138 29 718KB
English Pages 320 Year 2010
The Litigation State
princeton studies in american politics: historical, international, and comparative perspectives Ira Katznelson, Martin Shefter, and Theda Skocpol, series editors
A list of titles in this series appears at the back of the book
The Litigation State public regulation and private lawsuits in the u.s. Sean Farhang
princeton university press princeton and oxford
Copyright © 2010 by Princeton University Press Published by Princeton University Press, 41 William Street, Princeton, New Jersey 08540 In the United Kingdom: Princeton University Press, 6 Oxford Street, Woodstock, Oxfordshire OX20 1TW All Rights Reserved ISBN: 978-0-691-14381-1 ISBN (pbk.): 978-0-691-14382-8 Library of Congress Control Number: 2010927494 British Library Cataloging-in-Publication Data is available This book has been composed in Sabon Printed on acid-free paper ∞ press.princeton.edu Printed in the United States of America 1 3 5 7 9 10 8 6 4 2
To my father, Mansour Farhang who got me going down this road
This page intentionally left blank
CONTENTS
Illustrations and Tables
ix
Acknowledgments
xi
Part I: Private Enforcement Regimes in General Chapter 1 An Introduction to Private Enforcement Regimes
3
Chapter 2 Institutional Foundations of Private Enforcement Regimes
19
Chapter 3 An Empirical Model of Enactment of Private Enforcement Regimes
60
Part II: Private Enforcement Regimes and Civil Rights Introduction to Part II
85
Chapter 4 Foundations: The Civil Rights Act of 1964
94
Chapter 5 Reverberations: 1965–1976
129
Chapter 6 Escalation: The Civil Rights Act of 1991
172
Chapter 7 Conclusions and Implications
214
Notes
235
Index
293
This page intentionally left blank
ILLUSTRATIONS AND TABLES
Figure 1.1. Private and U.S. Plaintiff Statutory Litigation Rates in Federal Court, 1942–2005 Figure 1.2. Private Statutory and Tort Litigation Rates in Federal Court, 1942–2005 Figure 3.1. Private Enforcement Regimes, 1887–2005, and Private Statutory Litigation Rates, 1942–2005 Figure 6.1. Employment Discrimination Litigation in Federal Court, 1977–2005 Table 2.1. Table 2.2. Table 2.3. Table 2.4. Table 3.1. Table 3.2.
Calculating Expected Value of Suit Elements of Private Enforcement Regimes Legislative Manipulation of Expected Value Expected Value in a Separation of Powers Context Private Enforcement Regimes by Policy Area Autoregressive Poisson Models of Enactments
12 15 66 201 25 28 29 53 67 77
This page intentionally left blank
ACKNOWLEDGMENTS
I have incurred many debts along the way. Foremost is the one owed to my dissertation committee, comprised of Ira Katznelson (chair), Chuck Cameron, Robert Lieberman, and Greg Wawro. Beyond offering outstanding advice and generous support on the dissertation, they made for an excellent scholarly community. The plurality of their intellectual and methodological orientations, joined with their mutual respect for one another’s points of view, inspired the plurality of approaches brought to bear in this book. Many other people were generous with their time and gave me valuable feedback on work in progress. For all its shortcomings, the book is vastly improved because of their help. For laboring through the entire manuscript and offering important suggestions on how to make it better, I thank Tom Burke, Frank Cross, Charles Epp, Robert Kagan, Bert Kritzer, Susan Lawrence, Shep Melnick, Robert Mickey, and John Skrentny. I am grateful to the many research assistants who contributed along the way, and I wish to particularly acknowledge Avi Springer, who provided unflagging and exemplary research assistance on numerous facets of this book during his two years at UC Berkeley’s Goldman School of Public Policy. I am also grateful to the University of California Hellman Faculty Fund for providing financial support for the project. The faculty and students at UC Berkeley’s Goldman School of Public Policy provided a wonderful community that helped this project grow to fruition. Portions of chapters 2 and 3 are a revised version of “Public Regulation and Private Lawsuits in The American Separation of Powers System,” American Journal of Political Science 52 (2008): 821–39. Portions of chapters 4 and 5 are a revised version of “The Political Development of Job Discrimination Litigation, 1963–1976,” Studies in American Political Development 23 (2009): 23–60. For her toleration, patience, and love during the difficult passages in this book’s journey, I thank my wife Ingrid. For delaying that journey by giving me something more fun and important to do, I thank our son Silas.
This page intentionally left blank
The Litigation State
This page intentionally left blank
PA RT I
Private Enforcement Regimes in General
This page intentionally left blank
Chapter 1 AN INTRODUCTION TO PRIVATE ENFORCEMENT REGIMES
Next to petitions by prisoners to be set free, job discrimination lawsuits are the single largest category of litigation in federal courts. Over the past decade or so, the annual number of such lawsuits averaged about 20,000.1 Two percent of these job discrimination suits were prosecuted by the federal government, while 98 percent were litigated by private parties. The enormous volume of privately prosecuted employment discrimination litigation has earned it a prominent role among poster children for the much-maligned “litigation explosion.”2 As one commentator recently put it, “[F]rom malpractice suits to libel actions, from job discrimination to divorce, litigation has become a way of life in the United States,” making it “the world’s most litigious society.”3 How did job discrimination litigation become a part of the American way of life? A critical part of the answer concerns the way federal job discrimination statutes—the most important of which is the foundational Title VII of the Civil Rights Act of 1964—are written. The existence and extent of private litigation enforcing a statute is to an important degree the product of legislative choice over questions of statutory design.4 One need only consider two of the other largest federal interventions in the employment relationship—one before the CRA of 1964 and one after it— to drive this point home. While creating a wide array of rights for workers, neither the National Labor Relations Act of 1935 nor the Occupational Safety and Health Act of 1970 allowed private enforcement lawsuits for implementation. Instead, in those laws Congress opted for bureaucracycentered enforcement regimes that empowered administrators to undertake investigations, hold hearings, and issue orders. It is a legislative choice to rely upon private litigation in statutory implementation. And when Congress does choose to rely upon private litigation by including a private right of action in a statute, it faces a series of additional choices of statutory design concerning who has standing to sue, which parties will bear the costs of litigation, what damages will be available to winning plaintiffs, whether a judge or jury will make factual determinations and assess damages, and rules of liability, evidence, and proof that together can have profound consequences for how much or
4
CHAPTER 1
little private enforcement litigation will actually be mobilized. This book refers to this system of rules as a statute’s private enforcement regime. While private plaintiff-driven civil rights litigation is so familiar a part of the American legal landscape that it has an air of inevitability, this approach to implementing job discrimination laws was not foreordained. To the contrary, a resolutely bureaucracy-centered approach to remedying job discrimination, founded upon administrative cease-and-desist authority rather than the private right to sue, actually represented the dominant model in 1964. Of twenty-eight states with fair employment practice laws in 1964, twenty-one used the administrative cease-anddesist model, four used only criminal and no civil sanctions, and three lacked enforcement provisions and were strictly voluntary.5 Only a single United States territory—Puerto Rico—used the enforcement model that Congress would ultimately follow in the job discrimination provisions in the CRA of 1964: statutory provision for private civil actions in court, with economic damages and attorney’s fee awards for winning plaintiffs.6 The dominance of private litigation enforcing federal job discrimination laws that we take for granted today, widely regarded as emblematic of America’s litigious “way of life,” was a remarkably anomalous departure in 1964. Why did it happen? The answer to this particular policy history question, which is the focus of chapters 4 to 6, points to a conceptually broader argument taken up in this book about the large role of private litigation in the implementation of statutory policy in the United States. Legislators and the interest groups that influenced their behavior, with a high degree of self-consciousness, and centrally motivated by policy goals, constructed Title VII’s enforcement provisions with the objective of mobilizing private litigants to execute the enforcement function in court. As Senator James Abourezk (D-SD) would later put it, Title VII’s enforcement provisions were designed to provide for enforcement of the law “by enlisting private citizens as law enforcement officials.”7 The legislators who effectively deputized private litigants and their attorneys to enforce the law manifestly understood themselves to be facing a choice between building an authoritative bureaucratic enforcement apparatus on the one hand, and delegating enforcement to private litigants and courts on the other. The legislative choice of private litigation over administrative power emerged from conflict between ideologically antagonistic interests, channeled through America’s fragmented political institutions, particularly the dynamic of legislative-executive competition for control of the bureaucracy. The structure of American political institutions decisively shaped the outcome in Title VII of the CRA of 1964, and in subsequent important civil rights laws expanding the role of private enforcement. A
INTRODUCTION
5
contention at the heart of this book is that America’s fragmented state structures drives legislative enactment of private enforcement regimes. In elaborating this argument the book draws extensively on political science literature on American courts in the regulatory process (particularly Robert Kagan, Shep Melnick, and Thomas Burke), rational choice institutionalism (particularly Terry Moe, and the collaborative work of McCubbins, Noll, and Weingast), and historical institutionalism (particularly the literature on American political development). The core of the institutional arguments operates along several dimensions that are briefly previewed here, and that are spelled out in detail in chapter 2, where the literatures that the book builds upon are laid out and integrated. The first institutional relationship is by far the most central to the argument of the book, powerful in effect, and pivotal to explaining the steep rise in private enforcement litigation starting in the late 1960s, documented below. It is that conflict between Congress and the president over control of the bureaucracy, a perennial feature of the American state, creates incentives for Congress to bypass the bureaucracy and provide for enforcement via private litigation. This cause of private enforcement regimes has become much more significant to American public policy since the late 1960s, when divided party control of the legislative and executive branches became the norm and relations between Congress and the president became more antagonistic, a condition that was exacerbated by growing ideological polarization between the parties. These conjoint conditions of divided government and party polarization continued through the end of the twentieth century and into the beginning of the twentyfirst. The book will argue that this legislative-executive ideological polarization is an important cause of a coincident explosion of private lawsuits enforcing federal statutes since the Nixon administration, and of the corresponding, and widely remarked, role of private litigants, lawyers, and judges in American policy. Second, the many veto points that characterize America’s fragmented state structures render a very sticky status quo, making future amendments to a law, once passed, hard to accomplish. For reasons to be discussed in the next chapter, this sticky status quo creates an incentive for legislators and their interest group constituents to rely upon private enforcement regimes, which provide a form of auto-pilot enforcement, via market incentives, that will be difficult for future legislative majorities, or errant bureaucrats pursuing their own goals, to subvert. Third, the same veto-point-ridden institutional environment often necessitates compromising with many gatekeepers, which frequently entails scaling back ambitious policy proposals, and this institutional environment favors privatizing the enforcement function as opposed to administrative state-
6
CHAPTER 1
building. Before these institutional arguments are set forth in chapter 2, it is necessary first to establish that private enforcement regimes, as a policy instrument, are a critical component of American regulatory state capacity.
Private Statutory Litigation and Regulatory State Capacity Legislators’ construction of statutory private enforcement regimes has deep and underexplored implications for American regulatory state capacity. “State capacity” refers simply to the state’s capacity to effectively implement its policy choices.8 “Regulation,” as used here, to borrow a definition from Christopher Foreman, refers to “any governmental effort to control behavior by other entities, including small business firms, subordinate levels of government, or individuals.”9 Regulatory state capacity thus refers to the state’s capacity to successfully implement its efforts to control the behavior of other entities. Owing largely to the concepts and categories used to measure state capacity, scholars in political science and comparative political sociology have failed to adequately grasp private enforcement litigation as a form of state intervention. “State-centered” scholars have tended to operate within a narrow and executive-centered conception of what the state is. “State centered regimes,” explains James Q. Wilson, “are executivecentered regimes, and executive-centered regimes are dominated by their bureaucracies ... [and] mak[e] the administrative apparatus the center of official action.”10 Drawing implicitly upon the Weberian ideal of the modern state, this conceptual framework for studying state capacity has tended to privilege (1) the number of bureaucratic personnel, (2) the degree of their organizational centralization, and (3) the state’s capacity to extract resources, as the axial measures of state capacity.11 These dimensions of bureaucracy, according to Theda Skocpol, are the “universal sinews of state power.”12 Against this Weberian template, the scholarly literature on the American state has found it to be sorely wanting—a “weak state” as compared to the Weberian model of a “strong state,” most closely instantiated in France or Prussia. In assessing this scholarly literature, Ira Katznelson observes that the American state has been “widely portrayed as weak, amateur, decentralized, negligible.”13 William Novak, likewise, has observed more broadly that American social scientists and historians have represented the American state as “something not quite fully formed, . . . something less, something laggard, something underdeveloped when compared to the mature governmental regimes that dominate modern European history.”14
INTRODUCTION
7
This tendency is characteristic of the literature on American political development, the subfield within political science that has paid the most attention over the past quarter century to American state capacity. Indeed, where courts have entered this literature, it has often been as an explanation for state incapacity: courts contribute to the weakness of the American state by obstructing, constraining, and subverting the good works of the elected branches; they are a veto player against democracy; they delimit state capacity.15 If the role of courts in the literature on American political development has been limited, the role of ordinary litigation, proceeding at the bottom of the judicial hierarchy, orchestrated by private lawyers and litigants, overwhelmingly occurring outside of the courtroom and beyond the view of any official state actor, has been almost entirely ignored.16 An excessively executive-centered perspective on state capacity has caused observers to miss the significance of private enforcement litigation as a form of deliberate and effective state intervention. Private enforcement litigation falls outside the scope of vision characteristic of the dominant executive-centered approach to state capacity, and beyond conventional definitional boundaries of the state more broadly. Consequently, its crucial role in the functioning of the American regulatory state—indeed its very “stateness”—has been largely overlooked by students of American state capacity.17 Private Litigation, Policy Instruments, and Infrastructural Power One might readily acknowledge the importance of private statutory enforcement litigation in American policy implementation while resisting the characterization of it as a form of state action. Why should private enforcement litigation be regarded as a component of state capacity? After all, the decision to pursue enforcement litigation will lie with private actors pursuing their own interests, not with state officials, and the costs of enforcement will be borne by private citizens who bring suits and the defendants they sue, not by government agencies. If the state is neither undertaking enforcement activities with its own personnel nor funding them with its revenue, why talk of state capacity? The answer is that state capacity is not exhausted by the actions of state personnel or the expenditure of state resources. If the object of interest is the state’s capacity to implement its policy choices by controlling the behavior of other entities, then one must attend not only to the direct actions of state officers, but also to more indirect pathways of regulatory control. The concept of “policy instruments”—the repertoire of means available to policymakers to achieve their objectives—is useful for illuminating this point. Careful attention to the inventory of feasible policy
8
CHAPTER 1
instruments available to state actors is crucial for understanding state capacity because, as Peter Katzenstein has observed, “[t]he instruments which policy makers command largely determine whether stated objectives can be achieved in the process of policy implementation.”18 When, for example, the Department of Labor undertakes enforcement action under the Fair Labor Standards Act, it constitutes the archetypal exercise of state capacity. The bureaucratic arm of the executive engages in potentially invasive and coercive investigative activity to ascertain violations, and carries out the prosecutorial function, which can result in judicial commands against violators, backed by federal police powers. But what of private litigation to enforce federal statutes? The American civil discovery process effectively confers upon private litigants and their lawyers the same investigatory powers as federal agencies to compel sworn testimony and to disgorge documents; they can obtain the same court orders commanding a violator to cease its unlawful conduct and pay for its violations; and the court orders are backed by the same federal police powers. It is these litigant powers that Senator Abourezk was referring to when he characterized Title VII’s enforcement provisions as designed to provide for enforcement “by enlisting private citizens as law enforcement officials.” The remarkable level of authority and autonomy conferred upon private litigants and lawyers in American civil litigation is distinctive, contrasting sharply with European practices, where judges exercise far more power in managing the civil legal process.19 This facet of the American civil law system leads Robert Kagan to label it “lawyer-dominated,” as contrasted with a European “judge-dominated” approach.20 The standing statutorily conferred on private actors to enforce public regulatory laws in court effectively licenses them to wield the coercive instruments of state power. Marc Galanter and David Luban have referred to this phenomenon as the delegation of “enforcement endowments.”21 The details of the private enforcement regime built on top of the initial decision to license private enforcement determines how likely private litigants and their attorneys will be to actually use the license. Thus, rules comprising the private enforcement regime surrounding substantive statutory rights and prohibitions are, straightforwardly, policy instruments. There is, moreover, considerable evidence that private lawsuits can be an effective tool in shaping the behavior of the regulated population. Studies have found, ranging across such policy domains as job discrimination, sexual harassment, labor, playground safety, antitrust, and police brutality, that implementation of regulatory commands through private lawsuits can effectively encourage and induce compliance behavior by the regulated population, whether they be private entities or governmental subunits.22 The findings have established both “specific deter-
INTRODUCTION
9
rence” and “general deterrence” effects, where specific deterrence refers to the effects of enforcement against a particular violator on that violator’s future conduct, while general deterrence refers to effects of visible enforcement efforts in the legal environment on other would-be violators who have yet to actually be the targets of enforcement, and hope never to be.23 When private enforcement regimes are recognized as a policy instrument deployed by legislators to achieve state objectives, private enforcement litigation is easily integrated into the idea of state capacity. Sociologist Michael Mann’s influential distinction between despotic and infrastructural state power is useful to drive this point home. By despotic power, Mann refers to the “power of state elites over civil society . . . [which] derives from the range of actions that state elites can undertake without routine negotiation with civil society groups.” Infrastructural power, in contrast, is the “institutional capacity” of the state “to penetrate its territories and logistically implement decisions. This is collective power, ‘power through’ society, coordinating social life through state infrastructures.” Mann further observes that “[i]nfrastructural power is a two-way street: It also enables civil society parties to control the state.”24 Private litigants and their attorneys represent a core dimension of the American regulatory state’s infrastructural power in Mann’s sense. Private litigation is state power exercised through society, as Senator Abourezk put it, “by enlisting private citizens as law enforcement officials.” Moreover, private enforcement litigation is infrastructural power in that it enables civil society to control the state by compelling judicial intervention in policy conflicts between private actors, and because it is sometimes directed at state defendants. This form of infrastructural power can penetrate far more deeply into civil society than despotic power. A state whose infrastructural power allows it to “harness the pluralism of civil society by hitching ‘stateness’ to competing political coalitions,” suggests Katznelson, will be regarded by the polity as more legitimate, which will facilitate building extensive capacities.25 Viewing private statutory enforcement litigation from this angle dovetails with an emergent strand of scholarship in American political development that pushes beyond the narrow conception of the state highlighted above toward a study of governance more broadly, emphasizing that policy outcomes are shaped by interactions between the state and private actors. A number of studies, while not self-consciously connected, represent a growing recognition of how state actors mobilize private resources in the service of state building in multiple ways across a wide range of policy domains. Christopher Howard and Jacob Hacker both demonstrate that lawmakers in the United States have used tax incentives to induce the private provision of social benefits, such as health insur-
10
CHAPTER 1
ance, that are typically directly provided by more economically developed democratic states.26 Daniel Carpenter demonstrates that during the Progressive Era entrepreneurial bureaucrats in the United States Postal Service and the Department of Agriculture, by mobilizing the support of powerful networks of private interests, managed to enlarge the mission and authority of their agencies.27 Robert Lieberman shows that, during the early years of the Civil Rights Act of 1964, the Equal Employment Opportunity Commission, lacking formal enforcement powers, actively collaborated with private civil rights groups in an ultimately successful campaign for expansive judicial interpretations of the law.28 These works converge in regarding the boundary between public and private to be porous and, most fundamentally, in recognizing linkages between state and civil society as crucial for comprehending and explaining the development, reach, and efficacy of the American state. “[A] realistic and pragmatic model of state development,” argues William Novak in this vein, “focuses directly on the convergence of public and private authority. The hallmark of American politics from this perspective is the distinctive way in which state power has long been distributed among persons, associations, and institutions that are not easily categorized fully as either private or public.”29 Private statutory enforcement litigation, willfully mobilized by state actors, fits this characterization exactly. Counting the Litigation State Recognizing private statutory enforcement litigation as an important form of state capacity has important consequences for how one understands the reach and strength of the American regulatory state. In the domain of regulatory enforcement, it brings into view causal linkages between the state and a tremendous volume of enforcement activity that in the past has been regarded primarily as private dispute resolution. Currently, in most major areas of federal regulatory policy, private litigants, and not federal agencies charged with enforcement responsibilities, prosecute the overwhelming majority of cases in federal courts. In the past decade, there was an average of about 165,000 lawsuits filed per year to enforce federal statutes in United States district courts.30 These suits spanned the waterfront of federal policy, including antitrust, civil rights, labor and employment, environmental, banking, and securities/commodities exchange regulation. More than 97 percent of the suits were privately filed. At present, the role of private litigation in many important areas of federal policy in the United States is massive both in absolute terms and relative to enforcement by the national government. It deserves emphasis here that this, overwhelmingly, is not the kind of private statutory litigation that political scientists have paid much atten-
INTRODUCTION
11
tion to. Political scientists have shown a fairly keen interest in litigation filed or orchestrated by interest groups, and also in suits filed against government agencies challenging agency policymaking and seeking a court order enjoining or revising policy decisions of administrators.31 However, such suits comprise a very small fraction of total litigation brought under federal statutes. In a stratified random sample of 2,625 published federal court of appeals cases between 1900 and 2004, I find that in only 1 percent of the cases were interest groups either a plaintiff or counsel to a plaintiff, and in only 4 percent of the cases did a plaintiff seek a court order to enjoin or revise agency policy decisions.32 Even looking at the data only after 1960, when interest group litigation and challenges to agency policymaking are generally thought to have become more common, the figures are 2 percent and 5 percent, respectively. Moreover, these small numbers appear in published court of appeals decisions, which likely well exceed the representation of such cases in trial court filings, given that interest group litigation and challenges to agency policymaking are more likely to be cases with high policy salience, and therefore more likely to lead to published court of appeals decisions.33 The vast bulk of private litigation enforcing federal statutes (well over 90 percent) is neither a story of impact litigation by interest groups seeking to make policy, nor of suits challenging the policymaking prerogatives of national authorities. It is, rather, a radically decentralized story of private plaintiffs and their private attorneys pursuing their private interests. It is this kind of litigation that has been largely ignored by political scientists. As discussed in the next chapter, both law and economics and sociolegal scholars have paid considerable attention to ordinary litigation, focusing on the micro-level decision behavior of potential litigants and attorneys. However, they have done so largely in isolation from the broader political processes and institutions in which those potential litigants and lawyers are embedded, processes and institutions that this book argues shape litigants’ and lawyers’ behavior through the vehicle of statutory design. The phenomenon of private enforcement regimes explored in this book ties ordinary litigation that enforces statutes to those broader political processes and institutions. To be sure, at one level the importance of private litigation in American public policy is legend, and the foregoing figures reflecting the volume of private suits enforcing federal statutes will strike many as unsurprising. They may even tempt some to invoke an aphorism meant to show that the centrality of legal rights and courts in the American regime dates to its very founding. How about Alexis de Tocqueville’s famous 1835 dictum: “There is hardly a political question in the United States which does not sooner or later turn into a judicial one.”34 This impulse would be misguided, however. The overwhelming reliance upon private litigation to
12
CHAPTER 1
Litigation rates 40 Private statutory 30
20
10 U.S. plaintiff 1942 1946 1950 1954 1958 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006
Figure 1.1 Private and U.S. Plaintiff Statutory Litigation Rates in Federal Court, 1942–2005
enforce federal regulatory policy is not a static feature of the American state, but rather has varied dramatically over time. Figure 1.1 reflects the rate of lawsuits brought to enforce federal statutes by private parties and by the federal government from 1942 to 2005.35 Between 1942 and the end of the 1960s, the majority of judicial actions brought to enforce federal statutory policy were prosecuted by the national government, not private litigants. Only at the end of the 1960s did private litigation, which had long been an important part of the federal regulatory enforcement apparatus, begin to outstrip governmental prosecutions by a large margin. The interested reader may wish to know that the huge spike in enforcement actions prosecuted by the national government between 1942 and 1948 is accounted for overwhelmingly by suits enforcing the Emergency Price Control Act of 1942, which sought to stabilize prices and rents through the imposition of ceilings, and to ration scarce goods, after the outbreak of World War II. While the United States has relied heavily upon private litigation to enforce policies passed into law by Congress since the rise of the federal regulatory state in the late 1880s, the frequency with which it has done so exploded in the late 1960s. From a rate of 3 per 100,000 population in 1967—a rate that had been roughly stable for a quarter century— it climbed to 13 by 1976, to 21 by 1986, to 29 in 1996, increasing by about 1,000 percent during these three decades. Despite much vitriolic rhetoric, typically focused on tort litigation, serious empirical scholars have not established a “litigation explosion” across American court sys-
INTRODUCTION
13
tems as a whole during this period.36 Figure 1.1 makes it clear that there was, however, an utterly unmistakable explosion of private lawsuits filed to enforce federal statutes. It is well established that the modern American state relies to a vastly greater extent on litigation in policy implementation than most other industrial democratic countries.37 What explains the massive volume of private litigation implementing federal law in the United States? Why has it grown so dramatically since the late 1960s? These questions motivate this book.
