of appeals, exit interests can have a perverse effect on agency litigant behavior, creating a tension with the interests of other stakeholders who might hold the agency accountable for its negotiation.
Thus, beyond protecting exit interests, some attention also must be given to the voice interests of affected stakeholders. Voice interests differ from exit interests in that they are not limited to those who already are actual parties to an appeal, but encompass the possible universe of stakeholders who may participate in settlement proceedings. Attention to these interests broadens the universe beyond actual parties and those who are on actual notice of the settlement to include potential stakeholders who may not even know that a settlement has been agreed to or that an appellate proceeding is pending.
As discussed above, limited participation in appellate litigation allows an agency to commit itself to a course of action in settlement that is not necessarily consistent with the public interest. Especially since a settlement and its approval are not governed by the APA's procedural requirements, including APA ex parte contact requirements, expansive participation in settlement negotiations and broad intervention for purposes of judicial challenge of settlements can serve an important function in preserving the public interest. For judicial review to be effective, ex ante opportunities for participation in the initial settlement negotiations, as well as for judicial challenge to a settlement, also must be expansive. Unless an agency's regulatory process specifically provides for published notice of settlement proceedings involving rule appeals, participation in settlement negotiations typically will be limited to those parties present on appeal.
In the context of ordinary civil litigation, courts often accept the agreement of those present, because potential parties have incentives to participate when there is truly something at stake, typically money. [*pg 1048] In the context of rulemaking settlement, though, a heterogeneous range of interests -- such as workplace safety, the environment, diversity, the quality of life, or other non-monetary preferences -- coupled with less salient incentives for widespread private participation in appeals may make reliance on voluntary participation less reliable. In appeals of agency regulations, the interests of possible stakeholders are not nearly as homogenous as those quibbling over the amount of a monetary judgment in the typical civil tort suit. Certainly, some stakeholders may be concerned with the costs of compliance or the monetary benefits associated with a regulatory program. But the range of interests in the typical regulatory proceeding is far broader and more diverse. Interest groups may raise concerns with workplace safety, the environment, diversity, the quality of life, or other non-monetary preferences -- interests that also do not always create particularly salient incentives for participation in an appeal. Another factor thwarting participation in settlements is an informational imbalance. In many instances, interest groups may not even be aware of a serious judicial challenge or where that challenge has been mounted, let alone that settlement negotiations have commenced on topics potentially affecting the substance of a regulation.
In contrast to the range of interests at issue in a rulemaking settlement, consider the identity of the potential participants. Although the potential variety of interests at stake in a rulemaking settlement may be problematic, the identity of potential participants is more easily ascertainable than in, for instance, mass civil tort litigation. Since the notice and comment rulemaking process already has run its course by the time a regulation reaches the appellate posture, notice of the settlement and an opportunity to comment or participate could readily and fairly be limited to participants in the previous rulemaking proceeding.118 In monitoring settlement negotiations, courts may [*pg 1049] need to referee whether parties have been afforded adequate notice of a settlement. However, active judicial supervision is not required to ensure the airing of a settlement among the rulemaking's participants. Since regulations already have been generated through the notice and comment process, a court readily can consider the viewpoints of those who commented previously on the proposed rule and are likely to be interested in the settlement. To ensure that key participants in the notice and comment process are not ignored in settlement, a court might even require the agency itself to provide this type of opportunity for participation as a condition of approval of any settlement in appeals of substantive notice and comment rules.
Once a negotiation has run its course and settlement is reached, an opportunity for ex ante judicial challenges can assist in narrowing the principal-agent gap and increase accountability. Broad opportunities for such challenges seem uncontroversial to the extent that an agency uses the settlement process to effectively eviscerate or modify terms previously adopted through notice and comment rulemaking, especially to the extent that the implementation of the settlement will not necessarily result in final agency action subject to APA review. If settlement terms trigger interpretive rules and policy statements that modify the nature of a previously adopted regulation, broad ex ante participation that raises concerns with the settlement will help to ensure that the settlement is subject to the same opportunities for review afforded APA-promulgated interpretive rules or policy statements. It also will assist courts in evaluating the extent to which regulated stakeholders are able to collude with one or two other stakeholders in securing a settlement that would be unacceptable to other key stakeholders, if they too were present as primary parties to the appeal.119
B. Ex Ante Hard Look Review of the Merits of Settlements
A settlement of an appeal challenging a regulation risks widening the principal-agent gap in regulatory decisions. Since public participation in the rulemaking settlement process may be narrower and more secretive than that occurring through normal APA procedures, an agency can make concessions that are not acceptable to key principals, whether they be discrete and identifiable interest groups or the more amorphous "public interest." Indeed, an agency faced with the prospect of delayed rule enforcement, litigation costs, and protracted appeals may have strong incentives to settle with the parties on appeal even if the settlement is contrary to rules adopted through far broader and public notice and comment rulemaking. Because of those same incentives, agencies cannot be depended upon to advocate for those principals and interests that are not part of the settlement bargain when they are arguing in favor of judicial acceptance of a consent decree. Even where all key stakeholders are present, judicial deference to the bargain struck during a settlement invites an agency to ignore the law, the facts, and the public interest.120
Because the prospect of exit poses the threat of continued litigation, some monitoring of voice interests, both in participation in settlement negotiations and during judicial evaluation, is necessary to protect the public interest. Respecting voice through participation alone, however, is not sufficient to protect the public interest in the negotiation and approval of settlements. To expose the potential "blind spot" of rulemaking settlement, APA standards of review also must apply to judicial endorsement of a settlement. For judicial review to be effective, a court should not limit the applicability of APA standards to ex post evaluation, as the reviewing court scrutinizes final agency action implementing a settlement. Judicial review of the merits also must reach back ex ante, to settlement negotiation and settlement approval. In addition to exit and voice interests, loyalty interests -- the agency's fidelity to its statutory mandate and other key principals not present on appeal -- also must play a role in holding settlements to account. To the extent courts can use ex ante review to monitor the fidelity of agency concessions during settlement to key principals who are not necessarily present on appeal, judicial monitoring of settlements also can further loyalty interests.
To be sure, administrative agencies usually are considered effective institutions for protecting the public interest in the implementation of regulatory decisions. To a large extent, gaps in the principal-agent relationship between interest groups and the public interest, on the one hand, and an administrative agency, on the other, are narrowed through ex post judicial review of administrative regulations. When an agency's regulations are reviewed after adoption, ex post judicial review, coupled with political oversight such as congressional or executive review, helps to protect loyalty interests.
Such review, expressly provided for in the APA,121 helps to ensure that the public interest is not skirted through a rulemaking settlement and its implementation. Ex post review -- including a look back at the settlement after full implementation in regulation or some other agency action -- can be an effective vehicle for narrowing the principal-agent gap. On appeal, adopted regulations, as well as interpretive rules and policy statements, will be subject to review for their arbitrariness and capriciousness.