Explanations for Rising Litigation Rates The ambition of this book is to illuminate, in general, the institutional foundations of the congressional choice to mobilize private litigants for policy implementation, not specifically to explain the sharp growth in the rate of such litigation beginning in the late 1960s. However, as already suggested, the most important dimension of this institutional account does shed considerable light on this remarkable pattern of growth: legislativeexecutive conflict encourages Congress to enact private enforcement regimes, and such conflict intensified potently starting in the late 1960s. Before discussing these institutional dynamics further, I identify a number of other candidate explanations in the scholarly literature for perceived rising litigation rates, and highlight their inadequacy for explaining the puzzle at hand. The two dominant explanations for perceived rising litigation rates over time might be called the “cultural transformation” and “modernization/economic development” hypotheses. The cultural transformation explanation comes in two varieties, one pejorative and the other celebratory, or at least sympathetic. These cultural transformation explanations must be understood against the backdrop of the contention that American political culture is, to begin with, distinctively litigious. One particularly prominent strand of the notion of “American exceptionalism”—a perspective that sees the United States as differing fundamentally from other developed nations— understands the extensive role of legal rights, litigation, and courts in American policy as arising from an exceptionally individualist American political culture.38 Seymour Martin Lipset, in the fullest exploration of the cultural roots of American exceptionalism, observes that the United States is “a society profoundly rooted in law,” with a “potent orientation toward individual rights,” fostering “the American eagerness for legal settlements to disputes.”39 According to the pejorative cultural transformation explanation, escalating litigation rates in the United States are the result of a process of
14
CHAPTER 1
cultural degeneration from a rights-respecting people to a rights-abusing one. Beginning in the 1970s, legal scholars in particular diagnosed Americans as suffering from a “national disease”40 that prevented them from “tolerat[ing] more than five minutes of frustration without submitting to the temptation to sue.”41 According to then Chief Justice Warren Burger, there was “some form of mass neurosis developing in the country that leads people to think courts were created to solve all the problems of society,”42 and he located the temporal onset of this cultural malady somewhere in the late 1960s.43 During this period, America’s long-standing tradition of individualism morphed into the hyperindividualism of rabid “rights talk,” rights assertion became far more legalistic, and the American people became much more litigious.44 The celebratory cultural transformation story sees—rather than disease and neuroses—the emergence of a new and assertive form of American citizenship during the same period. Deeply influenced by the conjuncture of the civil rights movement on the one hand, and a Supreme Court that elaborated expansive understandings of individual and group rights on the other, there emerged an American citizen characterized both by a heightened state of “rights consciousness,” and by an increasing turn to courts to vindicate those rights.45 The civil rights movement and its accomplishments provided a model for the polity to seize rights through litigation, and this model was extended across the waterfront of American life. This view, it seems, broadly apprehends a similar cultural transformation as that which Chief Justice Berger regretted, but from a radically different normative vantage point. The modernization and economic development explanation finds the cause of increasing litigation in the development of complex modern economies and societies. With respect to economic development, an increasing volume of commercial activity in general engenders much legal disputing and litigation.46 Modern economic activity also entails the proliferation and wide dispersal of risks that result in increases in the kinds of harm for which legal redress is sought,47 while at the same time citizen expectations, fueled by the growing capacities of technology and the state, demand redress for all harms suffered.48 Finally, the corresponding social processes of urbanization and social mobility increase interaction and interdependency among strangers, which drives the demand for intervention by a formal, neutral, authoritative third party to resolve disputes.49 However, explanations rooted in national culture or broad macrohistorical transformations in economy and society cannot account for the changing level of private enforcement of federal statutory policy. While there are a number of problems with these explanations for the pattern of private enforcement of federal statutes, only one is addressed here. Most
15
INTRODUCTION Litigation rates 40 Private statutory 30
20 Tort 10
1942 1946 1950 1954 1958 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006
Figure 1.2 Private Statutory and Tort Litigation Rates in Federal Court, 1942–2005
importantly, such explanations are too global and reductive to convincingly account for sharp differences in the evolution of filing rates across different types of cases over time. This is well illustrated by figure 1.2, which shows the rate of private actions under federal statutes, alongside the rate of tort litigation filed in federal courts, from 1942 to 2005.50 Tort litigation is, by far, the type of litigation subject to the most scholarly study and public commentary by those interested in and concerned about the extent of litigation in the United States. While both types of federal litigation increased significantly over these roughly six and a half decades, the magnitude of the increase of federal statutory litigation vastly outstripped that of tort filings. Moreover, private statutory litigation and tort litigation followed distinct developmental trajectories. In the decade after the end of World War II, from 1946 to 1955, the rate (per 100,000 population) of tort litigation in federal court rose sharply, by about 350 percent, from 2 to 7, while the rate of private statutory litigation declined by 50 percent, from 3 to 2. During the decade from 1968 to 1977, the rate of private statutory litigation more than quadrupled, rocketing from 3 to 13, while the rate of tort litigation held constant at 7 over the same period. Between 1990 and 1996 alone, the rate of private statutory litigation increased by nearly 50 percent, from 20 to 29, during which time the rate of tort litigation held constant at 13.51 It is evident that important elements in the causal explanation for the level of private enforcement of federal statutes will be unique to that
16
CHAPTER 1
category of litigation as distinct from others such as tort litigation, which followed its own path. Propensities to assert legal rights in court driven by transformations in national culture, or by processes of modernization and economic development, apply as much to private law as to public law, and as much to rights under tort law as to rights under statutory law.52 It is simply not plausible that either cultural transformations with respect to legal claiming behavior, or processes of modernization and economic development, are critical in explaining the sea change in private enforcement of federal statutes in the late 1960s while having no similar effects on tort litigation filed in the very same courts. This is, emphatically, not to say that national culture or processes of modernization and economic development are not important. The point is to emphasize, instead, that there are other important forces in play that are obscured by too-easy reliance on such broad-gauged explanatory categories. This book turns to a narrower set of causal explanations—explanations rooted in political institutions—to explain the unique career of private enforcement of federal statutory policy.
Plan of the Book Chapter 2 lays out a theoretical framework for understanding how private enforcement regimes work, and why American state structures encourage Congress to enact them. With respect to how private enforcement regimes work, drawing on the law and economics paradigm for explaining rational litigant behavior in terms of the expected monetary value of claims, the chapter provides a systematic framework for understanding the core statutory elements from which private enforcement regimes are composed. The framework emphasizes the ways in which Congress utilizes these elements as mechanisms to regulate the volume of private enforcement litigation under a statute by effectively setting a market value for lawsuits, which can have the effect of creating an infrastructure of lawyers ready, willing, and able to prosecute enforcement actions. With respect to why Congress enacts private enforcement regimes, the chapter argues that the legislative choice is made by purposive, goaloriented, and strategically minded actors attempting to maximize on policy objectives. It argues, most importantly, that legislative-executive conflict over control of the bureaucracy drives Congress to rely upon litigants and courts as an alternative to administrative power. It further argues that the extreme institutional fragmentation of the American state, making statutes difficult to change once they are enacted, encourages congressional reliance upon private enforcement regimes because they will be difficult for bureaucrats and future legislative majorities to subvert. It
INTRODUCTION
17
argues, finally, that private enforcement regimes can often attract broader legislative support than bureaucratic regulatory state-building, and such broad support can be critical in overcoming the multiple veto players that institutional fragmentation empowers. The chapter also traces the important consequences of the legislative choice of private enforcement regimes for the allocation of power among branches of government in the separation of powers system, and between the state and civil society. In probing evidence on the causes of the congressional choice to mobilize private litigants for policy implementation, this book takes a mixedmethods approach, relying upon a combination of statistical and qualitative historical evidence. Chapter 3 presents statistical analysis of an empirical model of Congress’s propensity to enact private enforcement regimes. It analyzes new data on a measure of Congress’s efforts to mobilize private litigants from 1887 to 2004, testing hypotheses developed in chapter 2 linking American state structures to the congressional choice to enact private enforcement regimes. This data allows evaluation of additional hypotheses concerning legislative mobilization of private litigants that have long appeared in the scholarly literature but have never been tested because of a lack of appropriate data. These theories concern the role of (1) rent-seeking lawyers, (2) issue-oriented citizens groups, (3) the Democratic party, and (4) scarce tax revenues to fund bureaucracy, in driving the legislative enactment of private enforcement regimes. Analyzing a large number of regulatory laws over a long period of time, this model allows evaluation of whether hypothesized causes of congressional enactment of private enforcement regimes, such as legislative-executive conflict, are actually significantly correlated with passage of laws relying upon this form of implementation. While the first three chapters of this book focus on understanding and explaining private enforcement regimes in general, the next three chapters focus closely on the policy area of civil rights in order to shed light on dimensions of private enforcement regimes that can only be discerned in particular policy episodes and contexts. Turning from statistical analysis of patterns in legislative behavior to qualitative historical evidence, chapters 4 to 6 carefully trace the historical record to uncover the key causes of the policy outcome described in the opening paragraphs of this book: massive reliance on private litigation, with very modest administrative powers, in the implementation of federal job discrimination laws. The evidence in this part of the book will lead well beyond the domain of job discrimination and into that of civil rights more broadly. These three chapters provide a detailed historical analysis of the origins and development through time of this approach to implementation, allowing evaluation of whether hypothesized causes of congressional enactment of private enforcement regimes, such as legislative-executive conflict, are
18
CHAPTER 1
visible in the historical events and processes that produced the private enforcement outcome. Some hypotheses about congressional enactment of private enforcement regimes are readily susceptible to validation or rejection through quantitative analysis of correlation between variables, and others are more conducive to illumination through scrutinizing the qualitative historical record reflecting the events and processes that produced an outcome of interest. When the two forms of evidence combine to validate a hypothesis in a mutually reinforcing way, the ability to draw inferences in favor of the hypothesized cause will be especially strong and convincing.53 To anticipate one of the book’s most significant findings, both the statistical model and the qualitative historical analysis will strongly support the conclusion that conflict between legislative and executive preferences encourages Congress to rely upon private litigation for statutory implementation. This evidence strongly links polarization between Congress and the president since the late 1960s to the steep growth of private enforcement litigation during the same period, and correspondingly to the growing empowerment of private litigants, lawyers, and courts in the implementation of American policy.
Chapter 2 INSTITUTIONAL FOUNDATIONS OF PRIVATE ENFORCEMENT REGIMES
Any theoretical account of legislative mobilization of private litigants must address both how Congress mobilizes private litigants and why it chooses to do so. This chapter first addresses how Congress does so, laying out a framework for understanding the way in which Congress crafts the details of private enforcement regimes with the purpose of setting the economic value of claims at a level calculated to mobilize desired levels of private enforcement. It is necessary first to clearly establish how private enforcement regimes work as mechanisms of litigant mobilization in order to understand some of the subtleties of what causes Congress to adopt this strategy of regulatory implementation. Moreover, clearly specifying how Congress mobilizes private litigants will be indispensable to the task, in the next chapter, of developing a concrete measure of when it is endeavoring to do so, for we shall see shortly that merely allowing private litigation in a statute is a far cry from proactively mobilizing it. This chapter then turns to why Congress makes this choice, setting forth an account of the crucial role played by America’s fragmented state structures. The argument builds upon foundations laid in several subfields of political science, drawing together work on the role of American courts in policy implementation (especially Kagan, Melnick, and Burke), rational choice institutionalist work on congressional oversight of bureaucracy (particularly Moe, McCubbins, Noll, and Weingast), as well as historical institutionalist work on bureaucratic and welfare state-building. The most important institutional claim is that Congress frequently struggles with the president over control of the regulatory apparatus of the administrative state, and Congress’s difficulty in controlling the administrative state can cause it to look for alternative vehicles to carry out its will. Chief among these alternatives is reliance upon private lawsuits, which provide a mechanism for Congress to bypass unwilling agencies and opposing presidents. This argument is essential to explaining the precipitous increase in private enforcement litigation starting in the late 1960s. Other institutional incentives for Congress to enact private enforcement regimes are grounded in the notorious stickiness of the legal status quo in the United States. The fragmented character of the American state
20
CHAPTER 2
endows many actors with veto powers in the lawmaking process, making it difficult to pass a new law or change an existing one. Legislative coalitions operating in this institutional environment have reason to provide for regulatory enforcement through private lawsuits. Lawsuits provide a form of auto-pilot enforcement that will be difficult for bureaucrats or future legislative coalitions to subvert, short of passing a new law. Further, when a lawmaking process has many veto players, broad consensus often is required to achieve passage, necessitating extensive compromise. Assertive proposals to enlarge the scope, responsibilities, and power of government frequently encounter opposition from actors with veto powers. This institutional environment encourages reliance upon private litigation as an alternative to bureaucratic state-building. Of course, the potential causes of the legislative choice of private enforcement regimes are likely to be multifaceted, and thus the causal account offered in this chapter, focusing sharply on political institutions, can only paint a partial picture. Part of the next chapter will survey four other accounts of why Congress mobilizes private litigants for regulatory implementation: (1) rent-seeking lawyers’ interest groups lobby to create and maintain opportunities for remunerative litigation so as to enrich themselves; (2) issue-oriented citizens groups, for a variety of reasons that will be discussed, have successfully lobbied for court-based implementation; (3) the Democratic party, acting in the service of important core constituents, is responsible for the frequent use of private enforcement regimes in federal statutes; and (4) lack of adequate tax revenue, and the political undesirability of collecting it, encourages Congress to achieve public policy goals through private adversarial legal process because it shifts the costs of regulation away from the state and to private parties. Thus the partial theoretical account in this chapter, focusing on political institutions, will be filled out in the next. The empirical model presented in chapter 3, and the historical analysis in chapters 4 to 6, both of which endeavor to discern causes of the legislative choice of private enforcement regimes, will attend to the multiple causal theories discussed. Several of the four causal accounts just enumerated, we shall see over the course of this book, are importantly entwined with the institutional arguments set forth in this chapter. The arguments developed in this chapter—both about how private enforcement regimes affect the behavior of potential plaintiffs and attorneys, and about why Congress enacts them—represent relevant actors as rational, strategic, and purposive, and as capable of apprehending cause and effect in a complex political, social, and economic world reasonably well. In the course of the chapter, justifications for these assumptions of rationality and predictability will be offered, and in later chapters evidence to support them will be provided. This is not to say that these as-
INSTITUTIONAL FOUNDATIONS
21
sumptions represent the whole story, but rather only that they are an important part of it, and that a model based upon them has considerable explanatory power. Important limitations to these assumptions are discussed at the end of this chapter.
Legislative Instruments of Legal Mobilization This book’s claim about the capacity of legislators to exercise control over the volume of private enforcement litigation entails a claim about the rationality and predictability of the choice to litigate. If prospective litigant-enforcers do not respond in fairly predictable ways to the rules and procedures that comprise a private enforcement regime, then private enforcement regimes cannot serve as policy instruments. Law and economics scholars have long recognized the influence of legal rules and procedures on the choice to litigate, and, theoretically, have regarded the choice as one that could be manipulated through adjustment of such rules and procedures (a literature discussed shortly). This basic idea was integrated into a model of political decision-making by members of Congress endeavoring to control the bureaucracy, as McCubbins, Noll, and Weingast put it, by using “administrative procedures as instruments of political control.”1 These scholars argue that Congress responds to principal-agent problems inherent in delegation of authority to agencies by using administrative rules as mechanisms to keep policy from shifting away from congressional preferences. Legislators use procedures such as notice-andcomment rulemaking, and the information disclosure requirements under the Freedom of Information Act, to allow interested constituents points of entry into the political process to ensure that relevant information is brought to light (and to legislators’ attention) in the course of policymaking. Further, Congress “stacks the deck” in favor of intended beneficiaries of legislation by specifying statutory procedures such as rules of standing, evidence, proof, and judicial review that make it more probable that the intended beneficiaries will prevail in agency proceedings or in court challenges to agency decisions. The upshot is that Congress uses administrative rules and procedures to harness the energies and resources of private actors to achieve the purpose of controlling agency policymaking. While this line of research recognizes judicial review provisions as instruments that Congress uses to police agencies,2 it does not address the use of private enforcement regimes to mobilize direct enforcement against the objects of regulation. The oversight literature is about “how to regulate the regulators,” explain McCubbins and Schwartz, and “not how to regulate society.”3 In contrast, the subject of this book is precisely
22
CHAPTER 2
the regulation of society through the use of direct enforcement by private litigants as an alternative to bureaucratic power, not as a mechanism to monitor agencies. The private lawsuits focused on by the oversight literature are suits against government agencies seeking to enjoin or reverse agency policymaking. As noted in chapter 1, these suits comprise a tiny fraction of statutory enforcement actions. This book’s subject is the vastly larger number of private suits filed to enforce federal statutes against alleged violators—for example, a privately filed wage suit against Walmart, shareholder action against Enron, antitrust action against Microsoft, or consumer protection claim against Verizon. In order to systematically conceptualize the ways in which Congress controls the volume of private enforcement litigation, the chapter draws on the model of rational litigant behavior elaborated in the law and economics literature.4 This model generally contemplates a multistage process in which the prospective plaintiff alleging injury resulting from a violation of the law must decide whether to sue, the defendant must decide whether to make a settlement offer, and if so, for how much, and the plaintiff must decide whether to accept the settlement offer or opt instead to go to trial. According to this model, both parties are guided in their decisions at each stage by the expected monetary gain or loss should the case be tried. The expected monetary value of a claim from the plaintiff’s point of view is a function of the plaintiff’s estimate of the expected monetary benefit of the case if she prevails (EB), the probability that she will prevail if the case goes to trial (p), and the expected costs of litigating the claim (EC). Thus, EV = EB(p) − EC, and the rational plaintiff will file suit if the expected monetary value of the calim (EV) is positive. To be sure, the choice of whether or not to sue may be influenced by forms of utility or disutility distinct from and not reducible to money. Past research suggests that the choice to sue may also be influenced by utility from telling one’s side of the story in a conflict,5 utility derived from litigation as a form of political participation,6 or disutility resulting from feelings of embarrassment or victimization.7 Thus, the law and economics model of the choice to litigate is, without doubt, highly stylized, but usefully so from the standpoint of the puzzle that motivates this book. The chief purpose of this book is to identify and explain the causes and consequences of choices, when drafting statutes, made by legislators, and the interest groups that influence them, that determine private enforcement levels. The book maintains that a key lever legislators use to accomplish this purpose is to determine the expected value of claims (EV) by manipulating expected benefits (EB), probability of prevailing (p), and expected costs (EC) in the above model. The psychological pleasure or pain that plaintiffs experience as a result of litigation is vastly less within
INSTITUTIONAL FOUNDATIONS
23
Congress’s control than the expected monetary value of claims, and I am aware of no evidence that legislators pay significant attention to attempting to control such psychological payoffs or losses. Thus, the law and economics model of the choice to litigate is stylized in a manner that isolates not just any component of the choice to litigate, but rather the one most essential to the investigation undertaken in this book. The economic motive to litigate is, moreover, not just one coequal among many, but rather is a dominant one in the sense of being a threshold condition in the vast majority of cases. Whatever other motives inform a prospective plaintiff’s choice of whether or not to litigate, very few plaintiffs will be willing or able to do so under an expectation of incurring a substantial financial loss. This is not a claim that plaintiffs are driven by greed, but rather that an expectation of a positive economic outcome, rather than an expectation of suffering economic loss, will typically be a precondition to a plaintiff choosing to proceed with litigation, even if there are other political or psychological motives for litigating. As Robert MacCoun, studying litigation from a cognitive psychological perspective, put it in reference to the effect of monetary damages on the choice to litigate: “even a boundedly rational psychological model will assume that expectations play a central role in choice.”8 Further, the choice to litigate will not be left to the plaintiff alone, but will typically also require the agreement of an attorney. Given that plaintiffs’ lawyers in the United States are often dependent upon the profitable resolution of a case for some or all of their compensation, unless a plaintiff is willing and able to bankroll (often massive) litigation costs on her own, before filing suit her attorney will have to evaluate whether to invest limited resources in the case based upon an assessment of the case’s risks and potential returns.9 This is an economic judgment that will broadly follow the law and economics model. In making this claim, this book’s central focus is on ordinary enforcement litigation prosecuted by attorneys working in the for-profit sector, not on interest groups prosecuting impact litigation. Law reform organizations with external sources of funding, which strategically select cases to achieve long-range policy goals, can often litigate without an expectation of economic reward, and such litigation can be profoundly consequential. This is certainly not to say that economic recovery is insignificant to interest groups; as we will see in part II of this book, economic recoveries can provide significant resources to support and encourage the litigation activities of advocacy organizations. However, the profit motive will surely figure more prominently among lawyers in the forprofit sector. As discussed in chapter 1, interest group litigation comprises a minuscule fraction of total statutory enforcement litigation, with forprofit attorneys prosecuting the overwhelming majority of claims. The
24
CHAPTER 2
law and economics model of the choice to litigate is adopted here in order to explain ordinary claims by private parties and their for-profit attorneys, which represent the vast bulk of statutory enforcement actions. As Cooter and Ulen put it, in this context the bottom line is that “[n]ot every plaintiff with a cause of action can sue profitably,” and thus prospective plaintiffs and their attorneys contemplating suit must ask, “When does it pay to file suit?”10 Congress determines when it pays to file suit under a federal regulatory statute. It does so by deciding whether and to what extent to construct a private enforcement regime as part of the law. The motivation in using the law and economics model of the choice to litigate (EV = EB(p) − EC) is to systematically illuminate the ways in which Congress manipulates the expected monetary value of lawsuits for the purpose of regulating the volume of private enforcement litigation. It bears emphasis that this is a quite different explanatory goal than that of law and economics scholarship. Working from a perspective of new institutionalism in political science, this book stresses the political, strategic, and policy forces behind choices of statutory design, rather than assessing what combinations of rules produce efficient outcomes under various assumptions. Indeed, the argument of this book suggests that political variables may be pivotal in explaining outcomes that are quite inefficient from an economic point of view. Consider a case in which a plaintiff is evaluating whether to file an action for violation of a statute. She believes that she has a probability of prevailing (p) of .5, with an expected benefit (EB) of $1,000 if she prevails, and an expected cost (EC) of $750 to litigate the case through to final judgment. From the prospective plaintiff’s point of view, the expected value of the case (EV) = $1000(.5) − $750, or negative $250, and she will not file because she believes that doing so will lead to a financial loss. The main point to stress at this stage is that, from the point of view of drafting legislation, the expected value (EV) for the same case can be made positive by changes to the statute that work changes to expected benefits (EB), probability of prevailing (p), or expected costs (EC), or some combination of these changes. Relative to the above example, if legislators are able to increase the expected benefits (EB) of winning the same case to $2,000, or to increase the plaintiff’s probability of winning the case (p) to .8, or to eliminate the plaintiff’s costs of litigating the case if she prevails (EC), then the expected value of the case (EV) will become positive and the plaintiff will proceed with filing, as reflected in table 2.1. Each of the three terms in the equation used to calculate expected value (EV) represents an analytically distinct category of mechanisms that together comprise a private enforcement regime. In the discussion of each category below, Title VII of the Civil Rights Act of 1964, which prohib-
25
INSTITUTIONAL FOUNDATIONS
Table 2.1 Calculating Expected Value of Suit EB
*
p
−
EC
=
EV
File suit?
$1,000
*
.5
−
$750
=
−$250
Don’t file
$2,000
*
.5
−
$750
=
$250
File
$1,000
*
.8
−
$750
=
$50
File
$1,000
*
.5
−
$375
=
$125
File
its certain forms of employment discrimination, will be examined as a principal illustration, since its private enforcement regime comes extensively into play later in this book. Further, each element of Title VII’s private enforcement regime will be contrasted with other statutes in which Congress made very different choices. The point of these contrasts is to emphasize the extremely broad continuum of options available to legislators, and the wide range of choices that Congress has in fact long made, in specifying the details of each term of the equation EV = EB(p) − EC. Expected benefits (EB), probability of prevailing (p), and expected costs (EC), are continuous rather than dichotomous in nature, and thus legislators can make very fine-grained decisions of degree in constructing precisely what the incentive structure will be when a plaintiff and her attorney assess the expected value of a claim (EV). While the discussion below touches upon many of the most important elements of private enforcement regimes, it is ultimately illustrative only, for any attempt to map the entire universe of statutory attributes that could be incorporated into a private enforcement regime would surely try the reader’s patience. The discussion does, however, provide considerable detail, for the instrument of private enforcement regimes is at the center of this book from start to finish, and thus it will pay to be very clear at the outset about what that instrument is and how it works. Before discussing Congress’s construction of expected benefits (EB), probability of prevailing (p), and expected costs (EC), there is an additional and logically prior crucial dimension of private enforcement regimes not captured by the foregoing model of the choice to litigate, and that concerns whether an individual or entity is given access to plaintiff status at all. Given the creation of a substantive legal rule—for example, employers must not discriminate based upon race—Congress must also make rules about whether an individual or entity has a private right of action, which allows one to bring an enforcement action against a violator. Though Congress gave the right to sue to any “person claiming to be aggrieved” by unlawful discrimination under Title VII,11 it does not always do so under regulatory statutes. When Congress enacted the Oc-
26
CHAPTER 2
cupational Safety and Health Act of 1970, it elected not to allow workers injured by violations of the act to bring suit under it, and instead vested all enforcement authority with an agency.12 Between these two poles—standing for any aggrieved private party on the one hand, and no private party standing whatsoever on the other— there are many opportunities for Congress to delineate more finely the size of the pool of potential enforcers. For example, if Congress wanted to provide a private cause of action against entities who fraudulently label articles with respect to their content and quality, it could confer standing upon one, all, or some combination of the following categories of private plaintiffs: the defrauding party’s (1) competitors, (2) customers, (3) competitors of customers, (4) any subsequent purchaser of an article in the stream of commerce, and (5) trade associations. Congress elected explicitly to confer standing upon all five groups in 1970 amendments to the National Gold and Silver Stamping Act of 1906, which had not previously had any private right of action, for the express purpose of increasing enforcement pressure.13 Aside from the core decision of whether private parties will be permitted to enforce a statute in court, and if so, precisely which private parties, additional aspects of a private enforcement regime can affect access to plaintiff status. Perhaps most important among these aspects are statutes of limitations, which operate to deny the right to bring suit even for egregious and uncontested violations of the law if the violations occurred outside the limitations period. Under Title VII the limitations period is 180 days,14 which is very short in comparison to most federal statutes. By contrast, in the False Claims Act of 1863, which provides for private actions against persons who defraud the federal government of money— where the private plaintiff is awarded some share of the money recovered for the government—Congress provided for a lengthy six-year limitations period, which continues in operation to the present day.15 Thus, a federal civil rights claim under Title VII forever expires after six months, while a claim to recover money for the federal government lives on for six years. The larger the limitations period window, in general, the larger the population of potential claims included within it. Expected costs (EC) are the sum of the filing fee, which is generally small in the United States, attorney’s fees, and other costs of litigation, such as expenditures on expert witness fees and the conduct of discovery. The standard “American rule” is that each party pays its own attorney’s fees and other costs of litigation, whereas the “English rule”—which prevails in Europe, and most of the rest of the world—provides that the loser pays most of the winner’s fees and costs.16 As discussed more fully in chapter 3, Congress sometimes departs from the American rule in regulatory legislation and opts instead for a rule under which winning plaintiffs
INSTITUTIONAL FOUNDATIONS
27
may recover the costs of enforcement, including attorney’s fees, while similar recovery is not granted to winning defendants, which this book refers to as a “one-way plaintiffs’ shift.”17 There is a considerable theoretical literature on the comparative effects of different fee- and cost-shifting rules on litigation and settlement rates.18 While scholars disagree about the comparative effects of the American and English rules, it is clear that as compared to either the American rule or the English rule, under a one-way plaintiff’s shift, no matter who wins, expected costs (EC) will be equal or less than under the other systems for allocating the expenses of litigation, causing the expected value of the case (EV) to be equal or greater.19 Thus, among the alternative arrangements for allocating responsibility for paying litigation expenses, the one-way plaintiff’s shift creates the greatest incentives for plaintiffs to file enforcement actions.20 Congress included a fee-shifting provision in Title VII of the CRA of 1964, with the express purpose of stimulating private enforcement actions.21 Expected benefits (EB) are determined to an important extent by rules governing how much monetary damages a plaintiff will be able to recover, which directly increases the expected value of a case (EV), and with it the incentives for plaintiffs and their attorneys to file enforcement actions.22 Legislators have wide latitude to determine whether statutory cases, if won by plaintiffs, will be worth no money or a little money or a lot of money. In Title VII of the CRA of 1964, with respect to monetary damages, legislators elected to allow only for back pay in the way of monetary damages.23 Under this rule a woman denied a promotion to a position associated with a salary increase of $1,000 per year would stand to recover damages of $1,000 per year for the permissible period of recovery. Congress could have, if it wished, provided for damages in the amount of double the wages actually withheld, as it did in Equal Pay Act of 1963,24 or of triple the actual economic damages, as it did in the Sherman Antitrust Act of 189025 and the Emergency Price Control Act of 1942,26 which would have had the effect of doubling or tripling expected benefits (EB) for the hypothetical promotion discrimination claim described above. Probability of prevailing (p) is, of course, strongly influenced by the facts of each case, or how strong the evidence is that the defendant violated the law. But given a set of facts that present an arguable violation, the rules governing burdens of proof (who has to prove what), and standards of proof (to what degree of probability), can be potent determinants of probability of prevailing (p), and thus expected value (EV).27 Likewise for rules governing what evidence the parties are permitted to demand in the course of discovery and to introduce at trial.28 In the Interstate Commerce Act of 1887, for example, Congress included an express pro-
28
CHAPTER 2
Table 2.2 Elements of Private Enforcement Regimes Access to plaintiff status Private plaintiff standing Definition of plaintiff Statutes of limitations Claim aggregation devices (e.g., class actions)
Expected benefits
Probability of prevailing
Expected costs
Monetary damages Burdens of proof Filing fee rules (e.g., double, Standards of proof Costs of litigation triple, punitive, compensatory) Rules of evidence Attorney’s fees Injunctive relief
Judge versus jury Liability rules (e.g., joint and several liability)
vision that Interstate Commerce Commission findings of rate violations by a railroad could be offered in evidence by plaintiffs in private actions against the railroad. Courts were required to treat such commission findings as prima facie evidence of a violation—meaning that the commission findings would be regarded as proven unless rebutted—effectively shifting the burden of proof to the defendant, and significantly increasing the probability of prevailing (p) from the plaintiff’s point of view.29 Legislators made abundantly clear in floor debates that it was their intention to facilitate successful private litigation by making commission fact-finding available as proof to private plaintiffs.30 Conversely, the CRA of 1964 barred the use of evidence gathered by the EEOC during an investigation in any subsequent proceeding.31 It also included no provision directing courts to give any weight at all to EEOC findings in subsequent litigation, and in the absence of such a statutory directive courts have largely concluded that EEOC findings are not even presumptively admissible as evidence, and that even to the extent that they are admissible, they are entitled to only limited, and not burden-shifting, weight.32 To summarize, each attribute of a private enforcement regime discussed above, and some not discussed, is listed in table 2.2, grouped under the major organizing categories of access to plaintiff status, expected benefits, probability of prevailing, and expected costs. With this framework in hand, a final example illustrates the potentially potent effects of private enforcement regimes on incentives to litigate, as well as the cumulative effects of different attributes of private enforcement regimes. Consider a potential plaintiff who sustained $10,000 in actual damages because of an arguable violation of a statutory right, and assume that $10,000 is typical of damages caused by violations of this
29
INSTITUTIONAL FOUNDATIONS
Table 2.3 Legislative Manipulation of Expected Value
Hypothetical
Expectected benefits
Probability of success
Expected costs
Expected value
(1)
(Actual damages) $10,000
(No burden rule) .6
(American rule) $10,000
−$4,000
(2)
(Treble damages) $30,000
(No burden rule) .6
(American rule) $10,000
$8,000
(3)
(Treble damages) $30,000
(Proplaintiff burden rule) .8
(American rule) $10,000
$14,000
(4)
(Treble damages) $30,000
(Proplaintiff burden rule) .8
(Plaintiff’s fee shift) $2,000
$22,000
particular statute. Further assume that only actual monetary damages are recoverable, the American rule applies to attorney’s fees (no shift), and default burdens and standards of proof apply. The plaintiff and her attorney estimate her probability of prevailing as .6, and estimate the costs of litigating through to final judgment as $10,000. These facts, which are reflected in hypothetical 1 in table 2.3, yield an expected value of negative $4,000, such that the plaintiff will not file suit. Under this scenario, Congress has elected to provide a private right of action, but beyond simply allowing private litigants to enforce the statute, it has not particularly sought to affirmatively mobilize them. It is important to register the significance of this point. Legislators can elect to use private litigation for enforcement, while structuring the private enforcement regime so as to ensure that there will actually be very little enforcement. In hypothetical 2, the legislature included a treble damages provision, so that expected benefits are now increased from $10,000 to $30,000, and expected value is increased from negative $4,000 to positive $8,000, rendering a positive incentive for the plaintiff to file suit. In hypothetical 3, in addition to the treble damages provision, the legislature adds an explicit pro-plaintiff statutory burden of proof rule, which increases the plaintiff’s probability of prevailing (p) from .6 to .8, increasing the expected value of her case to $14,000. Finally, in hypothetical 4, Congress also adds a one-way plaintiff’s shift to the statute, which has the effect of reducing expected costs (EC) from $10,000 to $2,000 (since the plaintiff will be able to recover those costs if she wins, an outcome to which she assigns a probability of .8). This increases the expected value of the case commensurately, by $8,000, to $22,000. As we go from hypothetical 1
30
CHAPTER 2
(a minimal private enforcement regime) to hypothetical 4 (a robust one), the expected value of the case (EV), from the plaintiff’s point of view, increases from negative $4,000 to positive $22,000. The four hypotheticals represent four (among a vastly larger number) configurations of private enforcement regimes, each corresponding to a successively higher degree of mobilization of private enforcers, and each the product of legislative choice. Chapter 6 will discuss evidence demonstrating that such legislative choices are actually effective in influencing the behavior of potential private enforcers in the manner hypothesized. When such legislative choices are made in Congress, they are embedded within and structured by a set of institutional opportunities and constraints. The next section of this chapter examines the ways in which American state structures encourage the legislative choice of private enforcement regimes. In concluding this section it is important to draw out a connection between the argument developed so far and a line of research in the field of law and society concerning the significance of certain forms of legal infrastructures to the elaboration and protection of rights through courts. Marc Galanter famously argued that “repeat players” (who use the courts frequently, and are typically “haves”) possess significant advantages over “one-shotters” (who use the courts infrequently, and are typically “have-nots”).33 The advantages importantly include access to greater legal resources, skill, and expertise in the litigation process, and Galanter speculated that one way to counterbalance the inequality between repeat players and one-shotters could be through legal rules and policies that have the effect of increasing the supply of quality legal services for havenots. Building on this line of research, in a cross-national study focusing on civil rights and civil liberties, Charles Epp demonstrates that successful “rights revolutions” were more likely in countries characterized by more extensive “support structures” for legal mobilization, particularly activist organizations, skilled and committed lawyers, and sources of funding to support litigation campaigns.34 The logic of private enforcement regimes highlights that the existence of an important support structure to enforce regulatory statutes—a bar of ready, willing, and able lawyers—can be profoundly influenced by legislation. As discussed above, before deciding to prosecute a lawsuit, attorneys often must evaluate whether to invest limited resources in a case based upon their assessment of its potential financial risks and rewards. When the results of this process for selecting cases are aggregated across the legal system, the legislative creation of opportunities and economic rewards for lawyers to enforce regulatory statutes can importantly contribute to the growth of a bar of lawyers possessed of the skill and desire to execute the function of enforcing them. That is, when legislators craft
INSTITUTIONAL FOUNDATIONS
31
enforcement markets with sufficient incentives in a regulatory statute, they can stimulate the creation of support structures for implementation in the form of an infrastructure of private prosecutors who earn a living, at defendants’ expense, practicing in the relevant area of law.35 This argument extends far beyond litigation by have-nots, and beyond the policy domain of civil rights and civil liberties. It is an argument about regulation through private enforcement regimes in general.