In Motor Vehicle Manufacturers Ass'n v. State Farm Mutual Automobile Insurance Co.,122 the Supreme Court articulated the test for ex post review of adopted regulations, especially where there is a change from a previously adopted agency position. The case arose when the National Highway Traffic Safety Administration (NHTSA) decided to rescind a previously adopted regulation that would have mandated the inclusion of passive restraints on new cars, on the ground that the safety benefits of the rule did not justify its compliance costs for automobile manufacturers.123 NHTSA's decision, a small part of President Ronald Reagan's implementation of a broader regulatory relief agenda, followed years of dispute regarding the standard for automobile air bags.124
As the Court stated in State Farm, an agency rule is arbitrary and capricious if in adopting it
the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not [*pg 1052] be ascribed to a difference in view or the product of agency expertise.125
Under this standard, the Court rejected NHTSA's rule rescinding the passive restraint standard as arbitrary and capricious. The Court held that the agency had failed to explain its basis for completely abandoning the passive restraint rule where record evidence only faulted automatic seatbelts, not all passive restraints.126 The Court not only faulted the agency for failing to consider the alternatives, it also required the agency to explain the factual premises for its actions. In particular, the Court observed, the agency had failed to explain how the rulemaking record supported the agency's premise that drivers and passengers would, in fact, disable detachable automatic seatbelts.127 In elaboration, the Court held that an agency must consider the factors that Congress intended it to consider in making its decisions. In State Farm, the Court reasoned, "Congress intended safety to be the pre-eminent factor" in NHTSA's adoption of standards for automobile design.128
The judicial demand for reasoned decisionmaking articulated in State Farm works to narrow the principal-agent gap in an agency's policy choices. For example, one can understand the safety emphasis in the statute as a way for the political coalition that prevailed in Congress at the time of the statute's adoption to ensure NHTSA's compliance with congressional will. As positive political theorists observe, substantive statutory requirements coupled with evidentiary and procedural requirements, "create pressures on [the] agenc[y] that replicate the political pressures applied when the relevant legislation was enacted," working to "enhance the durability of the bargain struck among members of the coalition."129 So understood, hard look review is designed to minimize agency infidelities to key principals, minimizing the agent cost problem in administrative law.130 That is, [*pg 1053] hard look review serves loyalty interests by helping to hold agencies accountable to key principals for agency decisions.131
For similar reasons, Professor Richard Nagareda urges hard look review of settlements in mass tort cases. Professor Nagareda observes two agent cost problems with mass tort settlements. First, private counsel wield "sweeping power" through such agreements, without necessarily claiming valid authority to represent the interests of every member of the class.132 Second, as Judge Posner has observed, in the class action context, "the negotiator on the plaintiffs' side, that is, the lawyer for the class, is potentially an unreliable agent of his principals."133 To help guard against the accountability problems that these agent costs create, Professor Nagareda urges that courts apply the hard look test to approval of class action settlements in mass torts under Rule 23 of the Federal Rules of Civil Procedure.134 As he states, [*pg 1054] "It would be far more difficult for class counsel to advance their own interests at the expense of the class if counsel knew that in order to obtain judicial approval they must explain their decisions in light of contrary information."135 There may be some advantages to applying the hard look standard in this context, although it is not exactly clear what such a standard would entail in the context of evaluating a settlement, where in many instances the policy issues are few.
State Farm sets the bar for modifications of adopted regulations by agencies. As State Farm reminds, an agency cannot commit to repeal a rule without going through the notice and comment process and providing reasons -- apart from raw executive branch politics -- for a change in the agency's position.136 The dissenters in State Farm, including two current members of the Court (Chief Justice Rehnquist and Justice O'Connor), stated that a "change in administration brought about by the people casting their votes is a perfectly reasonable basis for an executive agency's reappraisal of the costs and benefits of its programs and regulations."137 The State Farm majority, however, rejected the position that "substantial uncertainty" that a regulation will accomplish its purpose is sufficient reason for rescinding a regulation.138 State Farm limits the nature of an agency's repeal or modification mechanisms, encouraging the agency to use reasons beyond political pressure in the executive branch.
At a minimum, as the APA provides, ex post review of settlements, at the enforcement or final rules appeal stage, is necessary to protect the public interest. But given agency opportunities for skirting administrative procedure in the settlement process, ex post review is not sufficient to protect the public interest. In contrast to class action settlements, which are only approved to the extent that they are fair, just, and reasonable,139 the judicial role in approving consent decrees and other settlements has been more limited. In many cases outside of the class action context, courts have been described as "rubber- [*pg 1055] stamping" consent decrees or settlements.140 However, as case law establishes, "[a] judge is not obligated to approve any settlement that is put forth by the parties."141 Where "the interests of individuals and organizations other than those approving the settlement may be implicated[,] [t]he presence of these other interests prevents . . . the trial court [] from taking a totally 'hands-off' attitude toward the settlement reached."142 Courts have held that "[s]ettlements are void against public policy . . . if they directly contravene a state or federal statute or policy."143
Certainly, ex post review is based on a far more extensive record than review ex ante -- at the time of a settlement's approval -- thus providing courts an opportunity to gauge the full process rather than assess piecemeal a small corner of a policy decision, as courts may ex ante. If all an agency faces is ex post review, however, there is a risk of premature, even unknown or unknowable, agency commitment to substantive policy positions that can influence the content of a final rule without necessarily becoming a part of the rulemaking record. Moreover, in some instances, settlement may never culminate in final agency action, or an agency may adopt what are effectively interpretive rules or policy statements in the course of settlement, without necessarily taking these through APA procedures or subjecting them to APA review.
Just as a glance into the rearview mirror will not alert the automobile driver to potential blind spots on the highway, a backward glance at the merits of settlement will not always suffice to expose the blind spots associated with rulemaking settlement. A direct glance at the settlement's merits and the circumstances surrounding it at the time of initial approval of the settlement or stay of an agency's action also is necessary. For example, in approving stays of agency regulatory action pending review or implementation of a settlement under the APA, assessment of the public interest serves to guard against the [*pg 1056] use of stays to undermine the regulatory process.144 Even where a stay is supported by the agency, independent judicial assessment of the public interest in granting the stay serves an important role in ensuring that stays do not undermine the regulatory process.
More broadly, ex ante review of a settlement's merits guards against the possibility that an agency will adopt a position in a settlement without some consideration of contrary views among principals, including principals such as Congress and interested stakeholders that are not present on appeal but who may have commented previously in the initial rulemaking. Ex ante review also would improve the quality of private participation in rulemaking settlement, since private parties will be less likely to view settlement opportunities as an occasion for private bargaining with the agency. Ex ante review further minimizes the risk of an agency's wasting resources if a settlement commits an agency to a course of action that later is rejected on appeal. The APA does not expressly provide for ex ante review of settlements and their terms outside of the context of judicial stays,145 suggesting that the textual support for such review is shaky. However, because a court applying ex ante review already has jurisdiction to hear the appeal over a previously adopted regulation, it is not controversial as a procedural matter for the courts to apply the same standard in reviewing the settlements that it would apply in substantive review of the challenged rule.146 The record for reviewing the settlement presumably would include the full appellate record on appeal, as well as any submissions by the agency and stakeholders in support of the settlement.
Even with ex ante review of the merits of a settlement, an agency still has broad authority to make concessions in the course of settlements on appeal. One way for an agency to do so is through an interpretative rule or policy statement, which agencies routinely adopt without notice and comment rulemaking. An interpretation endorsed in the course of a settlement agreement, however, ought to be subject to the same challenges as an interpretive rule adopted outside of the settlement context. Although the record for review may be a narrow one, subjecting the interpretation to ex ante hard look review prior to judicial approval of a settlement will create strong incentives for an agency to provide a robust explanation for the interpretation.
In sum, ex post review alone will not be sufficient to protect loyalty values in rulemaking settlement but must be coupled with ex ante review. Courts can apply this recommendation by extending the arbitrary and capricious test to settlement approval or to consent decrees. Outside of the context of staying agency action, the statutory basis for courts exercising such review, which the APA only provides for "final" agency action, is not express, but such review has strong potential to hold rulemaking settlement to account and is consistent with the procedural roles that courts play in reviewing agency regulations.
| |
CONCLUSION
Private stakeholders see value in rulemaking settlement to the extent that it can be used to bind an agency. The binding nature of settlement thus works as an important incentive for private litigants to engage in this strategy, rather than to litigate, when contesting an agency regulation.147 It is important to note, however, that settlements that bind the government can affect the public interest in ways that ordinary civil settlements do not.
Since private stakeholder bargaining in rulemaking settlements often takes place outside of the formal legal protections of administrative process, an agency may find it expedient to give short shrift to the public interest in order to settle lawsuits. Courts and litigants should not treat settlement of substantive challenges to rules as mere private contracts. Nor should strong deference to agency interpretations necessarily be the norm. Such a judicial approach heavily favors [*pg 1058] exit interests over other concerns such as voice and loyalty. Exit, voice, and loyalty interests each must play a part in the evaluation of settlements. Voice and loyalty interests, in particular, may be in tension at times, but this is the same tension with which courts struggle elsewhere in defining participation for administrative law.148
Due to a principal-agent gap that fails to protect the public interest in administrative agency concessions during settlement, some ex ante judicial role in policing the public interest is appropriate. Simultaneously, courts must evaluate carefully participation in such settlements, allowing broad collateral opportunities for challenge to ensure that key stakeholder perspectives are not omitted. With a balanced approach to monitoring participation and reviewing the merits of settlements, courts can help avoid the potential for undermining the public interest while also preserving incentives for private stakeholder participation in settlements of rule appeals.
| |
FOOTNOTES
Copyright © 2001 by Jim Rossi.
| † | Patricia A. Dore Associate Professor of State Administrative Procedure, Florida State University College of Law. This Essay is based on a paper presented at the Duke Law Journal's Administrative Law Conference held at Duke University School of Law on March 5, 2001. The Duke University School of Law's Program in Public Law provided additional financial support for the 2001 conference. Thanks to Liz Fisher, Adam Hirsch, Harold Krent, Richard Nagareda, Peter Oh, and Ansley Samson for their comments on a draft, and to Amy Comer for her dedicated research assistance. |
| 1. | Conceptually, the tension revealed in this debate is not new, as it reflects the longstanding effort to reconcile public and private interests in the administrative state. See Louis L. Jaffe, Law Making by Private Groups, 51 HARV. L. REV. 201, 203 (1937) (observing that powers exercised by majorities of groups not only often are binding on the entire group, but also indirectly affect the rest of the public). |
| 2. | See, e.g., IAN AYRES & JOHN BRAITHWAITE, RESPONSIVE REGULATION: TRAN- SCENDING THE DEREGULATION DEBATE 4 (1992) (proposing that the interplay between private and public regulation creates the best public policy results, as opposed to either strict regulation or total deregulation); Jody Freeman, Collaborative Governance in the Administrative State, 45 UCLA L. REV. 1, 22 (1997) (embracing a normative model of collaborative governance); Mark Seidenfeld, Empowering Stakeholders: Limits on Collaboration as the Basis for Flexible Regulation, 41 WM. & MARY L. REV. 411, 445 (2000) (eyeing skeptically a widespread shift in favor of collaborative approaches that empower stakeholders). |
| 3. | See infra notes 11, 18, 22, 23 (identifying sources discussing these alternatives). Of course, not all of this commentary praises these innovations. |
| 4. | Notable studies of the phenomenon that struggle with understanding settlements against the backdrop of agency discretion include Elizabeth Fisher & Patrick Schmidt, Seeing the "Blind Spots" in Administrative Law: Theory, Practice, and Rulemaking Settlements in the United States, 37 COMMON L. WORLD REV. 272 (2001); Peter M. Shane, Federal Policy Making by Consent Decree: An Analysis of Agency and Judicial Discretion, 1987 U. CHI. LEGAL F. 241. Many other studies express awareness of, but do not fully address, the phenomenon of rulemaking settlements. See, e.g., Patricia M. Wald, Negotiation of Environmental Disputes: A New Role for Courts?, 10 COLUM. J. ENVTL. L. 1, 17 (1985) (observing the shift in environmental policymaking toward collaborative efforts not involving the courts, and thereby away from traditional judicial oversight). Wald also proposes a new role for the courts in the wake of this shift-a role in negotiated regulations. Id. |
| 5. | See infra notes 26-27 and accompanying text. |
| 6. | Fisher & Schmidt, supra note 4, at 273. |
| 7. | For discussions of these types of settlements, see Jeffrey M. Gaba, Informal Rulemaking by Settlement Agreement, 73 GEO. L.J. 1241, 1243-48 (1985) (examining the differences between scheduling agreements, process agreements, and substantive agreements). |
| 8. | See, e.g., Michael Fix & George C. Eads, The Prospects for Regulatory Reform: The Legacy of Reagan's First Term, 2 YALE J. ON REG. 293, 308 (1985) (characterizing negotiated regulation as "pioneering work" and an "innovation"); Richard B. Stewart, The Discontents of Legalism: Interest Group Relations in Administrative Regulation, 1985 WIS. L. REV. 655, 682 (asserting that dysfunctions in a legalistic approach to regulation can be ameliorated by encouraging negotiation). |
| 9. | As of August 1, 2000, over twenty-nine federal agencies and agency components had published proposed or final alternative dispute resolution policies or implemented pilot alternative dispute resolution programs. Marshall J. Breger, The Administrative Dispute Resolution Act of 1996 and the Private Practitioner, in FEDERAL ADMINISTRATIVE DISPUTE RESOLUTION DESKBOOK 1, 10-11 (2001). The most prominent of these programs is the Department of Labor's mediation program. Id. |
| 10. | Even in the context of purely private, bilateral settlements, however, commentators have raised concerns about the protection of public values. See, e.g., Owen M. Fiss, Against Settlement, 93 YALE L.J. 1073, 1075 (1984) (suggesting that settlements are at best "a capitulation to the conditions of mass society" and "should be neither encouraged nor praised"); David Luban, Settlements and the Erosion of the Public Realm, 83 GEO. L.J. 2619, 2652-53 (1995) (decrying in particular "secret settlements" and their threat to public values). |
| 11. | For example, when OSHA finds a violation of its regulations, it often settles it for one-half or less of the assessed penalty. Given scarce inspection resources and limitations on OSHA's fining authority, this creates less than optimal incentives for compliance with workplace safety measures. Sidney A. Shapiro & Randy Rabinowitz, Voluntary Regulatory Compliance in Theory and Practice: The Case of OSHA, 52 ADMIN. L. REV. 97, 108-09 (2000). |
| 12. | 5 U.S.C. §§ 551-559, 701-706, 1305, 3105, 3344, 4301, 5335, 5372, 7521 (1994 and Supp. IV 1998). |
| 13. | Id. § 553(b)-(c). |
| 14. | Id. § 706. As a component of such review, courts sometimes invoke the hard look test, which is discussed infra at Part III.B. |
| 15. | See, e.g., United States v. Students Challenging Regulatory Agency Procedures, 412 U.S. 669, 686 (1973) (determining that standing should not be denied solely on the basis that many people are injured); Ass'n of Data Processing Serv. Orgs., Inc. v. Camp, 397 U.S. 150, 152 (1970) (noting that petitioners who sold data processing services had standing under a theory of "injury in fact"); Office of Communication of the United Church of Christ v. FCC, 359 F.2d 994, 1000 (D.C. Cir. 1966) (demonstrating that "courts have resolved questions of standing as they arose and have at no time manifested an intent to make economic interest and electrical interference the exclusive grounds for standing"). The rulemaking process depends on broad-based participation for its legitimacy. For a discussion of this theory, see Jim Rossi, Participation Run Amok: The Costs of Mass Participation for Deliberative Agency Decisionmaking, 92 NW. U. L. REV. 173, 221-46 (1997). |
| 16. | See Abbott Labs. v. Gardner, 387 U.S. 136, 153 (1967) (permitting pre-enforcement review "where a regulation requires an immediate and significant change in the plaintiffs' conduct of their affairs with serious penalties attached to noncompliance"). |