Institutional Fragmentation and Private Enforcement Regimes When Congress enacts a law to regulate some facet of economic or social life, it must make choices about how the law will be implemented. One classic characterization of a basic regulatory choice juxtaposes the creation of decentralized incentives to shape behavior through voluntary choice, such as tax rewards for desired behavior, and the creation of a “command and control” system to police for violations and punish violators.36 The first type of system relies upon rewards, and the second upon punishment. While regulation through systems of command and control is commonly associated with regulation through bureaucracy, the use of private enforcement regimes is also a form of command and control. As argued in chapter 1, in comparison to bureaucratic regulation, the use of private enforcement regimes entails no less commanding, controlling, or coercing, but rather is distinguished by delegating some of those powers of coercion to private actors pursuing private ends—albeit private ends constructed by Congress. Thus, when Congress enacts commandand-control regulatory legislation, it faces a choice between enforcement through bureaucratic machinery and the use of litigation, or what Morris Fiorina called the fundamental choice between “administrative process” and “legal process.”37 Fiorina’s dichotomy between administrative process and legal process as forms of command and control regulation is certainly stylized, but usefully so. To be sure, the two forms of regulation, while they are often used independently of one another, are also used in combination, with some powers being delegated to administrative actors, while others are left to private litigants and courts within the same statute. In hybrid regimes, for example, either administrative process or legal process can be given the dominant role, with the other playing a more ancillary one; the two forms can be given important and distinctly separate roles in a regulatory scheme, such as by authorizing administrators to act as prosecutors in court, or they can be given substantially overlapping roles; the two regulatory forms can be treated as substitutes for one another, or can be
32
CHAPTER 2
used in complementary fashion. The range of possible combinations of administrative and legal process is substantial and complex, and numerous concrete examples of hybrid regimes will be examined in part II of this book. Despite this complexity, however, any command-and-control regulatory enforcement regime distributes power between administrative process and legal process, and can be characterized in general terms as falling somewhere on a continuum between pure legal process and pure administrative process. The theoretical framework below is developed based upon the stylized choice between bureaucratic implementation and private enforcement regimes, as alternatives, in order to gain analytical leverage on and crystallize the core differences between the two regulatory forms. The insights gained will be no less useful for understanding the relative weight given to bureaucracy and private lawsuits in hybrid regimes than for understanding the choice between the two pure forms. Private Enforcement Regimes and Regime Politics Before the discussion turns to the specific linkages between American state structures and the legislative choice to use private litigation in policy implementation, it will be helpful to set the argument in the context of a stream of research in American politics situating courts in relation to nonjudicial actors’ pursuit of their policy agendas. The study of American courts by political scientists has long been dominated by analyses of appellate decision-making by judges, and particularly the degree to which it is driven by judges’ personal ideological preferences. There is a growing line of scholarship that seeks to shift from a focus on the judiciary as an autonomous bubble, in which judges make policy according to their own preferences, to a focus on the judiciary as an institution that is used, strategically and instrumentally, by the elected branches in support of their political and policy agendas. This line of research, often referred to as a “regime politics” approach to the study of courts, traces its lineage to early work by Robert Dahl and Martin Shapiro.38 It endeavors to retrieve courts from scholarly isolation from the broader field of American politics, and to integrate them into more general theories and understandings of how governing coalitions exercise power.39 The present study is aligned with and builds upon these themes, adding private litigation to the story of regime politics. This literature has identified a number of ways that an independent judiciary can and has served the interests of governing coalitions in the elected branches, three of which are particularly relevant here. First, focusing on constitutional doctrine, Balkin and Levinson suggest that the appointment of a sufficient number of ideologically kindred lifetime ten-
INSTITUTIONAL FOUNDATIONS
33
ured judges, which can long outlast the coalition that appointed them, can allow that coalition to “entrench” its policy agenda. They emphasize that this strategy will become more attractive as the governing coalition’s electoral future grows more uncertain.40 Second, Whittington argues that an independent judiciary with powers of judicial review can aid a governing coalition that is still in power by making policy choices that the coalition desires but is unable to achieve because of countermajoritarian “obstructions” arising from, for example, veto powers conferred upon electoral minorities by fragmented political institutions.41 Third, Landes and Posner have argued, influentially, that judges—regardless of who appointed them—can also serve legislators by applying statutes according to “norms legality,” most significantly by interpreting statutes enacted by past governing coalitions according to the intentions of those coalitions. This serves the interests of legislators in general, including current legislators who are ideologically opposed to past legislative bargains, because their ability to make a “credible commitment” that legislative bargains will endure beyond the life of the enacting coalition makes legislation far more valuable to interest groups, and thus allows legislators to extract a higher price for it.42 Each of these accounts of how courts can serve broader political regimes involves some supposition about the content of judicial decisions in relation to legislative preferences. In the “overcoming obstructions” scenario the instrumental utility of the judiciary depends upon its rendering substantive decisions in accord with the preferences of the current ruling coalition. Courts friendly to current majorities can help them overcome obstructions to governance. In the “political entrenchment” scenario the instrumental utility of the judiciary depends upon its rendering decisions in accord with the preferences of the coalition out of power, and presumably disfavored by the current ruling coalition. Courts friendly to past majorities, now defeated, can continue to press their agenda. In the “credible commitment” scenario, the instrumental utility of the judiciary depends upon its rendering decisions in accord with the preferences of the enacting coalition, whether in or out of power, and this norm serves the long-run interests of current legislators, even if they dislike the shortrun outcomes in particular cases that apply statutes enacted by their predecessors. Legislative enactment of private enforcement regimes is similar to these accounts in some ways, and critically different in others. Straightforwardly, when ruling coalitions enact private enforcement regimes in pursuit of their policy goals, they use the judiciary as an infrastructure for implementation of their agenda. As we shall see later in this chapter, when legislative motivations for this choice are assessed, private enforcement regimes can at times facilitate political entrenchment, at times help
34
CHAPTER 2
to overcome obstructions to lawmaking, and at times enable credible commitments. However, the instrumental utility of private enforcement regimes does not depend upon courts rendering decisions in accord with the preferences of past or present legislative majorities, or with the enacting coalition’s intent. The instrumental utility of private enforcement regimes depends upon the mobilization of private litigants and their attorneys to execute the prosecutorial function. This will, of course, bring the ideological preferences of courts into play, but as we shall see at the end of this chapter, a particular supposition about judicial behavior will not be a necessary precondition to the attractiveness of private enforcement regimes from Congress’s point of view. Private Enforcement Regimes and Legislative-Executive Conflict To understand this legislative choice, we must situate it within the institutional context of American state structures. In his foundational Adversarial Legalism, Robert Kagan has suggested that the large role of adversarial legal process in American public policy is rooted partly in the “weak” and “fragmented” character of American state structures, which are characterized by crosscutting institutional checks and the dispersion of authority across executive, legislative, and judicial branches.43 Kagan stresses that, notwithstanding these limitations on the capacity of the American state to implement ambitious public policy, the American polity demands a robustly interventionist and protective regulatory state.44 Consequently, Kagan argues, the large role of litigation in American public policy is driven in part by the mismatch between public demand for an activist state on the one hand, and a weak and fragmented administrative state on the other. Americans use litigation to try to get from courts what they cannot get from their hapless administrative state. Thomas Burke provides an important development of Kagan’s work, focusing more closely on the ways in which institutional fragmentation, identified by Kagan as a source of adversarial legalism, provides incentives for interest groups and policymakers to purposefully structure laws so as to encourage litigation as an alternative to bureaucracy. Burke observes that the work of Terry Moe on congressional choice of bureaucratic structure provides a useful analytical frame for specifying key institutional dynamics through which the American separation of powers system generates high levels of private litigation to enforce public policy, and I draw upon Burke’s discussion.45 A central theme of Moe’s work is that when creating agencies, rational legislators in the United States make choices about agency structure and procedure meant to insulate their preferences from political opponents who would subvert them in both the short and long run.46 The most fundamental and persistent threat of
INSTITUTIONAL FOUNDATIONS
35
subversion is subversion by the president, who has distinct institutional interests, and potentially divergent ideological preferences. This threat of subversion of administrative implementation is critical to understanding the legislative choice of private enforcement regimes. Moe argues that, even aside from ideological differences between Congress and the president, there are fundamental institutional divisions that will give the two branches different preferences regarding the exercise of bureaucratic authority. As compared to presidents, legislators are influenced more by particularistic than by national interests and are more subject to interest group pressure, differences that can lead to divergent preferences over regulatory implementation.47 Further, while legislators certainly have significant continuing power over agencies’ actions,48 presidents nevertheless possess considerable capacity to unilaterally influence agencies’ structure and behavior.49 Thus, legislators and the interest groups that influence them strive to create agency structures calculated to implement their policy preferences while tightly constraining bureaucratic discretion so as to insulate it, to the extent possible, from presidential subversion. Examples include writing extremely detailed laws that limit administrative discretion, limiting the number and power of political appointees, vesting more power instead in career civil service personnel, and denying the president hierarchical instruments of control over the agency.50 To the extent that these structural dynamics are driving Congress’s construction of the character and capacities of the American administrative state, the relationship should be intensified with increasing ideological conflict between Congress and the president. If institutionally rooted differences produce some baseline divergence between the preferences of the president and those of legislators, then ideological conflict between the two will exacerbate this divergence. The more congressional and presidential ideological preferences diverge, the more likely the president will be to use his significant institutional resources to subvert implementation of congressional policy choices, and the more likely Congress will be to constrain and limit delegations of power to the bureaucracy. Empirical research strongly bears out this prediction. Epstein and O’Halloran demonstrate that Congress does, indeed, respond to the ideological position of the president when deciding how much discretion to delegate to agencies and when designing agency structure.51 They find that under conditions of divided party government, Congress enacts more detailed laws, thus limiting agency discretion in implementation, and places more structural constraints on the exercise of bureaucratic implementation authority. Similarly, in studies comparing American states, researchers have found that divided party government between the executive and legislative branches leads legislators to enact more detailed laws and thus
36
CHAPTER 2
to delegate less discretion to bureaucrats,52 and to enact more structural constraints upon the exercise of bureaucratic power that is delegated.53 This institutional logic for delegating less authority to the bureaucracy, and structurally constraining its exercise of the powers delegated, simultaneously motivates Congress to enact private enforcement regimes.54 To the extent that Congress has concerns about whether the president will pursue an enforcement agenda inconsistent with congressional policy preferences, because of the distinct institutional and electoral imperatives of the presidency, Congress has reason to enact incentives for private actors to do so. To the extent that this structural cause for enactment of private enforcement regimes is in fact at play, it will be intensified under conditions of ideological conflict between Congress and the president. This is the flip side of the delegation literature just discussed. Under conditions in which that literature has found legislators delegating less implementation power to the bureaucracy—ideological conflict between the legislative and executive branches—legislators do not abandon implementation. Rather, under those conditions legislators marshal other resources to achieve desired levels of enforcement, including private litigants, who are substantially beyond the reach of presidential influence.55 This institutional logic suggests that the more ideologically distant Congress is from the president, the more likely it will be to enact private enforcement regimes. This hypothesis will be tested in the next chapter. This theoretical claim is not only about elevating the extent of enforcement with private enforcement regimes, as compared to bureaucratic implementation, but also about controlling the level of enforcement, and this is a critical point that has been neglected. As an enforcement strategy, private enforcement regimes can protect against either under- or overenforcement by bureaucrats. As discussed above, table 2.3 reflects the variability of private enforcement regimes, which, like bureaucratic regimes, can range from weak to strong. A private enforcement regime with limited opportunities and incentives for private enforcement can be calculated to produce less intervention than an overzealous agency (from Congress’s point of view), and a robust private enforcement regime can be calculated to produce more intervention than an agency disinclined toward vigorous implementation (from Congress’s point of view). Finally, it bears emphasis that the claim here is not that legislativeexecutive relations in a separation of powers system encourages Congress to enact private enforcement regimes only under conditions of ideological conflict with the president, but rather that those conditions will make it more likely to do so. The institutionally rooted difference between the preferences of legislators and presidents in the separation of powers sys-
INSTITUTIONAL FOUNDATIONS
37
tem can make private enforcement regimes appealing to Congress even when the president is an ideological ally, but their appeal will multiply when the president is an enemy. Thus, theory predicts that the American separation of powers system encourages enactment of private enforcement regimes in general, and that their use will be elevated under conditions of ideological conflict between Congress and the president. What about judicial ideology? If it is true that legislators sometimes regard bureaucracy and courts as alternative implementation venues, and that Congress is concerned about executive subversion of its policy preferences when confronting this choice, then the question naturally arises whether Congress is similarly concerned about judicial subversion of its policy preferences. How to incorporate judicial ideology into this discussion, however, turns out not to be theoretically straightforward. This question is deferred until later in this chapter because some intervening concepts will be helpful in unpacking it. Private Enforcement Regimes, Coalition Drift, and Uncertainty Moe identifies electoral uncertainty as a second potential source of subversion of Congress’s policy preferences that motivates it to insulate its decisions through strategic choice of bureaucratic structure. Members of Congress are famously mindful of the prospects of electoral defeat,56 and they and their constituents recognize the possibility that rival political forces may gain control of Congress in the future and seek to undo the good works of the enacting Congress.57 That is, in future periods the preferences of Congress may drift away from those of the enacting coalition. Kenneth Shepsle observes that, from interest groups’ point of view, this risk of “coalition drift” is present even aside from the question of electoral uncertainty because the existing legislative coalition’s preferences may be subject to change the future—most likely in response to shifts in demand by the public and interest groups—even if it remains in power.58 As a result of uncertainty about future legislative preferences in a democratic society, according to Moe, an enacting coalition’s goal “is to build agencies that are difficult for its opponents to gain control over later,” which will ultimately mean that an agency is difficult for any outside forces to gain control over. This goal is advanced, for example, by writing detailed laws that impose inflexible restrictions on agencies’ mandate and decisional processes, and that minimize the number and power of political appointees, and by opposing formal provisions that would foster legislative oversight and participation in agency affairs, such as sunset provisions stipulating the necessity of reauthorization of the agency after some period.59 This strategy of insulating the agency from oversight and
38
CHAPTER 2
intervention by legislative majorities in the future assumes, of course, that they won’t simply pass a new law to achieve their goals. This strategy of insulation can be effective in an environment of institutional fragmentation because the institutional stickiness of the status quo will make it difficult and costly for future coalitions to pass new laws. Fragmentation of American political institutions gives rise to a multitude of what Ellen Immergut has called “veto points,” which are the source of the stickiness of the status quo in the American lawmaking system. Working from a comparative institutional perspective, Immergut explains: “Political decisions are not single decisions made at one point in time. Rather, they are composed of sequences of decisions made by different actors at different institutional locations. Simply put, enacting a law requires successive affirmative votes at all decision points.” The degree of difficulty of passing a contested proposal depends, in part, on the number of veto points along the decision chain.60 New institutional scholars in political science have emphasized that the American separation of powers system, giving a considerable degree of constitutional independence, autonomy, and legitimacy to separate executive, legislative, and judicial branches, coupled with the strong norm of judicial review, is characterized by multiple veto points.61 Greatly adding to these veto points, the lawmaking process within Congress is particularly fragmented. Bicameralism, an elaborate committee system that gives disproportionate powers to committee members and chairs, and the filibuster in the Senate, combine to create a multitude of players with the power to kill or radically reshape legislation that would easily command a solid majority if only it could reach a floor vote.62 These scholars have also argued that the relative weakness of parties in the United States encourages pivotal lawmakers to respond to important, even if very narrow and particularized, constituencies and interest groups when deciding whether and how to exercise their veto powers. The net result is a very sticky status quo. In this lawmaking system, as Moe puts it, “Whatever is formalized will tend to endure,” leading current majorities to embed their preferences into the structure of bureaucracy.63 This institutional logic for constraining bureaucratic discretion, grounded in uncertainty about the preferences of future coalitions, constitutes a potent incentive for legislators and their constituents to opt for private enforcement regimes, which they know will tend to endure.64 Indeed—and this is critical—once formalized, private enforcement regimes provide better insulation on the enforcement front than rule-governed agency powers, which future Congresses will have more continuing control over. Most significant among forms of continuing congressional control over bureaucracy, even if future Congresses lack the political capac-
INSTITUTIONAL FOUNDATIONS
39
ity or will to change or extinguish an agency’s enforcement authority by formal legal enactment, they can exercise some leverage over enforcement efforts through investigation, oversight hearings, earmarking funds, formal reporting requirements, refusing to confirm appointees, and, of course, by threatening to reduce or actually reducing the agency’s budget.65 In contrast, if an enacting Congress utilizes a private enforcement regime, there is little if anything that future Congresses can do to influence private enforcement levels short of repealing or amending the law. This will be far more difficult than efforts at agency oversight because of the institutional stickiness of the status quo already discussed, the high transaction costs of passing new legislation, and uncertainty about the ultimate legislative outcome if the legislative process is opened up.66 Just as private enforcers will be largely beyond the reach of the president, they will also be beyond the reach of future legislative majorities, short of a new legal enactment. To the extent that this structural cause for enactment of private enforcement regimes is in fact operative, it will be intensified under conditions in which legislators are more apprehensive about losing power. That is, while private enforcement regimes will always have some appeal in a separation of powers system because of their ability make policy choices stick even when future legislative majorities with different preferences come to power, greater concern about electoral losses will make current majorities more likely to use them. Thus, as the risk of electoral losses increases for the majority party, it will be more likely to enact private enforcement regimes. This hypothesis will be tested in chapter 3. These effects of electoral risk, in conjunction with a sticky status quo, have a logic similar to that of legislators using judicial appointments for “political entrenchment,” and using an independent judiciary for purposes of making “credible commitments” that statutory policy will survive the enacting coalition. Private enforcement regimes can certainly allow a ruling coalition to entrench not just its statutory policy pronouncements, but also an enforcement apparatus designed to achieve desired enforcement levels. Private enforcement regimes likewise can be thought to enhance credible commitments by legislators to interest groups. Landes and Posner’s formulation of this idea focuses upon commitment to the substantive meaning of a statute, and hinges upon judges observing norms of legality according to which the enacting coalition’s intent guides judicial interpretation, a supposition about judicial behavior that many scholars have doubted.67 Private enforcement regimes represent a credible commitment to consistent levels of enforcement, as distinct from substantive interpretations of the law, where the credibility of the commitment hinges only upon potential plaintiffs and lawyers responding to opportunities
40
CHAPTER 2
and market incentives for enforcement, not expectations about norms of legality governing judicial behavior. We will return shortly to the issue of how judicial interpretation can influence the efficacy of private enforcement regimes. Private Enforcement Regimes and Bureaucratic Drift When delegating to the bureaucracy, legislators also have “bureaucratic drift” to worry about. The preceding discussion of legislative-executive conflict addressed the prospect of bureaucrats being used and influenced by the White House to pursue its policy goals at the expense of the enacting congressional coalition’s, and the discussion of coalition drift addressed the prospect of bureaucrats being used and influenced by future congressional coalitions to pursue their policy goals at the expense of the enacting congressional coalition’s. A third risk concerns the problem that bureaucrats are agents with their own preferences, at least partially distinct from any of their principals in the legislative or executive branches. McCubbins, Noll, and Weingast (McNollgast) observe that “bureaucrats have personal preferences . . . derived from some combination of private political values, personal career objectives, and, all else equal, an aversion to effort, especially effort that does not serve personal interests.” Consequences that arise from bureaucrats’ personal preferences include the risks of bureaucratic shirking of delegated work, capture of the bureaucracy (undue influence upon it) by the regulated population, and bureaucratic oligarchy, where bureaucrats pursue their own policy preferences rather than those of democratically elected officials.68 An important caveat is necessary about the parameters of the concept of bureaucratic drift deployed in this book. Viewed from Congress’s point of view, McNollgast’s account of bureaucratic drift overlaps with Moe’s account of legislative-executive conflict as a motivation for legislators’ choices about the structure of bureaucracy. McNollgast’s account contemplates pivotal legislative actors and the president as principals in an enacting coalition that is concerned about drift by the bureaucratic agents that it empowers. Because there are multiple principals in the coalition, one source of drift that each is worried about is the prospect that in the future other principals will seek to influence the bureaucracy to move policy in their own preferred direction.69 If we conceptualize Congress as a unitary actor, from its point of view the president is another principal that threatens to pull policy (cause it to drift) in her direction and away from the legislative bargain. This dynamic’s influence on the congressional choice of private enforcement regimes is already accounted for in the conceptual model developed in this chapter under the rubric of