| 17. | See Robert A. Kagan, Adversarial Legalism and American Government, 10 J. POL'Y ANALYSIS & MGMT. 369, 375-79 (1991) (proposing that adversarial legalism creates enormous costs to parties). |
| 18. | See CORNELIUS KERWIN, RULEMAKING: HOW GOVERNMENT AGENCIES WRITE LAW AND MAKE POLICY 246 (2d ed. 1999) (noting that the "frequency of lawsuits that challenge rules in certain agencies is breathtaking"); Cary Coglianese, Assessing Consensus: The Promise and Performance of Negotiated Rulemaking, 46 DUKE L.J. 1255, 1298-1301 (1997) (observing that the percentage of rules subject to judicial review ranges from twenty-six to eighty percent of all rules promulgated). |
| 19. | See, e.g., Thomas O. McGarity, Some Thoughts on Deossifying the Rulemaking Process, 41 DUKE L.J. 1385, 1396 (1992) (explaining that ossification is due in part to underfunding of agencies, but is due primarily to "underlying inertial forces" embedded in the process of informal rulemaking itself). |
| 20. | Negotiated regulation draws inspiration from the alternative dispute resolution movement, which favors negotiation over more litigious approaches to resolving disputes. For a discussion of the benefits and barriers to negotiation, see ROBERT H. MNOOKIN ET AL., BEYOND WINNING: NEGOTIATING TO CREATE VALUE IN DEALS AND DISPUTES 9-19 (2000); Joel Cutcher-Gershenfeld & Michael Watkins, Toward a Theory of Representation in Negotiation, in NEGOTIATING ON BEHALF OF OTHERS 23, 23-51 (Robert H. Mnookin & Lawrence E. Susskind eds., 1999). |
| 21. | For examples, see infra note 23 and accompanying text. |
| 22. | Well-known critiques of the implementation process include William Funk's case study of Environmental Protection Agency (EPA) negotiated regulation and Susan Rose-Ackerman's conceptual exploration of the tension between incentives and consensus. William Funk, When Smoke Gets in Your Eyes: Regulatory Negotiation and the Public Interest-EPA's Woodstove Standards, 18 ENVTL. L. 55 (1987) [hereinafter Funk, Woodstove Standards]; Susan Rose-Ackerman, Comment, Consensus Versus Incentives: A Skeptical Look at Regulatory Negotiation, 43 DUKE L.J. 1206 (1994); see also William Funk, Bargaining Toward the New Millenium: Regulatory Negotiation and the Subversion of the Public Interest, 46 DUKE L.J. 1351, 1387 (1997) [hereinafter Funk, Subversion] (concluding that "the incentives to make negotiated rulemaking succeed themselves undermine and subvert the principles underlying traditional administrative law by elevating the importance of consensus among the parties above the law, the facts, or the public interest"). |
| 23. | See Philip J. Harter, Negotiating Regulations: A Cure for Malaise, 71 GEO. L.J. 1, 28-31 (1982) (offering a classic defense of negotiated regulation); see also Deborah S. Dalton, Negotiated Rulemaking Changes EPA Culture, in FEDERAL ADMINISTRATIVE DISPUTE RESOLUTION DESKBOOK 135, 146 (2001) (noting that prior to the EPA's use of negotiated regulation, rulemaking in the agency was under fire). |
| 24. | See Funk, Woodstove Standards, supra note 22, at 80-81 (noting that groups with stakes in woodstove regulation put pressure on the EPA); Rose-Ackerman, supra note 22, at 1216-17 (encouraging an examination of the procedures of private organizations influencing federal policymaking). Coglianese also observes cases in which negotiated rules were subject to judicial challenge, sometimes by those who participated in negotiations. Coglianese, supra note 18, at 1286-1309. A comprehensive critique of negotiated regulation is presented in Funk, Subversion, supra note 22. |
| 25. | Funk, Woodstove Standards, supra note 22, at 80-81. |
| 26. | Coglianese calculates that negotiated regulation constitutes a mere one-tenth of one percent of all rulemakings. Coglianese, supra note 18, at 1277. Some claim that the appeals rate for rules is as high as 80 percent, but Coglianese presents convincing data that the appeals rate for the most significant EPA rules is 35 percent or even lower. Id. at 1297-1301. Rules still are far more likely to be appealed to the judicial system, and thus pose the opportunity for settlement, than to undergo a regulatory negotiation process prior to adoption. |
| 27. | In his study of EPA rulemaking litigation, Coglianese observes that nearly half of the cases filed between 1979 and 1990 resulted in voluntary dismissals prior to hearing. Cary Coglianese, Litigating Within Relationships: Disputes and Disturbance in the Regulatory Process, 30 LAW & SOC'Y REV. 735, 756 (1996). This is nearly twice the settlement rate for all appeals. Id. As Coglianese notes, "[w]ith so much settlement activity, interest groups sometimes file an action against the EPA just to maintain a seat at the bargaining table." Id. |
| 28. | Gaba, supra note 7, at 1264. In practice, as an enforcement device, parties to a voluntary settlement frequently will request a court to approve the terms of the settlement as a consent decree, providing a judicial imprimatur. Courts are inclined to comply if the settlement has unanimous support, including that of the agency. Alternatively, to effectuate enforcement of a voluntary settlement against the agency, private stakeholders in an appeal may request a stay of the appeal pending agency implementation of the terms of the settlement. Arguably, courts have broader review authority in approving consent decrees than in issuing stays. Id. |
| 29. | Occupational Exposure to Formaldehyde, 29 C.F.R. §§ 1910, 1926.1148 (2001). |
| 30. | Occupational Exposure to Formaldehyde, 52 Fed. Reg. 46,168, 46,168 (Dec. 4, 1987). OSHA's short-term formaldehyde exposure limit was set at 2 ppm. Id. § 1910.1048(c)(2). |
| 31. | For discussion of the appellate litigation and ensuing settlement, see Charles C. Caldart & Nicholas A. Ashford, Negotiation as a Means of Developing and Implementing Environmental and Occupational Health and Safety Policy, 23 HARV. ENVTL. L. REV. 141, 171-73 (1999). |
| 32. | UAW v. Pendergrass, 878 F.2d 389, 400 (D.C. Cir. 1989). |
| 33. | Caldart & Ashford, supra note 31, at 172-73. |
| 34. | See Occupational Exposure to Formaldehyde, 57 Fed. Reg. 22,290 (May 27, 1992) (codified at 29 C.F.R. § 1910.1048) (providing that these standards would take effect on June 26, 1992). |
| 35. | Caldart & Ashford, supra note 31, at 173. |
| 36. | At the same time, however, Funk argues that the notice and comment process alone is not a sufficient safeguard to protect the public interest in every case. Other administrative procedure requirements, such as the preparation of a statement of basis and purpose and the requirement of arbitrary and capricious review, also are necessary to protect the public interest in consensus solutions. Funk, Subversion, supra note 22, at 1371-72, 1374. As Funk observes, consensus may come at the price of the public interest, the facts, or the law, even when an agency utilizes notice and comment to implement the consensus. Id. at 1382-86. More than in the case of rulemaking settlement, ex post review of final rules adopted through notice and comment rulemaking following a consensus proposal helps to assure the protection of these values in the adoption of negotiated regulations. |
| 37. | Shane, supra note 4, at 265. Judge Wilkey of the District of Columbia Circuit has warned of the "evil(s) of government by consent decree," including "its potential to freeze the regulatory processes of representative democracy." Citizens for a Better Env't v. Gorsuch, 718 F.2d 1117, 1136 (D.C. Cir. 1983) (Wilkey, J., dissenting). |
| 38. | For example, in another context, commentators raise concerns with government's changing the terms of the "regulatory compact" governing public utility industries, such as the telecommunications, natural gas, and electric sectors, as firms in these industries are deregulated. J. GREGORY SIDAK & DANIEL F. SPULBER, DEREGULATORY TAKINGS AND THE REGULATORY CONTRACT: THE COMPETITIVE TRANSFORMATION OF NETWORK INDUSTRIES IN THE UNITED STATES 2-17 (1997) (arguing that deregulation can effectuate a taking or constitute breach of contract). Critics, including me, observe that this argument has little basis in takings and government contracts jurisprudence, thwarts flexibility in regulatory decisionmaking, and may lead to perverse results by encouraging private stakeholders to lobby government excessively for concessions that do not enhance the public welfare. E.g., Herbert Hovenkamp, The Takings Clause and Improvident Regulatory Bargains, 108 YALE L.J. 801, 821-28 (1999) (reviewing SIDAK & SPULBER, supra); Susan Rose-Ackerman & Jim Rossi, Disentangling Deregulatory Takings, 86 VA. L. REV. 1435, 1489-93 (2000); Jim Rossi, The Irony of Deregulatory Takings, 77 TEX. L. REV. 297, 298-99 (1998) (reviewing SIDAK & SPULBER, supra). |
| 39. | See generally Michael C. Jensen & William H. Meckling, Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure, 3 J. FIN. ECON. 305, 308-10 (1976) (observing the problems involved in inducing an agent to maximize a principal's welfare). |
| 40. | This distinction is rooted in what is now considered a classic book in corporate and managerial governance, Albert O. Hirschman's, Exit, Voice, and Loyalty Responses to Decline in Firms, Organizations, and States (1970). Others draw on this vocabulary to understand the class action litigation process. See, e.g., John C. Coffee, Jr., Class Action Accountability: Reconciling Exit, Voice, and Loyalty in Representative Litigation, 100 COLUM. L. REV. 370, 376-80 (2000) (observing that reform proposals for complex litigation can be channeled into exit, voice, and loyalty strategies); Samuel Issacharoff, Governance and Legitimacy in the Law of Class Actions, 1999 SUP. CT. REV. 337, 341-42 (characterizing the Court's recent decisions as using a functional approach, rather than an approach embracing notions of exit, voice, or loyalty). |
| 41. | Cognizant of this phenomenon, Coglianese observes that litigation itself plays an important institutional role in defining consensual relationships, and that the litigation process provides a fundamentally different forum for framing consensus discussions. Coglianese, supra note 27, at 761-63. |
| 42. | Id. at 758. |
| 43. | See Funk, Woodstove Standards, supra note 22, at 95 (arguing that an agency's "responsibility for determining the public interest . . . no longer provides the safeguard otherwise available for unrepresented interests"); Rose-Ackerman, supra note 22, at 1211 (observing that "regulatory negotiation is not democratically legitimate unless all interested parties are adequately represented"). |
| 44. | Control of Hazardous Energy Sources (Lockout/Tagout), 29 C.F.R. § 1910.147 (2001). |
| 45. | Patrick Delbert Schmidt, Between Government and Client: The Roles of Lawyers in Negotiating and Disputing Administrative Regulation 205 (2000) (unpublished Ph.D. dissertation, Johns Hopkins University) (on file with the Duke Law Journal). |
| 46. | Id. at 205-06. |
| 47. | Id. at 175-213. |
| 48. | 55 Fed. Reg. 38,677 (Sept. 20, 1990) (codified at 29 C.F.R. § 1910.147). |
| 49. | 5 U.S.C. § 553(b)(3)(A) (1994). |
| 50. | UAW v. OSHA, 938 F.2d 1310, 1325 (D.C. Cir. 1991). |
| 51. | The issue of making a policy choice to avoid litigation is further discussed infra at notes 83-85 and accompanying text. |
| 52. | The confidential nature of settlement also makes it a particularly difficult phenomenon to study: settlements associated with judicial approval of a consent decree are often published, but many settlements are made available only to those parties present in an appeal. Thus, because available data is not necessarily reflective of the practice, a study of settlements would not easily lend itself to generalizations about the practice. |
| 53. | Coglianese, supra note 27, at 757. |
| 54. | Id. |
| 55. | Interpretive rules and policy statements are not required to undergo the notice and comment process, but may be made effective upon publication. 5 U.S.C. § 553(b)(3)(A) (1994) (noting an exception to notice and comment for "interpretative rules, general statements of policy, or rules of agency organization, procedure, or practice"). An agency is required, of course, to publish such guidelines in the Federal Register to the extent that the guidelines adversely affect individuals who may later decide to challenge the guidelines' application to them. Id. § 552(a)(1)(D). Although not all interpretive rules or policy statements are published, all must be made available for public inspection and copying. Id. § 552(a)(2)(B). If an agency directs its interpretive rules or policy statements to only particular parties or circumstances in the course of a settlement, it may be able to take advantage of the exception to Federal Register publication. See Nguyen v. United States, 824 F.2d 697, 700 n.5 (9th Cir. 1987) (observing that § 552(a)(2) "applies to those interpretations and policy statements directed to particular parties or circumstances"). |
| 56. | See, e.g., Fiss, supra note 10, at 1075 (arguing that settlements are often unjust and that settlement should be "neither encouraged nor praised"). |
| 57. | See, e.g., Frank B. Cross, In Praise of Irrational Plaintiffs, 86 CORNELL L. REV. 1, 24-25 (2000) (arguing that settlement is often inefficient, especially in the context of repeat player litigants, and that policies should promote the corrective capacity of the irrational litigant). |
| 58. | The APA allows a judge to stay agency action pending conclusion of an appellate proceeding. 5 U.S.C. § 705 (1994). A pending appellate proceeding includes implementation of a settlement agreement, so long as a court retains jurisdiction over the case pending implementation. |
| 59. | See, e.g., USA Group Loan Servs., Inc. v. Riley, 82 F.3d 708, 714-15 (7th Cir. 1996) (allowing an agency to renege on a negotiated regulation). |
| 60. | See, e.g., Funk, Subversion, supra note 22, at 1374-76 (observing that negotiated regulation is not necessarily consistent with the administrative process). |
| 61. | See supra notes 29-36 and accompanying text. |
| 62. | Caldart & Ashford, supra note 31, at 171-73. |
| 63. | See infra note 106 and accompanying text. |
| 64. | See supra notes 44-51 and accompanying text. |
| 65. | Natural Res. Def. Council, Inc. v. Costle, 12 Env't Rep. Cas. (BNA) 1833, 1834 (D.D.C. 1979) [hereinafter Costle I]. |
| 66. | Id. at 1834. |
| 67. | As the District of Columbia Circuit described the consent decree, "[i]t did not specify the substantive result of any regulations EPA was to propose and only required EPA to initiate 'regulatory action' for other pollutants identified through the research program." Citizens for a Better Env't v. Gorsuch, 718 F.2d 1117, 1121 (D.C. Cir. 1983) [hereinafter Gorsuch II]. Importantly, "the Companies would be free to influence the content of the proposed regulations by participating in the rulemaking proceedings and then could attack the legality of any final regulations in court." Id. |
| 68. | Envtl. Def. Fund, Inc. v. Costle, 636 F.2d 1229, 1235 (D.C. Cir. 1980) [hereinafter Costle II]. |
| 69. | Id. at 1259. |
| 70. | Natural Res. Def. Council, Inc. v. Gorsuch, 16 Env't Rep. Cas. (BNA) 2084, 2087 (D.D.C. 1982) [hereinafter Gorsuch I]. |
| 71. | Gorusch II, 718 F.2d at 1128-30. |
| 72. | Id. at 1130. |
| 73. | Id. at 1136 (Wilkey, J., dissenting). Judge Wilkey observed:
[A] decree of this type binds not only those present Administrators who may welcome it, but also their successors who may vehemently oppose it. For reasons that ultimately have to do with preserving the democratic nature of our Republic, American courts have never allowed an agency chief to bind his successor in the exercise of his discretion.Id. at 1137. Judge Wilkey did not, however, address how regulations are binding on subsequent administrations. |
| 74. | Id. at 1131 ("In short, the court can only enter as a consent decree such relief as would have been within its jurisdictional power had the case gone to trial."). |
| 75. | See, e.g., Michael W. McConnell, Why Hold Elections? Using Consent Decrees to Insulate Policies from Political Change, 1987 U. CHI. LEGAL F. 295, 315-17 (criticizing such consent decrees for sidestepping the democratic process); Jeremy A. Rabkin & Neal E. Devins, Averting Government by Consent Decree: Constitutional Limits on the Enforcement of Settlements with the Federal Government, 40 STAN. L. REV. 203, 204 (1987) (observing the "obvious danger" that consent decrees allow "one administration to commit its successors to policies they might not otherwise have chosen"). |