INSTITUTIONAL FOUNDATIONS
41
legislative-executive conflict. Thus, the concept of bureaucratic drift used here is defined so as not to include drift caused by the president, which is treated as a separate phenomenon. McNollgast further observes that the stickiness of the status quo in the American lawmaking process, arising from the institutional fragmentation and multiple veto points discussed in the last section, can create significant latitude for bureaucrats to move policy (cause it to drift) in the direction of their personal preferences, and away from the preferences of the elected branches, before all necessary players in the lawmaking process will have an incentive to coordinate their actions in a legislative reversal.70 Recognizing these institutional limitations on corrections, through new legislation, to agencies’ deviation from the enacting coalition’s intent, that coalition will often strategically design into a law structures and procedures aimed at preventing bureaucratic drift. A key aspect of this legislative practice concerns allowing groups favored by the enacting coalition to intervene effectively in agency implementation, which is accomplished by requiring agencies to publicize important policy actions in advance, to keep policy proceedings open, to permit favored groups to provide information and otherwise participate in policymaking proceedings, and to use administrative procedures that militate in favor of outcomes preferred by favored groups.71 Note that there is generally a tradeoff between controlling coalition drift on the one hand, and controlling bureaucratic drift on the other. As Horn and Shepsle put it: “To deal with the bureaucratic drift, the arrangement must provide future coalitions with a relatively inexpensive way to intervene in implementation. To deal with the legislative drift, one must place obstacles in the path of such interventions.”72 Because each goal will typically be pursued at the expense of the other, both will necessarily remain far from fully realized. The problem of bureaucratic drift can create further institutional incentives for Congress, and the interest groups influencing it, to rely upon private enforcement regimes. Other things being equal, as an enacting coalition grows more concerned, with respect to its preferred level of enforcement of a regulatory statute, about bureaucratic drift (either in the direction of more or less enforcement than it wants), the appeal of private enforcement regimes, designed to produce desired enforcement levels via market incentives, will grow. The tradeoff relationship between controlling bureaucratic drift and coalition drift, described by Horn and Shepsle, further highlights the appeal of private enforcement regimes. On the regulatory enforcement front, private enforcement regimes present a potential solution to the tradeoff conundrum: they can provide for desired levels of enforcement in a manner that simultaneously insu-
42
CHAPTER 2
lates from both bureaucratic drift and coalition drift. To be sure, private enforcement regimes can involve their own costs. Later sections of this chapter consider some of them, after a number additional concepts are introduced that will aid in that assessment. Unlike the legislative-executive conflict and the coalition drift hypotheses, the bureaucratic drift hypothesis does not yield straightforwardly observable implications that can be directly tested in a statistical model. That is, while the hypothesized effects of legislative-executive conflict should be greater as the ideological distance between Congress and the president widens, and the hypothesized effects of coalition drift should be greater as electoral uncertainty increases, there is no straightforward direct empirical measure of the threat of bureaucratic drift.73 However, there is an indirect approach to evaluating the hypothesis. As discussed in the next chapter, scholars have theorized that issue-oriented interest groups are an important source of demand for private enforcement regimes, and that their demand is motivated by fear of bureaucratic drift. This claim can be tested by assessing the influence of issue group demand on legislative enactment of private enforcement regimes in the empirical model in the next chapter, and then investigating issue group motivations for preferring private enforcement regimes (if they do) in the historical evidence examined in chapters 4 to 6. If the evidence shows that issue groups are an important and effective source of demand for private enforcement regimes, and that they are motivated by fear of bureaucratic drift, this evidence will militate in favor of the bureaucratic drift hypothesis. Further, to the extent that the legislative-executive conflict and the coalition drift hypotheses are supported in the statistical model, revealing strategic enactment of private enforcement regimes to guard against subversion by executives and future legislative majorities, the closely related theoretical claim that an enacting coalition would also use them to guard against bureaucratic drift will be strengthened. Conversely, if the evidence does not suggest strategic use of private enforcement regimes in the other contexts, then the plausibility of the bureaucratic drift hypothesis will surely be weakened. Private Enforcement Regimes and Compromise The multitude of veto points in America’s fragmented lawmaking institutions encourage enactment of private enforcement regimes for an additional reason. A core theme of work focusing on fragmented institutions in the American lawmaking process, particularly in the subfield of American political development, is the way in which the many veto points truncate, limit, and curtail ambitious policy initiatives, if they can survive that lawmaking process at all.74 As Sven Steinmo puts it, an American law-
INSTITUTIONAL FOUNDATIONS
43
making system “replete with veto points . . . gives huge power to interests wishing to stop, alter, or modify governmental action,” and consequently by the time ambitious policy initiatives “wheedle their way through the labyrinth and past so many veto points,” they typically will have been considerably whittled down to satisfy many gatekeepers.75 This institutional environment has consequences for legislative choices concerning the deployment of public versus private power and resources in pursuit of policy goals. Christopher Howard and Jacob Hacker, in their respective studies on tax expenditures and policies encouraging private employment-based benefit plans, suggest that as compared to direct state expenditures and bureaucratic welfare state-building, laws privatizing the provision of social benefits are more likely to attract the support of a sufficiently broad coalition—including conservative Democrats and moderate Republicans, for example—to accomplish the supermajoritarian task of clearing the labyrinth of veto points in the American lawmaking process. Legislators occupying veto points who are unwilling to support direct bureaucratic welfare state-building may be willing to support the pursuit of similar policy goals, according to Howard and Hacker, by incentivizing private behavior. This is partly because it will be less visible and more ambiguous as a form of state intervention, making it more acceptable to those with antistatist preferences, and may strategically operate to diminish the likelihood of bureaucratic welfare state-building in the future. Even when advocates of expanded direct welfare statebuilding command a simple majority in Congress, they may be driven to embrace policies that operate through private incentives, for the alternative may be no legislation at all.76 The legislative creation of incentives for the private provision of benefits to achieve welfare state goals, and the legislative creation of incentives for private litigation to achieve regulatory state goals, are analogous in a number of important respects. Private litigation, with its air of private dispute resolution, is less visible and more ambiguous as a form of state intervention. Therefore, it may be preferred to bureaucratic statebuilding by legislators with antistatist preferences, a significant strand of the American political tradition, particularly as applied to the central state in the United States’ federalist system. Indeed, private enforcement regimes may be embraced by those legislators as a way of thwarting bureaucratic state-building. Just as the introduction to this book argued that state-centered scholars have neglected the significance of private enforcement regimes as a form of state intervention, legislators and the public likewise tend to regard them, as Kagan puts it, as “nonstatist mechanisms” of policy implementation. As compared to constructing and financing bureaucratic regulatory enforcement machinery and endowing it with coercive powers, for example, to investigate, prosecute,
44
CHAPTER 2
adjudicate, and issue cease-and-desist orders, an enforcement regime that is founded instead upon allowing aggrieved persons to prosecute their own complaints in court is likely to attract broader support, increasing its chances of emerging from the American legislative gauntlet. If there are pivotal lawmakers prepared to obstruct enactment of regulatory policy that entails bureaucratic state-building, utilizing private enforcement regimes may facilitate overcoming such obstructions.77 It is critical to highlight here that while fragmented institutions, by necessitating compromise, can drive enactment of private enforcement regimes as an alternative to bureaucratic regulatory state-building, this does not necessarily parallel the “weak state” and “whittling down” themes characteristic of the American political development literature. The introductory chapter suggested that private litigation may be regarded as a powerful form of state intervention. Whether to regard it as a form of intervention more or less powerful than bureaucratic regulatory state-building cannot be answered in the abstract. A robust administrative enforcement framework can certainly be more powerful than a feeble private enforcement regime, for example, in which most suits have a negative expected value. At the same time, however, a robust private enforcement regime can certainly be more powerful than a feeble bureaucratic enforcement framework characterized, for example, by modest formal administrative powers and inadequate resources. Thus, the claim advanced here is that veto points, by necessitating compromise, can encourage enactment of private enforcement regimes, distinct from the question of the strength or weakness of the enforcement framework to be constructed. This claim about the relationship between the multitude of veto points in American lawmaking institutions does not yield a straightforward and readily operationalizable hypothesis that can be tested in a statistical model. It is, however, an account that, if a significant explanatory factor, should show itself in the historical evidence surrounding legislative decision-making about regulatory enforcement. We will assess such evidence in the context of federal job civil rights laws in chapters 4 to 6. Private Enforcement Regimes and the Allocation of Power These institutional dynamics governing Congress’s enactment of private enforcement regimes bear directly upon Congress’s allocation of policymaking power among potential implementation agents. While discussion thus far has focused on the enforcement of policy, it is necessary here to distinguish between rule enforcement and rule articulation in the policy implementation process. Given a discrete set of statutory rules, rule enforcement concerns what resources are mobilized to monitor for and
INSTITUTIONAL FOUNDATIONS
45
investigate violations, and to prosecute and punish violators. However, for reasons discussed below, Congress generally does not, and cannot, provide a comprehensive set of legal rules to be enforced. As a result, the implementing authority, whether judicial or administrative, will be required to articulate rules elaborating the substantive meaning of a statute in the course of applying it. Contrasting which actors exercise these distinct rule enforcement and rule articulation functions under a bureaucratic implementation regime, as compared with a private enforcement regime, helps to crystallize how the two approaches to implementation differentially distribute power among the branches of government, and between the state and civil society. Bureaucracies can articulate and enforce rules. The power of agencies in the regulatory process is multifaceted and broad ranging, and fuses the powers associated with each of the three branches of government. Agencies can have rule-making powers that are fundamentally legislative in character, such as when the Environmental Protection Agency makes rules governing emissions from motor vehicles. They can have adjudicatory powers that are essentially judicial, such as when the National Labor Relations Board adjudicates disputes over alleged unfair labor practices, followed by issuance of judicial-looking opinions and binding orders. These quasi-legislative and quasi-judicial agency powers entail rule articulation in the course of elaborating the meaning of a statute in application. And, of course, agencies can be given the quintessentially executive function of rule enforcement by being authorized to enforce their own statutes and rules, such as by monitoring to detect violations, conducting compulsory inspections, demanding documents and testimony, and prosecuting enforcement proceedings. While a particular agency’s powers will be defined by its enabling statute, typically an agency will have both rule enforcement and rule articulation powers.78 As compared to bureaucratic implementation, private enforcement regimes empower private litigants and their attorneys to enforce rules at the expense of the executive. Note that this points to a falsehood (certainly not the only one) in the textbook characterization of the allocation of powers and functions in the American separation of powers system. On that view, Congress has the power to make laws, the president has the power to implement and enforce them (to “take care that the laws be faithfully executed,” as the Constitution puts it), and the federal courts have the power to decide cases and controversies presented to them. However, when Congress implements laws through the use of private enforcement regimes, it can cut the president out of this scenario, save for an indirect presidential influence via judicial nominations. When a law is implemented purely through a private enforcement regime, Congress makes laws, private actors and their attorneys “take care that the laws
46
CHAPTER 2
be faithfully executed,” and courts decide cases or controversies. Private actors are given what are classically thought of as executive responsibilities, and the president, along with the executive apparatus, is sidelined. This has significant policy consequences, and not just for how much or little enforcement there will be. It also has consequences for the distribution between state and civil society of the power to affect policy. The act and process of litigation can have far-reaching policy effects beyond the individual case being prosecuted.79 Most obviously, it sets the agenda for courts’ exercise of their rule articulation function (discussed below), since courts can only articulate rules in the context of cases presented to them. It can set legislative agendas by triggering legislative interventions or overrides.80 It can drive policy innovation by the regulated population, whose creativity is stimulated by a desire to avoid liability.81 It can draw national media coverage and bring mass attention to an issue, fostering public deliberation.82 And such public attention to an issue can itself stimulate more litigation in similar cases.83 More broadly, it can shape the ambition and efficacy of social movements.84 Choices about what, when, where, and how to enforce the law can have widely radiating policy effects, and private enforcement regimes, as compared to bureaucratic implementation, place those choices in private rather than public (executive) hands. As compared to bureaucratic implementation, private enforcement regimes empower courts to articulate rules at the expense of the executive. This is because the delegation of power to apply and interpret statutes is the delegation of power to give content and contours to statutory rights and prohibitions from a range of plausible alternatives. When courts interpret statutes, “[T]hey become an integral component of the legislative process.”85 When courts interpret statutes, they make policy.86 This is, of course, also true of agency interpretation of statutes,87 for the deep policymaking dimension of statutory interpretation inheres in the enterprise of interpreting and applying statutes, not in the title of the interpreters or the governmental branch in which they work. Recognizing this fact is critical to appreciating the power stakes implicated when Congress designates the agent that will engage in rule articulation in the course of statutory implementation. This view of statutory interpretation and application as policymaking does not depend upon the claim that judges or bureaucrats are overreaching usurpers of legislative power. Even an interpreter that struggles for fidelity to congressional intent will often be required to make policy. One important reason this is so is that Congress frequently legislates through broad and vague pronouncements that simply do not make clear what rights and prohibitions it intended to enact.88 In attempting to apply such broad and vague statutory language, courts and bureaucrats given implementation responsibility have no choice but to make policy.
INSTITUTIONAL FOUNDATIONS
47
Scholars have identified a number of factors that contribute to Congress’s propensity to enact broad and vague laws generating wide discretion to articulate substantive rules at the implementation stage. Legislators may wish to draw on the expertise of implementing agents, and allow them flexibility to adapt the regulatory regime to changing circumstances; legislators may wish to escape electoral blame for hard policy choices, which they can slough off on courts and bureaucrats with vague regulatory pronouncements; they may wish to avoid expending scarce time and energy that would be required to write a detailed law; they may be trying to appear responsive to contradictory constituent demands; and they may simply be unable to pass a law that is more clear and detailed in its policy prescriptions, and thus more likely to attract opposition in the legislative process.89 Moreover, even when legislators are able to overcome these obstacles to statutory specificity, they are simply not capable of predicting the multitudinous circumstances that will arise in the course of the statute’s application, and thus they are not capable of resolving them in the law’s text.90 The important point for present purposes is that many factors endemic to politics, American political institutions, and bounded rationality point to the inevitability of policymaking through rule articulation at the implementation stage. This view is consistent with extensive systematic empirical evidence demonstrating that judges’ partisan and ideological preferences influence their voting patterns in statutory interpretation cases across wide-ranging policy domains. Studies have found, for example, that more liberal and Democratic judges vote more liberally, and more conservative and Republican judges vote more conservatively, when interpreting and implementing federal statutes across the domains of civil rights policy,91 labor policy,92 environmental policy,93 tax policy,94 and immigration policy.95 The fact that judges actually decide statutory interpretation and implementation cases differently from one another according to their ideological preferences across the waterfront of federal policy domains provides ample confirmation that the delegation of power to interpret and apply statutes is the delegation of the power to make policy. In sum, whereas bureaucratic implementation typically fuses rule articulation and rule enforcement powers in an agency, private enforcement regimes bifurcate the two, with private plaintiffs and their attorneys exercising rule enforcement powers and courts exercising rule articulation powers. To the extent that America’s fragmented political institutions create incentives for Congress to enact private enforcement regimes, they create incentives for Congress to empower courts to articulate rules and private litigants and lawyers to enforce them. To the extent that ideological polarization between Congress and the president starting in the late 1960s led Congress to rely more heavily upon private enforcement
48
CHAPTER 2
regimes and less upon administrative machinery to implement its policy choices, these conditions shifted policymaking power away from the executive branch and into the hands of private litigants, their attorneys, and judges. If private enforcement regimes, relative to an administrative delegation, redirect power from the executive to private litigants, lawyers, and courts, one might be tempted to ask why a president would facilitate her own disempowerment by signing such laws. There are many possible reasons that presidents sign laws with private enforcement regimes, and three are identified here. First, presidents don’t get everything they want. While a president who dislikes some aspect of a legislative proposal can most certainly use the veto threat or an actual veto to attempt to induce legislative revisions, in the executive-legislative dance that Charles Cameron calls “veto bargaining,” the president’s preferences will typically only be partially fulfilled, with multiple other factors, most importantly legislative preferences, also shaping the law’s content.96 The notion is not new that executives will sign laws that give them less power than a differently drafted law could give them. As already discussed, the empirical literature on legislative delegation to the bureaucracy has provided powerful evidence that under conditions of divided government, executives sign laws that contain greater limits and constraints on executive discretion than under conditions of unified government.97 A second reason is that it is not, in fact, at all clear that presidents facilitate their own disempowerment by signing regulatory laws with private enforcement regimes. It may be that legislative resistance to bureaucratic state-building is such that the choice is between a regulatory law with a private enforcement regime versus no law at all. In chapter 4 we will see that this may well have been the case with respect to Title VII of the CRA of 1964. Some pivotal conservative legislators credibly threatened to kill the legislative proposal if it contained a meaningful administrative enforcement apparatus, and in exchange for their support they insisted upon the private enforcement alternative. While President Johnson supported stronger executive enforcement powers, clearly he preferred an enacted Title VII with a private enforcement regime to no Title VII at all. A third and related reason is that a president may believe that a private enforcement regime will produce levels of enforcement more to her liking than an administrative enforcement apparatus. For example, a president may prefer a strong private enforcement regime if she prefers a high level of enforcement and is concerned that Congress would not adequately fund bureaucratic enforcement over time; or that problems of bureaucratic drift would lead to weak enforcement; or that there would be undesirable political fallout from vigorous executive enforcement. That is,
INSTITUTIONAL FOUNDATIONS
49
in some circumstances a president’s pursuit of policy goals, served by a private enforcement regime, may trump her preference to enlarge her own power. Thus, there are numerous plausible theoretical explanations for the observed fact that presidents have often signed regulatory laws utilizing private enforcement regimes rather than administrative implementation, just as they have signed laws explicitly constraining their exercise of administrative power. Private Enforcement Regimes and Judicial Ideology This chapter’s discussion of the legislative choice to enact private enforcement regimes points to the importance of accounting for the ideological position of the judiciary. In this chapter it has been argued that (1) Congress can regard bureaucratic implementation and private enforcement regimes as alternative implementation strategies; (2) Congress’s willingness to delegate power to the executive to implement statutes is diminished as the executive’s ideological distance from Congress increases; (3) enactment of private enforcement regimes effectively delegates policymaking power to courts; and (4) courts’ exercise of that power is influenced by judges’ ideology. Under these circumstances it seems unavoidable that Congress’s choices regarding enactment of private enforcement regimes will be influenced by the ideological position of the federal judiciary. Before discussing how this might play out, brief mention is warranted of possible institutional differences between courts and agencies as rule articulators. Courts versus Agencies as Rule Articulators
Earlier discussions of the threats of subversion of administrative implementation emanating from legislative-executive conflict, coalition drift, and bureaucratic drift, each contrasted private lawsuits with administrators in the domain of rule enforcement (the central focus of this book), emphasizing the relative insulation of private lawyers and litigants. Considering the role of judicial ideology raises the prospect of what might be called judicial drift (judges pursuing their own preferences, through rule articulation, at the expense of the enacting coalition’s). The same institutional fragmentation that makes it difficult to pass new legislation to correct bureaucratic drift and coalition drift likewise makes it difficult to pass new legislation to correct judicial drift. From the standpoint of the delegating Congress, this suggests the possible relevance of differences between judges and administrators regarding the threat of subversion, and raises the question of whether, aside from the simple ideological position of the two relative to Congress, there are institutional differences
50
CHAPTER 2