| 76. | Memorandum from Edwin Meese III, Attorney General, to All Assistant Attorneys General and All United States Attorneys 3, 4 (Mar. 13, 1986), reprinted in U.S. DEP'T OF JUSTICE, OFFICE OF LEGAL POL'Y, GUIDELINES ON CONSTITUTIONAL LITIGATION 150, 152-53 (Feb. 19, 1988). For an overview of this memorandum, see generally Timothy Stoltzfus Jost, The Attorney General's Policy on Consent Decrees and Settlement Agreements, 39 ADMIN. L. REV. 101 (1987) (suggesting that the new policy may increase litigation among agencies because of the directive not to enter into constraining consent decrees or settlement agreements). |
| 77. | For criticisms of the Meese memorandum, see Robert V. Percival, The Bounds of Consent: Consent Decrees, Settlements and Federal Environmental Policy Making, 1987 U. CHI. LEGAL F. 327, 337-41 (arguing that guidelines were based on "an unreasonably restrictive view of the permissible scope of settlement commitments"); Shane, supra note 4, at 266 (demonstrating that legal backing for the attorney general's argument was weak because of the distinction between enforcing consent decrees and settlement agreements). In the last term of President William Jefferson Clinton's administration, an acting assistant attorney general in the Office of Legal Counsel concluded that Attorney General Meese's concerns were based in public policy, not in legal limitations on the scope of the executive branch's powers to settle litigation in manners that limit the future exercise of executive branch discretion. Memorandum from Randolph D. Moss, Acting Assistant Attorney General, Office of Legal Counsel, to Raymond C. Fisher, Associate Attorney General, Authority of the United States to Enter Settlements Limiting the Future Exercise of Executive Branch Discretion (June 15, 1999), available at http://www.usdoj.gov/olc/consent_decrees2.htm (on file with the Duke Law Journal). Nevertheless, remnants of the Meese memorandum survive in Subpart Y of Chapter 28 of the Code of Federal Regulations, which addresses the authority of attorneys general to compromise and close civil claims. See 28 C.F.R. § 0.160 (2001) (requiring assistant attorneys general to refer to the deputy attorney general or the associate attorney general a "proposed settlement [that] converts into a mandatory duty the otherwise discretionary authority of a department or agency to promulgate, revise, or rescind regulations"). |
| 78. | Cindy Skrzycki, Pesticide Firms Seek EPA Intervention, WASH. POST, Feb. 27, 2001, at E1 (documenting the response of the pesticide industry to an EPA settlement reached at the end of the Clinton administration). |
| 79. | These chemicals include organophosphates, commonly used insecticides. See id. |
| 80. | 402 U.S. 673 (1971). |
| 81. | Id. at 681. |
| 82. | Under Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984), courts will defer to an agency's interpretation of a statute to the extent that the statute does not speak precisely to the question at issue. Id. at 843. In other words, to the extent that Congress delegated interpretive discretion to the agency, courts will not second-guess the agency's interpretive choice. Where statutes are ambiguous, Chevron deference depends on an assessment of whether an agency has been delegated authority to carry lawmaking force in its interpretations. See infra notes 86, 108 and accompanying text. The Chevron doctrine promotes both agency expertise and political accountability. Chevron, 467 U.S. at 865. |
| 83. | See supra notes 29-36 and accompanying text (discussing OSHA's formaldehyde settlement). |
| 84. | 402 U.S. at 681. |
| 85. | See Phillip G. Oldham, Regulatory Consent Decrees: An Argument for Deference to Agency Interpretations, 62 U. CHI. L. REV. 393, 394 (1995) (arguing that the "fundamental shift in the relationship between courts and agencies embodied in Chevron should encompass consent decree interpretation"). |
| 86. | Wash. Urban League v. Fed. Energy Regulatory Comm'n, 886 F.2d 1381, 1386 (3d Cir. 1989). The Eighth Circuit requires substantial evidence to support an agency's interpretation of a settlement agreement. Miernicki v. R.R. Ret. Bd., 20 F.3d 354, 356 (8th Cir. 1994). Also, courts appear to defer to agency interpretations of settlement agreements between private stakeholders that the agency had the power to approve. See Transwestern Pipeline Co. v. Fed. Energy Regulatory Comm'n, 988 F.2d 169, 172-73 (D.C. Cir. 1993) (supporting judicial deference to the Federal Energy Regulatory Commission where its reading of a contractual term serves a limited regulatory purpose and does not "give rise to the specter of res judicata"); Nat'l Fuel Gas Supply Corp. v. Fed. Energy Regulatory Comm'n, 811 F.2d 1563, 1569 (D.C. Cir. 1987) (concluding that a court must give deference to an agency's resolution of a settlement agreement where the issue involves merely the construction of language). |
| 87. | Consider, for example, Southern Union Gas Co. v. Federal Energy Regulatory Commission, 840 F.2d 964, 971 (D.C. Cir. 1988):
The Commission's interest in ordering its docket is especially strong here, where petitioners are seeking to reopen a settlement agreement to which they were themselves signatories. . . . We have had occasion more than once to observe the breadth of the Commission's discretion with respect to settlements, as well as the broad public interest favoring the settlement of complex rate matters. |
| 88. | See, e.g., United States v. Mead Corp., 533 U.S. 218, 121 S. Ct. 2164, 2167 (2001) (observing that "the overwhelming number of our cases applying Chevron deference have reviewed the fruits of notice-and-comment rulemaking or formal adjudication," not interpretive rules and policy statements); Christensen v. Harris County, 529 U.S. 576, 587 (2000) (finding that interpretations contained in an opinion letter do not warrant Chevron-type deference). |
| 89. | For this reason, even strong advocates of Chevron deference couple it with rigorous arbitrary and capricious review of the agency's interpretation of ambiguous terms. See Oldham, supra note 85, at 414 (noting that the review for arbitrariness "should closely mirror arbitrary-and-capricious review under Chevron, largely based on the standard articulated in the Administrative Procedure Act"). |
| 90. | See John B. Judis, Round Midnight, AM. PROSPECT, Feb. 12, 2001, at 11 (noting the last-minute regulations and executive orders in such fields as the environment, workers' health and safety laws, and trade); Cindy Skrzyzcki, "Midnight Regulations" Swell Register, WASH. POST, Jan. 23, 2001, at E1 (outlining the response of the Bush administration to late regulations, including the instruction to agencies to write to the Federal Register to terminate publication). |
| 91. | According to one account:
[V]irtually all the regulations finished by federal agencies shortly before Clinton left office had been developed over years, according to government documents, outside policy analysts, and officials of the Bush and Clinton administrations. Some had been delayed by lawsuits or because Republican-led Congresses of the mid- to late-1990s had explicitly forbidden federal agencies to work on them. Moreover, the regulations completed during Clinton's final weeks in office were in step with a brisk pace of regulatory work throughout his two terms-and with a longstanding practice in which presidents of both political parties have issued many regulations just before they departed.Amy Goldstein, "Last-Minute" Spin on Regulatory Rite, Bush Review of Clinton Initiatives Is Bid to Reshape Rules, WASH. POST, June 9, 2001, at A1. |
| 92. | This vote was the first time that Congress used its fast-track, joint approval process, established in 1996, to suspend a rule. Steven Greenhouse, House Joins Senate in Repealing Rules on Workplace Injuries, N.Y. TIMES, Mar. 7, 2001, at A19; Steven Greenhouse, Senate Votes to Repeal Rules Clinton Set on Work Injuries, N.Y. TIMES, Mar. 6, 2001, at A1. |
| 93. | See Alice Lipowicz, Injury Rule Puts Strain on Businesses; New Standards Too Costly, Employers Say; Delay is Sought, CRAIN'S N.Y. BUS., Feb. 12, 2001, at 22 (discussing the efforts of the National Coalition for Ergonomics, a coalition of business groups, to overturn the rule). |
| 94. | Natural Res. Def. Council, Inc. v. EPA, 683 F.2d 752, 766 (3d Cir. 1982) (holding that indefinite suspension of rules issued under the Clean Water Act constitutes repeal of the rule). |
| 95. | Enforceability of such a settlement term probably raises some separation of powers concerns. See, e.g., McConnell, supra note 75, at 297-99 (discussing the constitutional limits on consent decrees); Rabkin & Devins, supra note 75, at 226-28 (outlining a general theory of constitutional limitations). |
| 96. | Katharine Q. Seeyle, U.S. to Reassess Snowmobile Ban in a Park, N.Y. TIMES, June 30, 2001, at A10. |
| 97. | The Fund for Animals Condemns Snowmobile Litigation Settlement Agreement, PR NEWSWIRE, July 5, 2001; Anuj Gupta, Yellowstone Snowmobile Ban May Lift, L.A. TIMES, June 30, 2001, at A11; Mr. Bush's Miscalculation, N.Y. TIMES, July 1, 2001, at A12. |
| 98. | The significance of the distinction between adoption and publication was that the rule had been formally endorsed by the agency but was not yet "final" for purposes of APA review. |
| 99. | Publication of the rule had been held up in part due to a temporary restraining order and in part due to Andrew Card's memorandum. See Memorandum from Andrew H. Card, Jr., to the Heads and Acting Heads of Executive Departments and Agencies, Regulatory Review Plan (Jan. 20, 2001) (on file with the Duke Law Journal) (directing agencies to withhold sending regulations to the Office of the Federal Register pending review by the Office of President Bush, postponing for sixty days the effective date of regulations published in the Federal Register but not yet effective, and withdrawing all regulations sent to the Office of the Federal Register but not yet published). |
| 100. | William Claiborne, Hog Farmers Vow to Sue USDA; Agency's Reversal on Vote Against Promotion Fee Stirs Anger, WASH. POST, Mar. 2, 2001, at A7. |
| 101. | Id. Since the rule had not been published, it was not final and this lawsuit was not an appeal of the agency's final action under the APA. |
| 102. | These arguments, in general terms are: (1) the Secretary of Agriculture did not have the legal authority to call for a referendum on the pork promotion program given that fifteen percent of the eligible pork producers and importers did not sign a petition to request a referendum; (2) the Secretary of Agriculture did not have the legal authority to terminate the pork promotion program at all, and; (3) given the numerous irregularities identified in the implementation and voting on the referendum, there was no clear determination on the true outcome of the vote. Compl. ¶¶ 2, 4, Michigan Pork Producers Ass'n, Inc. v. Glickman, No. 01CV-34, U.S. Dist. LEXIS 1359 (W.D. Mich. Jan. 19, 2001). |
| 103. | See U.S. Reverses Plan to End Pork Promotion, N.Y. TIMES, Feb. 28, 2001, at A19 (discussing the settlement by the Department of Agriculture to continue promoting pork). |
| 104. | For more on the case, see Agric. Mktg. Serv., Pork Checkoff Program, USDA, at http://www.ams.usda.gov/lsg/mpb/pork.htm (last visited Nov. 10, 2001) (on file with the Duke Law Journal). According to a New York Times account, this settlement has transformed small hog farmers into "angry activists." Elizabeth Becker, Unpopular Fee Makes Activists of Hog Farmers, N.Y. TIMES, June 8, 2001, at A1. A small group of hog farmers from four Midwestern states, Campaign for Family Farms, has requested a federal court to uphold the initial vote, because it had been certified by the General Accounting Office and the inspector general of the Agriculture Department, as well as the previous agency head. Id. |
| 105. | The settlement is available online at Agric. Mktg. Serv., Settlement Agreement, USDA, at http://www.ams.usda.gov/hotissues/fnlsettl.htm (last visited Nov. 10, 2001) (on file with the Duke Law Journal). |
| 106. | 486 F.2d 375 (D.C. Cir. 1973). |
| 107. | Id. at 392-94 ("[I]nformation should generally be disclosed as to the basis of a proposed rule at the time of issuance."). |
| 108. | In Christensen v. Harris County, 592 U.S. 576 (2000), Justice Scalia's concurrence recognized that a legal interpretation adopted in a solicitor general's brief, co-signed by an agency head, is entitled to Chevron deference. Id. at 591 (Scalia, J., concurring). After United States v. Mead Corp., 533 U.S. 218, 121 S. Ct. 2164 (2001), however, a majority of the Court's members recognize an agency's interpretation is entitled to Chevron deference only where an agency has been delegated authority to speak with the force of law and where the agency makes its interpretation with a lawmaking pretense in mind. Id. at 2172-74. Agency legal interpretations adopted in the course of notice and comment rulemaking or formal adjudication are subject to the safe harbor of Chevron, but interpretations adopted in other contexts will be evaluated on a case by case basis. Id. Significantly, Mead suggests that a factor in assessing whether the agency had a lawmaking purpose is the agency's intent to bind third parties to the interpretation. Id. at 2174. A litigation position in a brief presumably binds no one. |
| 109. | This paradigm for evaluating the various ways of influencing institutional behavior in the context of civil settlements is presented in Coffee, supra note 40, at 376, which draws on the classic work by Albert Hirschman, supra note 40. In other litigation contexts not involving the creation of law and policy by administrative agencies, the role of exit may be more significant. See, e.g., Coffee, supra note 40, at 376 (arguing for attention to "exit" and "voice" over "loyalty" in civil class action litigation). |
| 110. | See, e.g., Hansberry v. Lee, 311 U.S. 32, 40 (1940) (holding that those privy to the settlement agreement but not made parties to the litigation could not be bound by a decree on a theory that the suit was a class suit in which they were adequately represented); see also Patrick Woolley, The Availability of Collateral Attack for Inadequate Representation in Class Suits, 79 TEX. L. REV. 383, 432-46 (2000) (discussing the continued applicability of Hansberry notwithstanding recent appellate cases rejecting the rule). |
| 111. | 490 U.S. 755 (1989). |
| 112. | Id. at 763-65. |
| 113. | Id. at 759-60. |
| 114. | Id. at 760. |
| 115. | Id. at 761. |
| 116. | Id. at 765. |
| 117. | Although collateral attack may be permissible in federal courts, state courts may not allow it to the same degree. In state courts, the Full Faith and Credit Clause may preclude later collateral attack to a settlement or consent decree. See Kremer v. Chem. Constr. Corp., 456 U.S. 461, 481-82 (1982) (stating that "state proceedings need do no more than satisfy the minimum procedural requirements of the Fourteenth Amendment's Due Process Clause in order to qualify for the full faith and credit guaranteed by federal law"); State v. Canney, 562 A.2d 1315, 1316 (N.H. 1989) (noting that a collateral attack on administrative determinations is precluded where the allegedly aggrieved party failed to exercise his statutory right to appeal); United Am. Ins. Co. v. Whaland, 337 A.2d 358, 360 (N.H. 1975) (rejecting an insurer's belated challenge to an administrative order because its finality did not offend due process). |