between judicial and administrative rule articulation that are relevant to the comparative risk of drift. This is a question different from the relative competence of judges versus administrators, a subject of much scholarly disagreement. Here the focus is upon the relative fidelity to Congress of judges versus administrators. This is a difficult question on which the scholarly literature lacks consensus. Some, such as Landes and Posner, have argued that institutional norms of legal interpretation within the judiciary will lead judges to be more faithful agents of enacting coalitions in the course of rule articulation, insulating the substance of legislative bargains from the vicissitudes of electoral politics.98 Others, such as Segal and Spaeth, propound an account of judicial rule articulation suggesting that judges will be no less prone to pursue their own ideological agenda than administrators.99 Further, some have argued that the organizational culture of single-mission agencies will lead bureaucrats within them to assume a more aggressive regulatory posture toward rule articulation than judges, who will be more inclined to see themselves as neutral adjudicators of narrow legal disputes.100 Others have argued that agencies regulating in a specific area, and having ongoing contact with the regulated population, will be more prone to be captured by that population, while the judiciary’s decentralized structure and independence from the regulated population will facilitate more vigorous regulatory implementation in the course of rule articulation.101 This discussion could go on identifying institutional characteristics of courts and agencies pointing in contradictory directions. Existing theory and evidence simply do not provide a clear answer to the question of whether, in general, courts or agencies will be more prone to drift away from the preferences of the enacting coalition when exercising rule articulation powers. One thing can be said with relative confidence, however: judicial drift will be harder for Congress to control than bureaucratic drift. I argued above, in connection with the threat to administrative implementation posed by coalition drift, that private enforcement regimes produce a rule enforcement mechanism, carried out by lawyers and litigants, that is more insulated from future Congresses than agency enforcement powers, because Congress has many more tools for ongoing supervision of agencies. It is likewise the case that private enforcement regimes produce rule articulation powers, exercised by lifetime tenured judges in an independent judiciary, that are more insulated from future Congresses than agency rule articulation powers, because Congress has many more tools for ongoing supervision of agencies.102 While this much seems fairly clear, the inference that one can draw from it about incentives faced by Congress is not. For example, if Congress believes that institutional norms of legal
INSTITUTIONAL FOUNDATIONS
51
interpretation, and institutional independence from the regulated population, will make the judiciary substantially more faithful to the enacting coalition than an administrator would be, then it might regard tools for continuing oversight as substantially less important. In sum, it is difficult to reach conclusions about generalizable net effects of institutional differences between courts and agencies as they bear upon the risk of policy drift in the sphere of rule articulation. There are, though, certainly reasons to believe that the judiciary’s ideological distance from Congress will influence Congress’s choice of private enforcement regimes, other things being equal. The Judicial Friendliness Hypothesis
Given the linkage already noted between judicial ideology and judicial interpretation of statutes, at first blush intuition suggests that Congress would be more likely to enact private enforcement regimes the more ideologically proximate courts are, and less likely to enact them the more ideologically distant courts are. This seems most clear in the rule articulation dimension of implementation. If the judiciary’s ideological position is such that it will elaborate the substantive meaning of statutes’ regulatory commands in a manner favored by Congress, this naturally should militate in favor of congressional enactment of private enforcement regimes, holding other variables constant. Congressional concerns about the rule enforcement dimension of private enforcement regimes can also contribute to this expected relationship. This is because the way in which courts exercise their rule articulation function in the course of statutory interpretation is entwined with execution of the rule enforcement function by private litigants; that is, the way that courts elaborate doctrine giving substantive meaning to statutory text can have important effects on how much or how little private litigation is mobilized to enforce it. When a statute is ambiguous regarding whether Congress intended to allow private enforcement, courts will have some interpretive latitude in deciding whether to “imply” a private right of action. Judicial interpretations of statutes’ substantive principles of liability can either favor or disfavor plaintiffs’ probability of prevailing (p), influencing expected value (EV). Judicial interpretations of statutory provisions governing damages and the allocation of litigation expenses, too, can either help or hurt plaintiffs, thereby influencing expected value (EV) via expected benefits (EB) and expected costs (EC). In these ways and others, courts can mobilize or demobilize private litigants.103 The potential effects of judicial ideology on the level of private statutory enforcement, like its effects on substantive policy, suggest that as Congress moves closer to the judiciary, it is reasonable to anticipate the expected value of
52
CHAPTER 2
cases (EV) to move in a direction more attractive to Congress, making the judiciary a friendlier enforcement venue from Congress’s point of view, increasing its propensity to enact private enforcement regimes. The Judicial Hostility Hypothesis
However, contrary to initial intuition, when we focus on rule enforcement, there are also theoretical reasons to expect countervailing forces to incline Congress toward increasing incentives for private litigation as courts move ideologically further away from Congress. This logic is implicit in the discussion above suggesting that, as judicial ideology becomes more distant from Congress, courts will be more likely to engage in judicial interpretation that moves EV in a direction objectionable to Congress. In a scenario in which Congress is seeking to mobilize private enforcers, for example, but distant courts would hold plaintiffs’ probability of prevailing (p) below that preferred by Congress, Congress can counteract this potential outcome by increasing expected benefits (EB) or reducing expected costs (EC) in the equation EV = EB(p) − EC. A concrete example helps to illustrate the point. Consider the hypotheticals in table 2.4 as a set of sequential moves in a scenario in which there is a Republican president, a Democratic Congress, and a suit by a plaintiff under a consumer protection statute. Starting with an initial baseline state of hypothetical 1, with actual damages only and the American rule on fees, the plaintiff’s expected benefit if she wins the case is $40,000, her estimated probability of prevailing is .6, and her expected costs are $20,000, netting an expected value for the case of $4,000. The plaintiff will sue. Next, for hypothetical 2, assume that after a series of Republican judicial appointments the federal bench becomes more probusiness, and the plaintiff’s probability of prevailing is reduced to .4, with the effect of driving down the expected value of her case to negative $4,000. Now the plaintiff will not sue. The Democratic Congress can respond to this more hostile (from its point of view) judiciary by increasing expected benefits (EB) or reducing expected costs (EC). If it amended the consumer protection statute to add a double damages provision, as in hypothetical 3a, this would raise the plaintiff’s expected benefits from $40,000 to $80,000, which would increase the expected value of the plaintiff’s claim to $12,000, thus ratcheting up enforcement pressure to a level higher than in hypothetical 1. If Congress, instead, amended the consumer protection statute to add a plaintiffs’ fee shift, as in hypothetical 3b, this would reduce the plaintiff’s expected costs from $20,000 to $12,000, which would increase the expected value of the plaintiff’s claim to $4,000, thus exactly offsetting the increased conservatism of the court and returning enforcement pressure
53
INSTITUTIONAL FOUNDATIONS
Table 2.4 Expected Value in a Separation of Powers Context
Hypothetical
Expectected benefits
Probability of success
Expected costs
Expected value
(1)
(Actual damages) $40,000
(More liberal court) .6
(American rule) $20,000
$4,000
(2)
(Actual damages) $40,000
(More conservative court) .4
(American rule) $20,000
−$4,000
(3.a)
(Congress doubles EB) $80,000
(More conservative court) .4
(American rule) $20,000
$12,000
(3.b)
(Actual damages) $40,000
(More conservative court) .4
(Congress enacts plaintiffs’ shift) $12,000
$4,000
(as measured by expected value) to the same level as in the initial baseline state in hypothetical 1. In chapter 6’s analysis of the genesis of the CRA of 1991, where Congress responded to the judiciary’s growing conservatism in Title VII cases partly by increasing available damages, we will observe an actual scenario resembling the move from hypothetical 1, to 2, to 3a. While a sequential example has been used for purposes of illustration, at the time that a regulatory statute is originally drafted, it can be shaped by legislators’ anticipation of how courts will interpret it. Thus it is evident that in some circumstances Congress may rationally respond to an increasingly ideologically distant court by enacting ever more robust private enforcement regimes. Because there are countervailing forces that cause increases in the judiciary’s distance from Congress to create incentives to enact private enforcement regimes, and not to enact them, theory does not support firm expectations about which causal force will predominate, if either does. These contradictory predictions about the effects of judicial ideology on Congress’s propensity to enact private enforcement regimes are not merely a peculiar artifact of theory. When examining Congress’s legislative choices about enforcement of civil rights laws in chapters 4 to 6, we will observe qualitative evidence in support of both the judicial friendliness and the judicial hostility hypotheses at different points in time. Whatever the effect of the judiciary’s ideological position, there is reason to expect that the judiciary’s ideological position will be weighed sig-
54
CHAPTER 2
nificantly less by Congress than that of the executive, and the logic here again concerns the distinction between rule articulation and rule enforcement. As discussed above, bureaucratic implementation typically gives agencies powers of both rule articulation and rule enforcement, whereas private enforcement regimes divide the two powers between courts (rule articulation) and private plaintiffs and their attorneys (rule enforcement), who will execute enforcement functions guided, and insulated from subversion, by market incentives. Assuming, for purposes of illustration, that Congress is substantially and equally distant from the executive and the courts, and is weighing only its ideological distance from these potential implementation venues, it would prefer a private enforcement regime over a bureaucratic delegation. Whereas in the bureaucratic case the president and his officers could subvert congressional preferences with respect to both rule enforcement and rule articulation, in the case of a private enforcement regime the rule enforcement function is largely selfexecuting and insulated. This logic suggests that while Congress may be influenced by the judiciary’s ideological distance from it, the magnitude of this effect will be substantially weaker than the president’s ideological distance. A Note on Private Enforcement Regimes and Federalism A number of scholars have suggested that the federal structure of the American state is a significant factor explaining the large role played by private litigation in policy implementation, frequently identifying constitutional litigation by individuals or interest groups against states as evidence.104 When individuals with grievances against states cannot enlist the assistance of the federal legislative or executive branches, they can assert constitutional rights in federal courts. Within this book’s narrow focus upon the statutory choice by Congress of private enforcement regimes versus bureaucracy, the causal significance of federalism is not theoretically clear. To the extent that the claim is that the American federalist tradition fosters normative resistance, in legislators’ preferences, against growing the central state,105 that was incorporated into the discussion above of private enforcement regimes and compromise in a veto-pointridden lawmaking environment. Some have also suggested, though, that within the domain of statute, litigation is somehow more effective than bureaucracy as a tool to regulate state and local governments.106 If there is federal legislative authority to regulate states on some issue, however, it is not clear why regulation of states as entities, as contrasted, for example, with federal regulation of corporations, or federal regulation of sectors of the federal government itself, would render the choice of legal process over administrative process more likely to be effective.
INSTITUTIONAL FOUNDATIONS
55
When empirically examining congressional enactment of private enforcement regimes in the next chapter, we will see that states have not been a particularly large part of the story. A measure is introduced of Congress’s proactive efforts to mobilize private litigants by statutorily providing economic rewards for successful plaintiffs, and data is presented on such statutory litigation incentives from 1887 to 2004. Only 9 percent of the incentives provide for actions against states, while 17 percent do so against the federal government, and 84 percent apply against private sector entities. These categories, of course, are nonexclusive. Only 3 percent of the litigation incentives apply exclusively to states, and 12 percent apply exclusively to the federal government. While private enforcement regimes have been used against states in the service of very important policy goals, Congress actually has been about twice as likely to mobilize private litigants against the central government itself than against states, and about four times as likely to target central government exclusively. Its primary target, by far, has been the private sector. Are Private Enforcement Regimes a Free Lunch? Some might be tempted to ask why a Congress enacting a regulatory prohibition would not always include a private enforcement regime given that it privatizes a huge majority of the costs of enforcement, insulates enforcement powers from subversion by the president, future legislative majorities, and disobedient bureaucrats, and will appeal to a broader coalition than bureaucratic regulatory state-building. However, private enforcement regimes are not really a free lunch. The ideological distribution of preferences among the branches may be such that Congress prefers that the executive branch, rather than courts, be responsible for rule articulation, which is a profoundly important policymaking function. Private enforcement regimes clearly consume judicial resources, and can clog court dockets. Moreover, a flip side of the insulation phenomenon is that private litigants, lawyers, and lifetime tenured judges are less susceptible to ongoing supervision even by the enacting Congress than are bureaucrats, who can be called into hearings and have their budgets slashed, and thus the use of private enforcement regimes may entail a greater loss of control over policy by the enacting Congress. If a private enforcement regime produces either significantly more or less rule enforcement (litigation) than Congress intended, or a judge hijacks the meaning of a statute through rule articulation, this state of affairs will become a part of the sticky status quo, which Congress will only be able to rectify if it can overcome barriers to new legislation.
56
CHAPTER 2
Finally, private enforcement regimes, as policy instruments, have many detractors. Some regulation scholars have suggested that, as compared to administrative regulation, private enforcement regimes produce policy inconsistency and uncertainty because policy emanates from a multitude of litigants and judges; bias policy toward private and away from public interests; fail to mobilize adequate policy expertise; are needlessly adversarial, subverting cooperation and voluntary compliance; are extremely costly to the regulated population; are painfully slow and cumbersome; are antidemocratic because they allow minorities to extort policy concessions and monetary side-payments that they could not secure through ordinary legislative politics; and may undermine the efficacy administrative power in a hybrid regime.107 While supporters of private enforcement regimes contest this view,108 suffice it to say that critics of private litigation as a regulatory tool have suggested ample reasons that Congress does not simply include a private enforcement regime in every regulatory law it enacts. Comparative Institutional Implications of Private Enforcement Regimes This chapter’s discussion of the relationship between separation of powers structures and legislative decisions about regulatory implementation suggest an important comparative institutional point. Moe’s arguments outlined above were partly marshaled in response to a comparative institutional puzzle raised in the work of James Q. Wilson.109 In decrying the extent to which the strength and effectiveness of American government was undermined by procedural and structural encumbrances on bureaucratic discretion, Wilson observed that, by comparison, bureaucracies in parliamentary regimes were less burdened by formal rules, procedures, and constraints, and thus were able to pursue their policy missions more forcefully and effectively.110 In response, Moe stressed that it is precisely the institutional structure of the separation of powers system that motivates Congress to cabin bureaucratic discretion with formal rules and structural constraints. Threats of subversion from legislative-executive conflict, coalition drift, and bureaucratic drift all have this effect. In parliamentary systems, by comparison, Moe writes: [T]he executive arises out of the legislature and both are controlled by the majority party. Thus, unlike in the United States, the executive and the legislature do not take distinctive approaches to issues of structure; they do not struggle with one another in the design and control of public agencies; they do not push for structures that protect against or compensate for the other’s political influence.111 Likewise, in parliamentary regimes coalition drift is not a significant incentive to formalize into law rules and procedures meant to insulate
INSTITUTIONAL FOUNDATIONS
57
bureaucratic power from manipulation by future coalitions. Though certainly stylized, the simple two-party case is illustrative. Moe continues: [W]hichever party gains a majority of seats in parliament gets to form a government and, through cohesive voting on policy issues, is in a position to pass its own program at will. Similarly, should the other party gain majority status down the road, that party would be able to pass its own program at will—and, if it wants, to subvert or completely destroy everything the first party has put in place. . . . This means that formal structure does not work as a protective strategy— at least not in the simple, direct way that it works in a separationof-powers system.112 Moe’s logic suggests that the prospect of bureaucratic drift also does not pose the same threat in the parliamentary as contrasted with the separation of powers context. Because legislative coalitions in the less institutionally fragmented parliamentary setting are far more capable of acting decisively against errant bureaucrats, they are less in need of formal structure as an ex ante protective strategy. Moe’s comparative institutional argument has provocative implications for the present discussion. He provides an institutional explanation for the development of larger, more coherent, unified, and centralized bureaucratic state machinery in parliamentary regimes, and the failure of a “strong state” on this European model to develop in the United States. American political development scholars, as discussed above, have made similar arguments. America’s fragmented, veto-point-ridden political institutions cause a whittling down of ambitious state-building projects, according to Steinmo, rendering a “weak state” relative to the usual parliamentary European comparators, measured against the benchmark of Weberian administrative centralization. Scholars have long recognized that private litigation plays an unusually pervasive and important role in the implementation of public policy in the United States as compared to industrial democratic countries with predominantly parliamentary systems.113 Alongside this observation, the institutional arguments laid out in this chapter point ineluctably to a question: viewing the United States in comparison to European parliamentary regimes, are the twin phenomena of a more modest administrative state, and a greater role for private litigation in policy implementation, opposite sides of the same institutional coin? Kagan offers an affirmative answer to this question, stating: “It is only a slight oversimplification to say that in the United States lawyers, legal rights, judges, and lawsuits are the functional equivalent of the large central bureaucracies that dominate governance in high-tax, activist welfare states.”114 Kagan makes this statement in the context of a book that maps the operation of adversarial legal process in the United States in the areas of both criminal and civil
58
CHAPTER 2
law; across constitutional, statutory, and common law; in the context of both regulatory and welfare state functions; and at both the state and federal levels. His main evidentiary goal in the book is to demonstrate the pervasiveness of distinctive patterns of “adversarial legalism” in the United States, not its causes. This book’s more modest canvas—federal statutory regulation—and its central theoretical and evidentiary focus on what causes legislative enactment of private enforcement regimes, can shed light on the comparative institutional question within the domain of federal regulation. The institutional arguments presented in this chapter and empirically investigated in the next four chapters link the same institutional distinctions between separation of powers and parliamentary systems, upon which Moe’s and Steinmo’s arguments about the “weak” American state hinge, to congressional enactment of private enforcement regimes. If the evidence bears out these claims about the institutional foundations of private enforcement regimes, it will suggest that the institutional explanations for modest administrative state capacity in the United States, and for the unusually large role of private litigation in public policy implementation, are one and the same. That is, it will suggest that limited administrative state capacity on the one hand, and extensive private litigation in policy implementation on the other, are indeed linked outcomes of the same institutional causes and processes. This would provide potentially fruitful comparative institutional lessons for understanding the role of private litigation, and its relationship to bureaucratic forms of regulatory implementation, in different national regulatory systems. Additional Theoretical Perspectives on Institutional Creation and Development This chapter developed an argument about how American state structures create incentives for enactment of private enforcement regimes. The theoretical account portrays legislators as rational, strategic actors pursuing regulatory policy goals in the face of a distinctive set of institutional constraints and opportunities. The purpose of the theoretical model is to explain the choices of members of Congress when enacting private enforcement regimes in general, across many policy domains, and over a long stretch of time. Such a general model of legislative choice of regulatory form, by its very nature, is based partly upon stylized assumptions about human behavior and capacity, portraying actors’ preferences and understandings of policy options as static, and representing them as capable of effectively predicting the future effects of institutional choices.115 Political, social, economic, and historical processes are, of course, typically more complex than these stylized assumptions suppose. When one
INSTITUTIONAL FOUNDATIONS
59
examines the concrete historical experiences of specific policy interventions and communities over time, one may find that institutions do not always behave consistently with their creators’ expectations, and that policy interventions can have “feedback” effects that transform the preferences of important actors by generating learning effects and by changing the incentive structures that they face.116 It is readily acknowledged that historically specific and textured inquiries into the reasons for the genesis and evolution of particular implementation regimes over time will require the consideration of causal processes in addition to those laid out in this chapter. Part II of this book—which provides a detailed historical analysis of the origins and development of the enforcement framework governing many important federal civil rights laws—will argue that policy feedback processes flowing from the Civil Rights Act of 1964, involving both policy learning and shifting incentive str uctures faced by civil rights advocates, are critical to understanding that case. The operation of such feedback processes, it should be stressed, is not at all in conflict with the rationalist institutional hypotheses developed in this chapter linking American state structures to the legislative choice of private enforcement regimes. Instead, as we shall see in part II of this book, the feedback processes flowed through, and were shaped by, the structural-institutional dynamics laid out in this chapter.
Chapter 3 AN EMPIRICAL MODEL OF ENACTMENT OF PRIVATE ENFORCEMENT REGIMES
Chapter 1 suggested that scholars have paid insufficient attention to private enforcement regimes. Perhaps nothing illustrates this better than the fact that no general measure of congressional enactment of private enforcement regimes has been devised to date, and, naturally then, no data has ever been collected on Congress’s propensity to enact private enforcement regimes over time. Scholars who have treated the statutory structure of judicial remedies as an important political variable have generally been limited, empirically, to case studies focusing on major legislative events.1 Rich and excellent as this research is, the development of an objective measure of congressional enactment of private enforcement regimes over time can add to it by allowing hypothesis testing about causes of this form of state building, such as the separation of powers hypotheses set forth in chapter 2. This chapter first explains and defends a measure of Congress’s enactment of private enforcement regimes over time, and provides a brief portrait of the variable between 1887—the year most commonly associated with the birth of the modern American regulatory state, marked by passage of the Interstate Commerce Act—and 2004. It then presents an empirical model to test a series of hypotheses about what causes Congress to mobilize private litigants to accomplish regulatory enforcement, including separation of powers hypotheses developed in chapter 2. The empirical findings provide robust support for the central separation of powers hypothesis: ideological conflict between Congress and the president is a statistically significant, consistent, and substantively powerful predictor of congressional enactment of incentives to mobilize private litigants. These findings link long-run historical patterns of divided government and legislative-executive polarization, which increased in frequency and intensity starting in the late 1960s, with the coincident growth of the role of litigation and courts in the implementation and elaboration of federal statutory policy. The model also provides support for the coalition drift hypothesis, and qualified support for the bureaucratic drift hypothesis, suggesting that legislators and the interest groups that influence them
EMPIRICAL MODEL OF ENACTMENT
61
understand that private enforcement regimes can insulate not just from presidential subversion, but also, given the stickiness of the status quo in the American separation of powers system, from subversion by future legislative majorities and errant bureaucrats.