| 118. | Similarly, in granting intervention to challenge settlements, courts do not necessarily need to decide whether a party has standing to appeal ex post a regulation implementing a settlement. Intervention can be limited for purposes of participating in a settlement or challenging the settlement agreement. See, e.g., Carlough v. Amchem Prods., Inc., 5 F.3d 707, 710 (3d Cir. 1993) (upholding a lower court order allowing intervenors "active participation in the status of objectors to the settlement agreement"); Howard v. McLucas, 782 F.2d 956, 960-61 (11th Cir. 1986) (holding that white federal employees may participate in developing a remedy to discrimination against black employees); Kirkland v. N.Y. Dep't of Corr. Servs., 711 F.2d 1117, 1128 (2d Cir. 1983) (allowing white corrections officers to intervene in a suit between black officers and the corrections department for the limited purposes of participating in hearings on proposed settlement); Bradley v. Milliken, 620 F.2d 1141, 1142 (6th Cir. 1980) (permitting intervention for the limited purpose of providing evidence about past discrimination against Hispanics); Shore v. Parklane Hosiery Co., 606 F.2d 354, 358 (2d Cir. 1979) (allowing a group of investors to intervene by participating in hearings between party-investors and a corporation over settlement terms). |
| 119. | As John Coffee has described class action settlements, "At its worse, this process can develop into a reverse auction, with the low bidder among the plaintiffs' attorneys winning the right to settle with the defendant." John C. Coffee, Jr., Class Wars: The Dilemma of the Mass Tort Class Action, 95 COLUM. L. REV. 1343, 1354 (1995) (arguing that class action settlements present an opportunity for defense lawyers to collude with weak plaintiff's lawyers, who may not adequately represent the interests of their clients in settlement). |
| 120. | For a similar critique of negotiated regulation, see Funk, Subversion, supra note 22, at 1387. |
| 121. | 5 U.S.C. § 706 (1994). |
| 122. | 463 U.S. 29 (1983). |
| 123. | Id. at 38-39. |
| 124. | Id. at 34-39. |
| 125. | Id. at 43. |
| 126. | Id. at 56-57. |
| 127. | Id. at 54. |
| 128. | Id. at 55 (citing legislative history). |
| 129. | Mathew D. McCubbins et al., Administrative Procedures as Instruments of Political Control, 3 J.L. ECON. & ORG. 243, 255 (1987). |
| 130. | See, e.g., Merrick B. Garland, Deregulation and Judicial Review, 98 HARV. L. REV. 505, 586-91 (1985) (characterizing hard look review as a reinforcement of agency fidelity to Congress). |
| 131. | Some scholars argue that courts should view judicial review of measures enacted by direct democratic means, such as voter initiative, more skeptically than those put into place via ordinary legislation. See, e.g., Julian N. Eule, Judicial Review of Direct Democracy, 99 YALE L.J. 1503, 1558-73 (1990) (arguing for a "hard judicial look" when challenges are raised "against laws enacted by substitutive plebiscite"); Hans A. Linde, When Initiative Lawmaking Is Not "Republican Government": The Campaign Against Homosexuality, 72 OR. L. REV. 19, 41-45 (1993) (describing test for defining manageable "standards for initiative lawmaking within a republican form of state government"); Jane S. Schacter, The Pursuit of "Popular Intent": Interpretive Dilemmas in Direct Democracy, 105 YALE L.J. 107, 123-67 (1995) (exploring the interpretation process used by courts in applying statutory law enacted through the initiative process and describing the problems of intentionalist methodology in the search for "popular intent"). Although this argument bears some analogy to my call for rigorous review of settlement-in the sense that I am skeptical about the accountability of private ordering-I am not proposing that settlement be reviewed more rigorously than other administrative actions that implement law or policy, such as the adoption of rules, policy statements, or interpretations. I am only proposing that arbitrary and capricious review extend to this sort of private ordering as well as to final agency action otherwise subject to review. |
| 132. | Richard A. Nagareda, Turning from Tort to Administration, 94 MICH. L. REV. 899, 904 (1996). |
| 133. | Mars Steel Corp. v. Cont'l Ill. Nat'l Bank & Trust, 834 F.2d 677, 681 (7th Cir. 1987); see also Nagareda, supra note 132, at 908-14 (discussing the consequences of contingent fee arrangements and franchise agreements within the context of the mass torts bar). |
| 134. | Nagareda, supra note 132, at 946-52. Under Rule 23, trial courts require class action settlements to be fair, adequate, and reasonable prior to their approval. See FED. R. CIV. P. 23(e) ("A class action shall not be dismissed or compromised without the approval of the court . . . ."); see also Williams v. City of New Orleans, 729 F.2d 1554, 1559-60 (5th Cir. 1984) (noting the active role of a district court in the settlement of a Title VII consent decree case and stating that the court will approve settlement if it is "fair, adequate and reasonable"); Weinberger v. Kendrick, 698 F.2d 61, 73 (2d Cir. 1982) (Friendly, J.) (recognizing that a "district court's disposition of a proposed class action settlement should be accorded considerable deference"). For further discussion of how courts have struggled in implementing this standard, see Roger C. Cramton, Individualized Justice, Mass Torts, and "Settlement Class Actions": An Introduction, 80 CORNELL L. REV. 811, 819-23 (1995). |
| 135. | Nagareda, supra note 132, at 946. |
| 136. | State Farm requires an agency to explain the reasons for the change in its position, pursuant to the arbitrary and capricious test. Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 41 (1983). Of course, State Farm should not be viewed as a case that automatically favors the initially adopted rules. State Farm merely requires an agency to explain the basis for its regulations. Hastily enacted regulations will be more likely to face challenge under the State Farm test. Hastily enacted rules, even under State Farm, may certainly be hastily repealed. |
| 137. | Id. at 59 (Rehnquist, J., dissenting in part and concurring in part). |
| 138. | Id. at 52. |
| 139. | See supra note 134 and accompanying text. |
| 140. | See, e.g., Thomas M. Mengler, Consent Decree Paradigms: Models Without Meaning, 29 B.C. L. REV. 291, 291 n.1 (1988) (defining the role of courts in administering consent judgments). |
| 141. | In re Smith, 926 F.2d 1027, 1029 (11th Cir. 1991) (quoting JAMES WM. MOORE ET AL., MOORE'S FEDERAL PRACTICE, MANUAL FOR COMPLEX LITIGATION 2d, § 23.14 (1986)). |
| 142. | United States v. City of Miami, 614 F.2d 1322, 1331 (5th Cir. 1980). |
| 143. | Smith, 926 F.2d at 1029; see also Atl. Co. v. Broughton, 146 F.2d 480, 482 (5th Cir. 1944) ("Though settlements in accord and satisfaction are favored in law, they may not be sanctioned and enforced when they contravene and tend to nullify the letter and spirit of an Act of Congress."). |
| 144. | Under the APA, judges reviewing agency action can issue "all necessary and appropriate process to postpone the effective date of an agency action or to preserve status or rights pending conclusion of the review proceedings." 5 U.S.C. § 705 (1994). In assessing the appropriateness of § 705 stays, some courts have evaluated whether issuance of the stay is consistent with the public interest. See Hamlin Testing Labs., Inc. v. U.S. Atomic Energy Comm'n, 337 F.2d 221, 222-23 (6th Cir. 1964) (weighing the public interest heavily in denying a stay); Associated Sec. Corp. v. SEC, 283 F.2d 773, 775 (10th Cir. 1960) (denying a stay because it was not established that granting the stay will not be contrary to the public interest). |
| 145. | See supra note 144. |
| 146. | See Mark Seidenfeld, An Apology for Administrative Law in The Contracting State, 28 FLA. ST. U. L. REV. 215, 236-37 (2001) (suggesting that courts review the enforcement settlement process as part of the ongoing process of regulatory implementation, subjecting it to APA standards of review). A consent decree must "further the objectives of the law upon which the complaint was based." Local No. 93, Int'l Ass'n of Firefighters v. City of Cleveland, 478 U.S. 501, 525 (1986). In the context of judicial review of regulations or enforcement proceedings under the APA, that law includes the APA and its standards of review. |
| 147. | Of course, a stakeholder will face other process-oriented incentives for settling, including reputation benefits that might benefit the stakeholder in other proceedings with the agency. |
| 148. | See, e.g., Rossi, supra note 15, at 211-41 (addressing the tension between voice interests, which are associated with mass participation, and deliberation, a type of loyalty concern). |