Measuring Private Enforcement Regimes An exhaustive measure of the larger constellation of elements that comprise private enforcement regimes, such as rules governing damages, evidence, proof, liability, the scope of standing, statutes of limitations, judge versus jury fact-finding, allocation of responsibility for attorney’s fees, and so on, is not feasible. This may well be the reason that no effort has previously been made to collect data that would allow testing of hypotheses regarding causes of the legislative mobilization of private litigants. A few clear and discrete indicators of Congress’s efforts to mobilize private litigants are needed. The best such indicators are (1) statutory provisions requiring that defendants pay successful plaintiffs’ attorney’s fees, and (2) statutory provisions providing that successful plaintiffs are entitled to monetary damages that exceed the actual material harm suffered. Data was collected on the full universe of such provisions enacted by Congress from 1887 to 2004. Because no previous attempt has been made to devise a general measure of congressional mobilization of private litigants, it is necessary to provide a detailed explanation of why these measures have been selected. These empirical measures focus on statutory attributes that depart from the default rules of practice in the federal legal system and directly and intentionally change the rules of the game so as to increase the expected monetary value of successful legal claims. They measure Congress’s propensity not just to allow, but to stimulate, private enforcement litigation. As discussed in chapter 2, merely allowing private enforcement is a far cry from assuring that it will actually be mobilized. The high costs of litigation and the modest actual economic damages caused by violations in many federal regulatory contexts will often produce negative net value claims unless Congress elects to include rules aimed at bolstering their expected value. The standard “American rule” is that each party pays its own attorney’s fees and other costs of litigation, whereas the “English rule” (which prevails in Europe, and most of the rest of the world) provides that the loser pays most of the winner’s fees and costs.2 Accordingly, as the Supreme Court has explained, federal courts follow “a general practice of not awarding fees to a prevailing party absent explicit statutory author-
62
CHAPTER 3
ity.”3 When Congress explicitly departs from this default rule and enacts a statutory fee-shifting provisions in favor of plaintiffs, defendants under the statute (the objects of regulation) are compelled to pay the legal fees and expenses (the costs of regulatory enforcement) of a successful plaintiff-enforcer. These fee shifts are asymmetrical, however, and a prevailing defendant is not similarly entitled to recover its attorney’s fees from the plaintiff. In the language of the rational litigant model in chapter 2, a plaintiffs’ fee shift increases the expected value of her claim by reducing her expected costs. Among the multiple potential arrangements for allocating responsibility for paying the costs of litigation, the plaintiff’s shift creates the greatest incentives for plaintiffs to file enforcement actions.4 The Supreme Court has referred to plaintiffs’ fee shifts as “congressional utilization of the private-attorney-general concept,” while noting that “under some, if not most, of the statutes providing for the allowance of reasonable fees, Congress has opted to rely heavily on private enforcement to implement public policy and to allow counsel fees so as to encourage private litigation.”5 The general baseline rule governing monetary damages under a federal statute is that, in the absence of a contrary statutory direction, successful plaintiffs are entitled to monetary damages proportional to the harm or loss they suffered, not more.6 However, Congress sometimes enacts express statutory provisions that confer monetary damages greater than a plaintiff’s actual material damages, such as double, triple, or punitive damages. Double or triple damages operate as multiples on the actual monetary damages suffered by the plaintiff, and punitive damages can be awarded separately in an amount that need not be tied to actual monetary harm at all, and can far exceed it. In the language of the rational litigant model in chapter 2, damages enhancements increase the expected value of a plaintiff’s claim by increasing her expected benefits. Courts have recognized that such damages enhancements “are justified as a ‘bounty’ that encourages private lawsuits seeking to assert legal rights.”7 Like plaintiffs’ fee shifts, using damages enhancements as a “bounty” operates to “reward individuals who serve as ‘private attorneys general’ in bringing wrongdoers to account,”8 and provides an “incentive to litigate” that is “designed to fill prosecutorial gaps.”9 There are three principal reasons for selecting plaintiffs’ fee shifts and damages enhancements as key indicators of the larger constellation of elements that comprise private enforcement regimes. First, they are unambiguous in their purpose and influence. With respect to some other elements of private enforcement regimes, it cannot be clearly generalized which party benefits outside the context of a particular policy area at a particular point in time. The use of jury trials in the civil rights context
EMPIRICAL MODEL OF ENACTMENT
63
illustrates this point. Advocates of civil rights enforcement originally opposed jury trials in favor of bench trials (where the judge acts as fact finder) under the Civil Rights Act of 1964’s job discrimination provisions because of fear that local juries would nullify the law in the South, while some southern opponents of the law unsuccessfully sought to have a jury trial provision added.10 In the CRA of 1991, however, it was plaintiffs’ advocates who successfully obtained the right to trial by jury for the CRA of 1964’s job discrimination title—significantly motivated by a belief that federal judges had grown increasingly antiplaintiff in the civil rights context—over the objections of probusiness interests who feared juries would award higher damages than judges.11 About a quarter century after the original law was passed, under changed historical conditions, civil rights advocates believed that they would be benefited rather than harmed by jury trials. In contrast, plaintiffs’ fee shifts and damages enhancements, perforce, only cut in one direction because they necessarily lower the monetary costs and increase the monetary gains for successful enforcement by plaintiffs, and thus their purpose and influence are robust to changing historical conditions. Second, plaintiffs’ fee shifts and damages enhancements are effective across all policy areas. Other elements of private enforcement regimes may be more likely to be used in some policy areas than others. Consider “citizen suit” provisions, which allow individuals to bring suit under a statute even if they have not suffered a concrete harm. Such provisions have rarely been used outside the context of allowing suits to complain of governmental action or inaction, and allowing suits under environmental statutes, areas in which the requirement of concrete harm for standing could lead to underenforcement.12 The use of such provisions as a general measure of Congress’s propensity to mobilize private litigants would be strongly biased toward reflecting legislation to regulate government officials and to protect the environment. Plaintiffs’ fee shifts and damages enhancements, on the other hand, can serve as the bedrock of a private enforcement regime in any policy area by lowering the costs and increasing the rewards of enforcement. Indeed, as discussed below, they have been used in virtually every significant domain of policy in which Congress has elected to intervene, from antitrust, to banking, to securities, to labor, to civil rights, to environmental protection. They thus serve as an effective measure of Congress’s propensity to mobilize private litigants in the service of its regulatory purposes, regardless of policy area. Third, plaintiffs’ fee shifts and damages enhancements have been purposefully and self-consciously used by legislators since the founding of the modern American regulatory state in the late nineteenth century, most commonly associated with passage of the Interstate Commerce Act of 1887. Thus, unlike some later procedural innovations, these measures
64
CHAPTER 3
are effective over the life span of the American federal regulatory state, which is the duration covered by the empirical model presented below. The use of damages multiples has ancient roots and has been used by American legislators since the colonial period.13 Plaintiffs’ fee shifting was inaugurated in federal statutory law with the Civil Rights Act of 1870 (also known as the Force Act of 1870), which was enacted to protect African American voting rights against Ku Klux Klan terrorism in the Reconstruction South, where states had proven unwilling or unable to control widespread violence against African Americans.14 The Interstate Commerce Act of 1887 contained a plaintiffs’ fee-shifting provision, and the Sherman Antitrust Act of 1890 contained the first combination of a plaintiffs’ fee shift and a damages multiple (it was triple). In the floor debates on the Interstate Commerce Act and the Sherman Act, legislators said explicitly that they were departing from the default commonlaw rules on attorney’s fees and damages for the purpose of mobilizing private litigants in the regulatory struggle to bring powerful economic interests under control. This point about legislative self-consciousness in using plaintiffs’ fee shifts and damages enhancements as instruments to mobilize private litigants during the full duration of the study is critical for two reasons. First, the main arguments of this book hinge upon the purposeful use of private enforcement regimes as a regulatory tool. The notion that legislators deploy private litigants and plaintiffs’ attorneys as a source of state capacity, that they understand them as an adjunct or alternative to administrative power, and that choices to enact private enforcement regimes are sometimes made strategically in the face of institutional constraints and opportunities, all contemplate a high degree of intentionality. Second, because these variables are used as indicators of legislative recourse to a whole body of rules whose unifying purpose is to mobilize private litigants, their efficacy as indicators depends critically upon the claim that when legislators have enacted them, litigant mobilization has been their purpose from the start. This assumption is borne out by the legislative historical evidence. The common law actually provided causes of action for some of the harms that the Interstate Commerce Act and the Sherman Act sought to prevent, but the American common law’s default rules on attorney’s fees and damages were such that, from legislators’ point of view, an inadequate level of private litigation was being mobilized. In debates over the ICA of 1887, Senator Spooner of Wisconsin argued that because the damages caused to individual shippers by unlawful railroad rate practices were a “trifling amount,” and protracted litigation against railroads was “burdensome and expensive,” in economic terms “even success at the end of a long litigation was to the complainant defeat.” As a result, common-
EMPIRICAL MODEL OF ENACTMENT
65
law rules had “been utterly futile as a shield against wrong by the carrier.” It was necessary, therefore, to create a new enforcement regime that would provide, among other things, that “if the railway company is defeated it shall pay the costs of the prosecution.” Spooner further noted that “[i]f incorporated into law,” this approach to enforcement “will be a new feature of law in this country.”15 While not strictly true, since the Force Act of 1870 was actually the first plaintiffs’ fee shift in federal law, Spooner was broadly right that the ICA’s departure from the American common law’s default rules governing attorney’s fees in private litigation was a historically significant departure. Representative Hopkins of Illinois stated in reference to the ICA’s private enforcement regime that the “mode of enforcing” the law was “different from common law remedies—made so to meet a condition of affairs which was not contemplated at common law.” That condition was, according to Hopkins, that because of “the growing sovereignty of the rail roads” they were “more powerful and exercise[d] a more direct influence upon the people than the State,” and thus they had to be brought under regulatory control by fashioning new rules to govern private enforcement litigation that departed from common-law practices.16 Representative Crisp of Georgia characterized the ICA’s plaintiffs’ fee shift as an “exceptional” provision needed to counterbalance the vast economic power of railroads and thereby “enable you to enforce your rights by paying the counsel that you employ for that purpose.”17 In debates over the Sherman Antitrust Act of 1890, the legislation’s chief sponsor, Senator John Sherman of Ohio, stated in reference to the triple damages provision: “The measure of damages, whether merely compensatory, putative, or vindictive, is a matter of detail depending upon the judgment of Congress. My own opinion is that the damages should be commensurate with the difficulty of maintaining a private suit against a combination,” rather than being limited to the harm actually suffered.18 Echoing Sherman, Senator James George of Mississippi stated: “Treble damages were provided in part for punitive purposes, but also to make the remedy meaningful by counterbalancing the difficulty of maintaining a private suit against a combination.”19 Senator James was concerned that remedies under the law should “further facilitate the small men, the men of small capital in getting redress for the great wrongs that are perpetrated against them by these trusts.”20 From the dawn of the American regulatory state, legislators’ enactment of private enforcement regimes—their self-conscious adjustment of the expected value of lawsuits by manipulating expected costs and benefits, with the goal of stimulating private litigation to enforce statutes—was no story of unintended consequences. They knew what they were doing from the start.
66 Cumulative enactments 400
CHAPTER 3 Private statutory litigation rate 40 Cumulative enactments
300
30
200
20 Private statutory litigation rate
100
10
1885 1895 1905 1915 1925 1935 1945 1955 1965 1975 1985 1995 2005
Figure 3.1 Private Enforcement Regimes, 1887–2005, and Private Statutory Litigation Rates, 1942–2005
Data was collected on the full universe of plaintiffs’ fee shifts and damages enhancements enacted into federal law from 1887 to 2004.21 The search yielded 275 plaintiffs’ fee shifts and 104 damages enhancements, for a total of 379 litigation incentives. Each such provision was coded for when it was enacted and the year it ceased to be operative if the statute was repealed by Congress (8 percent), expired of its own terms (3 percent), or was struck down by the Supreme Court (2 percent). Of plaintiffs’ fee shifts and damages enhancements enacted since 1887, 87 percent remain in effect. This is consistent with the theoretical expectation, discussed above, that formal legal enactments will tend to endure in the American separation of powers system, and that private enforcement regimes, once enacted, will likely persist through the reign of future governing coalitions. Of the 13 percent to exit the United States Code, in all instances the exit occurred because the statute or some substantive portion of it was repealed, expired, or struck, and in no instance was the fee shift or damages enhancement itself specifically cut out. Thus, while it would be sensible to deduct from annual counts proactive decisions by Congress to eliminate fee shifts and damages enhancements while leaving the underlying right intact, no such events occurred.22 The darker line in figure 3.1 represents the cumulative number of plaintiffs’ fee shifts and damages enhancements in effect annually (accounting for provisions that exit the United States Code), reflecting the structural environment of private enforcement regimes in existence annually. The lighter line in figure 3.1 is the annual rate, per 100,000 population, of pri-
EMPIRICAL MODEL OF ENACTMENT
67
Table 3.1 Private Enforcement Regimes by Policy Area Policy area
% of total enactments
Property
12%
Civil rights
9%
Consumer protection
9%
Labor
9%
Public health and safety
8%
Banking
6%
Environmental
6%
Interstate commerce
6%
Securities and commodities exchange
6%
Housing
5%
Antitrust
4%
Bankruptcy
2%
Communications
2%
Elections Other*
2% 14%
*Includes all policy areas with less than 2%.
vate federal statutory enforcement litigation (it is only possible to distinguish privately from governmentally filed actions beginning in 1942).23 The strikingly close association between these two variables, and particularly the coincident sharp upward shift in both at the end of the 1960s, reinforces the plausibility of plaintiffs’ fee shifts and damages enhancements as measures of the broader phenomena of private enforcement regimes, and of the efficacy of private enforcement regimes in mobilizing private litigants. The figure also illustrates a ratchet effect on statutory litigation in the federal system produced by the combination of enactments over time and their durability through time. The enactment of private enforcement regimes has occurred across the waterfront of federal regulation, and is not predominantly driven by some few policy domains. Table 3.1 reflects the distribution of fee shifts and damages enhancements enacted between 1887 and 2004 across policy domains. Nine percent of the incentives provide for actions against states, 17 percent do so against the federal government, and 84 percent apply against private entities (the categories are nonexclusive). Just as private enforcement regimes are not relegated to some narrow set of policy domains, it is evident from examining the substance
68
CHAPTER 3
of statutes including them that private enforcement regimes are not the unique province of any particular ideological or partisan program, nor have they been exclusively deployed to serve any specific type of constituency. Consistent with conventional expectations (discussed in the next section), more liberal and Democratic Congresses have enacted private enforcement regimes directed at business regulation, for example, serving constituencies of low-wage workers under the Fair Labor Standards Act of 1938,24 minority groups under the Civil Rights Act of 1964,25 and consumers under the Telemarketing and Consumer Fraud and Abuse Prevention Act of 1994.26 Less consistent with conventional expectations, more conservative and Republican Congresses have also found private enforcement regimes to be a useful regulatory strategy to serve their constituents as well. For example, in the Taft-Hartley Act of 1947 they gave companies a private right of action with economic damages against unions engaged in labor actions proscribed by the act;27 in the Cuban Liberty and Democratic Solidarity (Libertad) Act of 1996, they gave United States nationals whose property was confiscated by the Cuban government during or following the Cuban revolution a private right of action, with attorney’s fees for successful plaintiffs, against “traffickers” in such property;28 and in the “Partial-Birth Abortion” Ban Act of 2003 they created a private right of action with treble damages, and damages for emotional pain and suffering, for fathers (if married to the woman on whom the procedure is performed), and for “maternal grandparents of the fetus” if the woman is a minor, against a doctor who performs an abortion in violation of the act.29 Whatever a congressional majority’s partisan affiliation, cause, or constituency, if it wishes to control the behavior of other entities, private enforcement regimes are a tool that it can draw upon. This is not to discount the possibility of partisan effects on Congress’s propensity to enact private enforcement regimes, which will be discussed in the next section and tested in the empirical model presented in the final section.
Other Causes of Private Enforcement Regimes Chapter 2 discussed a number of potential institutional causes of the legislative choice to enact private enforcement regimes. There are an additional four potential causes of legislative enactment of private enforcement regimes that appear repeatedly in the scholarly literature, and which must be controlled for in the empirical model in order to isolate the effects, if any, of separation of powers structures. Moreover, they are of considerable interest in themselves because, although they have appeared in the scholarly literature for some time, none has been empirically tested
EMPIRICAL MODEL OF ENACTMENT
69
because of a lack of appropriate data. They also represent important perspectives on causes of private enforcement regimes that will reappear in subsequent chapters, and some are importantly entwined with the institutional variables that are the main focus of this book. Rent-Seeking Lawyer Hypothesis. Two main arguments causally link interest group activity to enactment of private enforcement regimes. In the first, rent-seeking lawyer interest groups, such as the Association of Trial Lawyers of America, securities class action lawyers, labor and employment lawyers, and the American Bar Association, lobby to create and maintain opportunities for remunerative litigation so as to enrich themselves.30 Professional associations frequently seek to advance and protect the economic interests of their members, and lawyers associations are no different. Under the sway of lawyer interest groups, “[L]egislators frequently benefit lawyers by passing . . . legislation known popularly as ‘lawyers’ relief acts,’ ” which make fee-generating litigation central to implementation.31 According to the rent-seeking lawyer hypothesis, greater lobbying influence by lawyers’ associations will be associated with increased enactment of private enforcement regimes. Issue Group Hypothesis. A second line of explanation is that beginning in the middle to late 1960s, issue-oriented citizens groups with proregulatory agendas, such as environmental, civil rights, and consumer protection organizations, burst on the American policy scene, flourished in their number and lobbying intensity, and successfully demanded courtbased implementation. This view links, implicitly, the interest group “explosion” of the late 1960s and early 1970s32 to the storied “explosion” of at least some kinds of litigation. As contrasted with the personal economic motivations of the rent-seeking lawyer groups, these issue groups are guided by policy preferences. Their rising preference for private enforcement regimes represented a turn away from the New Deal vision of governance through expert bureaucracy. It was motivated by a growing distrust of bureaucracy, which they regard as timid, apathetic, establishment oriented, and prone to be captured by the entities it was supposed to regulate. The growth in citizens groups included groups specifically dedicated to reform through litigation, and these “public interest” law groups were especially focused on lobbying Congress for private enforcement regimes with sufficient economic rewards to bankroll their continued litigation campaigns.According to the issue group hypothesis, greater influence and participation by issue groups in the legislative process will be associated with increased enactment of private enforcement regimes.33 It is important to stress that the issue group hypothesis is fundamentally entwined with the bureaucratic drift hypothesis discussed in the last
70
CHAPTER 3
chapter. The logic of the bureaucratic drift hypothesis is that an enacting legislative coalition, mindful that the stickiness of the status quo will make legislative override of bureaucrats difficult, will want to safeguard in advance against bureaucrats deviating from the coalition’s preferences based upon bureaucrats’ own policy preferences, careerism, propensity to shirk, or acquiescence to capture. The claim of the issue group hypothesis that groups regard bureaucrats as establishment oriented, timid, apathetic, and susceptible to capture is, manifestly, of a piece with the concerns animating the bureaucratic drift hypothesis. That is, the issue group and the bureaucratic drift hypotheses are predicated upon the same concerns about bureaucratic behavior. Further, the logic of the issue group hypothesis contemplates that interest groups may be influential members of a legislative coalition. Such coalitions are partly constituted by an alignment of the policy positions of legislator voting blocs, and the preferences of interest groups that are key constituents and supporters of the legislators comprising the blocs.34 When one adds that American interest groups are likely to understand that they are operating in an institutional environment in which the stickiness of the status quo can give bureaucrats considerable space to pursue personal preferences before legislative reversal will occur, the issue group and bureaucratic drift hypotheses converge. One notable difference is that the issue group hypothesis identifies interest groups as a particular locus of concern about bureaucratic drift, and it argues that, historically speaking, the concern emerged as a more important influence on regulatory lawmaking starting in the middle to late 1960s. Party Alignment Hypothesis. Arguments that interest groups—whether of the rent-seeking lawyer or issue-oriented variety—are key causes of adversarial legalism in the United States have been closely tied to political party. These arguments link the plaintiffs bar to the Democratic party, and link business opponents of civil legal liability to the Republican party. Claims about linkages between the plaintiffs bar and the Democratic party nearly always focuses on the for-profit plaintiffs bar. While connections of the plaintiffs personal injury bar to the Democratic party have received the most extensive attention,35 scholars have pointed to apparent influence by plaintiffs’ lawyers on Democrats more broadly, fostering litigation across such areas as civil rights, consumer protection, the environment, securities and exchange, and health care;36 seeking to shape rules of civil procedure, such as the class action device and the standard for entering judgment against a plaintiff without a trial, so as to strengthen plaintiffs’ position in litigation;37 and discouraging alternative dispute resolution mechanisms that would divert disputes away from adversarial legal venues.38 In an analysis of the parties’ respective
EMPIRICAL MODEL OF ENACTMENT
71
positions on federal securities regulation, Roberta Romano observes that “Republicans’ general support for and Democrats’ opposition to litigation reform that restricted liability . . . paralleled the perspective of key party constituencies, the business community for Republicans and the plaintiffs’ bar for the Democrats.”39 Other scholars have suggested that pro-regulatory issue groups— including but not limited to legal advocacy groups, which proliferated starting in the middle to late 1960s and supported court-based implementation—emerged as a core constituency of the Democratic party coalition, which was broadly more likely to support regulatory state intervention than Republicans.40 The rent-seeking lawyer and the issue group hypotheses, discussed above, identify the motives of for-profit lawyers and issue advocacy groups in seeking to shape the Democratic party’s positions: the pursuit of personal enrichment for the former, and the pursuit of policy goals for the latter. According to the party alignment hypothesis, Democratic control of Congress will be associated with enactment of private enforcement regimes. The party alignment hypothesis can be construed weakly or strongly. Construed weakly, it simply suggests that, other things being equal, Democratic-controlled Congresses will be more likely to enact private enforcement regimes than Republican ones. Construed strongly, it suggests that the construction of private enforcement regimes is a uniquely Democratic phenomenon, abhorred by Republicans and their constituents, and thus the influence of other theorized causes of legislative enactment of private enforcement regimes, such as divided government, electoral uncertainty, and interest group mobilization may all be conditional upon Democratic control of Congress. Both versions will be tested in the model presented below. Budget Constraint Hypothesis. A number of scholars have argued that lack of adequate tax revenue encourages Congress to achieve public policy goals through private adversarial legal process because it shifts the costs of regulation away from the state and to private parties.41 Among the more economically developed democracies, the United States extracts the least revenue from the polity as a proportion of gross national product.42 This places obvious limits on state building of a bureaucratic form. As distinguished from funding an agency to carry out enforcement activities, private enforcement regimes are, from Congress’s standpoint, more or less self-enforcing. Moreover, though increasing rates of litigation will cause some increase in the costs of maintaining the federal judiciary, these costs are not easily traceable by voters to legislators’ support for a piece of regulatory legislation with a private enforcement regime. Thus, with private enforcement regimes Congress can hope to achieve its aims on
72
CHAPTER 3
the cheap, and to minimize blame for what implementation costs are borne by the government. According to the budget constraint hypothesis, when resources are tight, Congress will be relatively more likely to enact private enforcement regimes.
An Empirical Model of Private Enforcement Regimes
Separation of Powers Conflicts. In order to test the hypotheses set out in chapter 2 regarding the relationship between interbranch conflict and the use of private enforcement regimes, two alternative sets of measures are used of political actors’ ideological preferences and the extent of congruence or conflict between them, one based on partisan identification and the other on NOMINATE scores. On the partisan side, in model 1 a simple Divided Government variable is used, coded 0 when the president’s party controls both chambers of Congress, and coded 1 otherwise. As a more sensitive alternative to measuring interbranch partisan conflict, in model 2 an Opposition Seat Share variable is substituted, which is defined as the proportion of seats held by the party opposite the president minus the proportion held by the president’s party, averaged across both chambers.43 It is coded 0 when the House and Senate are controlled by different parties. The variable will be positive when the party opposed to the president controls Congress, negative when the president’s party controls Congress, and of increasing magnitude the larger the margin of control. As an alternative approach to measuring ideological preferences and interbranch conflict, Poole and Rosenthal’s first-dimension common space NOMINATE scores are used, which are only available beginning in 1937.44 The NOMINATE procedure is based upon a Downsian spatial theory of voting and creates estimates of the ideological positions of legislators on an interval scale based upon their pattern of roll call voting behavior. The scores range between −1 and 1. In Poole and Rosenthal’s original calculation, negative scores are more liberal and positive scores are more conservative, but the direction has been inverted for purposes of consistency with the party variables discussed below, so positive scores are more liberal and negative scores are more conservative. The common space ideology scale is standardized so as to render the scores comparable across chambers and over time. To model presidential ideology, McCarty’s presidential NOMINATE scores are used, which are based upon public positions taken by the president on roll call votes.45 Using this set of roll calls for each president, the president is then treated as a legislator for purposes of estimating his ideological position.46
EMPIRICAL MODEL OF ENACTMENT
73
In measuring judicial ideology, the federal court of appeals is incorporated. Judicial ideology at the court of appeals level is incorporated rather than the trial or Supreme Court level because each litigant is entitled to have any trial court decision reviewed by the court of appeals as a matter of right, whereas the overwhelming majority of cases have no meaningful prospect of obtaining Supreme Court review. Further, because of the Supreme Court’s limited opportunity to extensively flesh out the meaning of particular statutes, and the trial court’s lack of authority to issue decisions that bind in future cases, a huge majority of binding statutory interpretation comes from the court of appeals. A combination of appointing presidents’ and senators’ NOMINATE scores is used to capture the effects of senatorial courtesy on appointments. Each judge is assigned the value of the president’s NOMINATE score when an appointment is in a state in which neither senator is from the president’s party, and each judge is assigned the value of the senator’s (or the average of the two senators’) NOMINATE score(s) when an appointment is in a state in which one or both senators belong to the president’s party. Giles, Hettinger, and Pepper demonstrate that incorporating senatorial courtesy in this fashion produces a better predictor of court of appeals judges’ voting behavior than imputing only the president’s NOMINATE score or party to a judge.47 The judiciary’s ideological position is then measured with the average score of federal court of appeals judges. Relative to the party variables, common space NOMINATE scores have the advantage of allowing one to locate the ideological position of the judiciary relative to the president and Congress at the same time. Presidential Distance is the absolute value of the distance between the presidential NOMINATE score and the congressional mean, measured as the average of the House and Senate median NOMINATE scores. Judicial Distance is the absolute value of the distance between the average court of appeals NOMINATE score and the congressional mean. Alternative measures of Presidential Distance and Judicial Distance, operationalizing Congress’s ideological position differently, yielded nearly identical results.48 Judicial Direction is coded 0 when the president and Congress are on the same side of the judiciary in ideological space, such that the appointment of additional judges will draw the judiciary closer to Congress, and coded 1 when Congress and the president are on opposite sides of the judiciary, such that appointment of judges will draw the judiciary away from Congress. Risk of Electoral Losses. Testing the hypothesis that majority coalitions in Congress will be more likely to use private enforcement regimes when faced with increases in the risk of electoral losses poses an empirical chal-
74
CHAPTER 3
lenge. It would be ideal to have a measure of legislators’ contemporaneous beliefs about electoral risk, but no data exists to construct such a variable going back as many decades as the present study calls for. Researchers studying nineteenth-century Congresses have dealt with this problem by projecting backward based upon actual electoral outcomes.49 Theoretically, the expectation that the risk of electoral losses will influence policy design is predicated on the assumption that current majorities are able to gauge, to some meaningful extent, the degree of risk they face in an upcoming election. Electoral Risk reflects seat gains or losses as a proportion of total seats in the next election by majority parties, averaged across the two chambers. The variable has positive values when the majority party loses seats in the next election, and negative values when it gains seats, and it will indicate whether congressional majorities heading into electoral losses are more likely to use private enforcement regimes for regulatory implementation. Partisan Control of Congress. Partisan Seat Share is positive when Democrats control Congress and reflects their margin of total seats over Republicans, and is negative when Republicans control Congress and reflects their margin of total seats over Democrats. It is important to be clear about what this variable is isolating. As discussed below, the models will contain controls for how much and what type of regulatory legislation Congress is enacting. Thus, for example, if more Democratic Congresses enact more economic regulation than more Republican ones, and as a function of this behavior they enact more private enforcement regimes, this tendency will be picked up by the controls for the extent and type of regulatory legislation Congress is enacting. The Partisan Seat Share variable therefore isolates whether, given some level and kind of regulatory intervention, partisan control influences the selection of private enforcement regimes as an instrument of implementation. Magnitude and Nature of Regulatory Productivity. Congress’s enactment of private enforcement regimes may vary over time merely as a function of the ebb and flow of Congress’s production of regulatory legislation in general, and of shifts in the nature of its regulatory agenda. The literature on regulation recognizes a basic distinction between economic and social regulation.50 Economic regulation, which typically targets particular industries, is generally aimed at promoting market stability, efficiency, and competition, and is “concerned with preventing undue economic concentration, regulating natural monopolies, eliminating economic windfalls, ensuring adequate distribution of goods and services, and reducing fraud in economic transactions.”51 Social regulation, which typically cuts across all industries and sectors, is generally aimed at addressing problems of
EMPIRICAL MODEL OF ENACTMENT
75
externalities, inadequate information, and public goods, and it is concerned with promoting public health and safety, consumer protection, environmental protection, equal opportunity, and quality of life in general. During the Progressive and New Deal periods, economic regulation was the main source of the growth in federal regulatory statutes, whereas by the early 1970s social regulation became the main focus of regulatory legislation.52 It is possible that Congress’s use of private enforcement regimes will vary across these domains. In order to capture these factors, a random sample was drawn of 20 percent of all public laws enacted between 1887 and 2004, and within that sample regulatory laws were identified (554 laws), and each was coded as economic regulation, social regulation, or other. For each classification, annual counts were calculated of the number of pages of regulatory legislation in the Statutes at Large from 1887 to 2004, and these counts were used to create the variables Economic Regulation, Social Regulation, and Other Regulation. Interest Groups. To the extent that interest groups are participating in shaping regulatory legislation, a visible venue in which this activity occurs is committee hearings, where committee members select representatives of groups and invite them to offer their views on legislation. The witnesses who appear reflect whose interests and preferences legislators decide to involve in the policymaking process. In order to measure interest group influence upon and participation in the legislative process, with respect to the random sample of 554 regulatory laws, all committee hearings held on the laws were identified, witness lists for each hearing were examined, and each witness was coded for the type of organization or interest she represented. Variables were then generated defined as the annual count of witnesses of various types appearing in the hearings. Issue Witnesses is the number of witnesses appearing in support of issueoriented citizens groups, such as Friends of the Earth, American Association of Retired Persons, and National Council of La Raza. Issue groups focused on law reform through litigation as a principal tactic, frequently referred to as “public interest law” groups, such as the National Women’s Law Center, the National Housing Law Project, and the National Consumer Law Center, are included in Issue Groups. When incorporated into the model as a separate variable, they are not independently significant.53 Lawyer Witnesses is the number of witnesses appearing on behalf of lawyers associations, such as the American Bar Association and the Association of Trial Lawyers of America (ATLA). Lawyers associations that exclusively represent for-profit plaintiffs’ lawyers, such as ATLA, constituted only about 5 percent of lawyers testifying on behalf of lawyers associations. When incorporated into the model separately, in an alternative specification, that variable was not significant. Business Witnesses is the
76
CHAPTER 3
annual count of witnesses appearing on behalf of businesses or business organizations, such as the National Association of Manufacturers, the Chamber of Commerce of the United States, and the Securities Industry Association. A number of scholars have suggested that business interests, at times, have mobilized against the use of private enforcement regimes,54 and thus this variable is necessary to control for the possibility that it will be negatively associated with enactment of private enforcement regimes. Budget Constraints. Budget Constraint is the size of the federal budget surplus or deficit relative to total expenditures.55 Larger negative numbers indicate relatively sizeable deficits, and larger positive numbers indicate substantial surpluses.56 Findings The statistical model employed is designed to analyze an “event count,” such as the number of statutory litigation incentives enacted by Congress each year, which is the dependent variable in the models below. The selection, application, and interpretation of the statistical models are explained in the appendix to this chapter. The substantive meaning of coefficients in a count model are not directly interpretable. The last paragraph of the appendix’s discussion of statistical methods explains how the magnitude of the effects of independent variables, discussed below, were calculated. Legislative-Executive Conflict. Regarding the most straightforward measure of legislative-executive conflict, model 1 in table 3.2 shows that divided government increases Congress’s enactment of private enforcement regimes. The Divided Government variable is statistically significant and positive, with a large substantive effect. A move from unified to divided government nearly doubles the predicted number of enactments, increasing it by 93 percent. Model 2 substitutes Opposition Seat Share for the divided government dummy variable as the measure of legislative-executive conflict. Opposition Seat Share is the proportion of seats held by the party opposite the president minus the proportion held by the president’s party, averaged across both chambers. It will be positive when the party opposed to the president controls Congress, and of increasing magnitude the larger the margin of control. It is statistically and substantively significant as well. An increase in Opposition Seat Share by 19 percent (one standard deviation) is associated with a 36 percent increase in enactments. To put this in perspective, the increase in Opposition Seat Share when Nixon replaced Carter, holding other variables constant, rendered a 77 percent increase in predicted enactments.
77
EMPIRICAL MODEL OF ENACTMENT
Table 3.2 Autoregressive Poisson Models of Enactments Variable Divided Government Opposition Seat Share
1887–2004 Model 1 .655** (.254)
1887–2004 Model 2
1.63** (.567)
Presidential Distance Judicial Distance Judicial Direction Electoral Risk Partisan Seat Share Lawyer Witnesses Issue Witnesses Business Witnesses Budget Constraint Social Regulation Economic Regulation Other Regulation Constant Residuals Lag 1 Residuals Lag 2 R2 N AEG
3.00** (.942) 1.44* (.680) −.024 (.096) .025** (.007) −.003 (.004) .045 (.630) −.001 (.001) .005* (.002) .033* (.015) .073 (.250) .330** (.102) .309** (.111) .60 116 −6.51*
1937–2004 Model 3
2.70** (.872) 1.55* (.704) −.014 (.092) .024** (.007) −.003 (.004) −.184 (.609) −.001 (.001) .005* (.002) .029* (.013) .479 * (.189) .364** (.101) .304** (.108) .61 116 −8.10*
Note: *p < .05, **p < .01; standard errors in parentheses
2.86** (.954) .876 (1.76) .272 (.309) 3.53* (1.44) −.060 (1.04) −.107 (.101) .019* (.009) .001 (.006) .280 (.960) −.001 (.001) .003 (.003) .010 (.016) −.627 (.697) .072 (.158) .207 (.159) .57 66 −5.32*
78
CHAPTER 3
Model 3 substitutes the Presidential Distance variable as the measure of legislative-executive conflict. Whereas Divided Government and Opposition Seat Share measure divisions in party control of Congress and the White House, this measure is based upon differences in presidential and congressional ideology as measured by positions taken by members of Congress and the president on a wide range of issues that were subjected to recorded votes in Congress. It is again positive and statistically and substantively significant. As the ideological distance between the president and Congress increases, Congress becomes more prone to use private enforcement regimes. An increase of one standard deviation (.206) translates into an increase of 80 percent in predicted enactments. To put the magnitude in practical perspective, the increase in Presidential Distance moving from Jimmy Carter to Ronald Reagan elevated predicted enactments by 99 percent, which is, sensibly enough, quite comparable to the substantive effects of the divided government variable in model 1. The central hypothesis about the relationship between legislativeexecutive conflict in the American separation of powers system and enactment of private enforcement regimes is strongly borne out by the data: conflict between the legislative and executive branches causes Congress to rely more heavily upon the mobilization of private litigants for regulatory enforcement. These findings are robust across multiple operationalizations of interbranch conflict, whether one uses a simple divided government dummy, opposition seat share, or a party-neutral measure of the ideological distance between Congress and the president. Judicial Ideology. Judicial Distance and Judicial Direction are insignificant. Neither the direct measure of the judiciary’s distance from Congress, nor whether the judiciary is moving toward or away from Congress, explains the frequency with which Congress enacts private enforcement regimes. As discussed in chapter 2, there are theoretical grounds to expect greater proximity of the judiciary to Congress to increase enactment of private enforcement regimes, and to expect greater distance between the judiciary and Congress to increase enactment of private enforcement regimes, generating the conflicting predictions of the judicial friendliness and the judicial hostility hypotheses. It may be that these conflicting forces are actually operative but are canceling one another out in the data, or it may be that Congress is simply indifferent to the ideological position of the judiciary when it is deciding whether to enact private enforcement regimes. The statistical model cannot adjudicate between these alternative interpretations. The qualitative historical evidence examined in part II of the book, however, is capable in principle of rendering visible the operation of the conflicting hypotheses, and thus will provide an opportunity
EMPIRICAL MODEL OF ENACTMENT
79
to further examine the significance of judicial ideology to the legislative choice of private enforcement regimes. Coalition Drift. The Electoral Risk variable is significant and has the expected sign in all three models, supporting the coalition drift hypothesis, which predicts that the prospect of electoral losses increases Congress’s enactment of private enforcement regimes. A seat share loss of 12 percent for the majority party (about average in years that the majority lost seats in the next election, which was true in 72 of 118 years), is associated with the current majority increasing its enactments by 43 percent in model 1, 38 percent in model 2, and 53 percent in model 3. The findings suggest that legislators understand that private enforcement regimes can insulate not just from presidential subversion, but also, given the stickiness of the status quo in the American separation of powers system, from subversion by future legislative majorities. Issue Groups and Bureaucratic Drift. The issue group hypothesis is supported in all three models. Issue Witnesses is positive and significant, indicating that the presence of more witnesses representing issue-oriented citizens groups in hearings on regulatory legislation is associated with increased enactment of private enforcement regimes. An increase in Issue Witnesses by one standard deviation (15.35) is associated with an increase in predicted enactments by 47 percent in model 1, 44 percent in model 2, and 34 percent in model 3. These findings, of course, only tell us that the presence of issue groups in hearings is associated with enactment of private enforcement regimes, and not why this is so. The reason posited by the issue group hypothesis is that beginning in the middle to late 1960s such groups, growing in number, became more active as lobbyists and began to advocate for private enforcement regimes because of a growing fear of bureaucratic drift. They had come to regard bureaucrats as timid, apathetic, establishment oriented, and prone to capture, and they sought private litigation as an alternative. The qualitative historical evidence in part II of this book will provide an opportunity to investigate whether this account of interest groups’ motivations is accurate, and thus whether the statistical findings regarding issue group influence on enactment of private enforcement regimes can reasonably be read as support for the bureaucratic drift hypothesis. Rent-Seeking Lawyers. Lawyer Witnesses is highly insignificant and, contrary to the rent-seeking lawyer hypothesis, has a negative sign in all three models. The presence of lawyers’ associations in congressional hearings on regulatory legislation is not associated with enactment of private en-
80
CHAPTER 3
forcement regimes. While lawyers on both sides of disputes get paid, the rent-seeking lawyer hypothesis sometimes focuses on lawyers who represent plaintiffs. In an alternative specification, a variable was included that separately incorporated the number of witnesses on behalf of lawyers associations that exclusively represent plaintiffs’ lawyers. Such groups, it turns out, are very rarely invited to testify in hearings on federal regulation, representing only 5 percent of lawyers association witnesses. This variable, too, was insignificant. It should be observed, though, that since the hypothesis contemplates lawyers shaping the construction of the American state as a vehicle for their personal enrichment, it is plausible that they would prefer to do so outside the public view rather than in congressional hearings. That is, public hearings may be the wrong place to look for the influence of rent-seeking lawyers. This is not necessarily so, since lawyers associations could certainly make policy arguments in favor of court-centered implementation, whatever their true motives. However, it does counsel in favor of remaining open to the possibility that other forms of evidence, such as the more broad-ranging qualitative evidence examined in part II of this book, will tell a different story. Party Alignments. The Partisan Seat Share variable is significant with the expected sign in models 1 and 2, showing that, even controlling for how much Congress is regulating, higher levels of Democratic control of Congress are associated with more extensive enactment of private enforcement regimes. An increase of 15 percent in Democrats’ margin of control is associated with a 24 percent increase of predicted enactments in model 1, and 26 percent in model 2. Partisan Seat Share is insignificant in model 3, which has a substantially smaller sample size. If congressional ideology is incorporated in model 3 with the average of the House and Senate median NOMINATE scores, as an alternative to Partisan Seat Share, that variable too is insignificant. The data provide no support for the strong version of the party alignment hypothesis, which holds that creation of private enforcement regimes is uniquely the province of Democrats, and accordingly that the effects of other independent variables will be conditional upon Democratic control of Congress. To test this possibility, interaction terms were included that were created by multiplying Partisan Seat Share by each of the other independent variables. These interaction terms allow evaluating whether independent variables have the predicted effects on enactment of private enforcement regimes only or especially when Democrats control Congress. All of the interaction terms proved insignificant. Recall that the party alignment hypothesis has two strands of explanation regarding why, given the passage of a regulatory intervention, party would be associated with support for private enforcement regimes as an
EMPIRICAL MODEL OF ENACTMENT
81
implementation tool. The dominant strand emphasizes linkages between the Democratic party and the for-profit plaintiffs bar, and a secondary one emphasizes linkages between the Democratic party and nonprofit issue groups, including legal advocacy organizations. When the model’s support for the issue group hypothesis on the one hand, and its lack of support for the rent-seeking lawyer hypothesis on the other, are examined alongside the significant effects of Democratic control of Congress, the weight of the statistical evidence points toward the conclusion that Democratic party support for private enforcement regimes in federal regulation is primarily explained by its issue group constituency. Part II of this book will provide considerably more evidence on the role of party and interest groups in explaining enactment of private enforcement regimes. Budget Constraints. The budget constraint hypothesis is not supported by the data. The insignificance of Budget Constraint in all three models indicates that the relative size of budget deficits or surpluses is not significantly related to the enactment of private enforcement regimes. This finding does not rule out the possibility that shifting the costs of regulatory enforcement onto the private sector is regarded by some legislators as an important virtue of private enforcement regimes, or that this is an important explanation for the legislative production of private enforcement regimes. Rather, the data shows that if such cost-shifting motivations are at play, their effects in producing private enforcement regimes are not significantly correlated with actual budgetary conditions. If such cost-shifting motivations do comprise some durable facet of regulatory lawmaking in the United States, to such an extent that they do not vary with actual budgetary conditions, then there should be traces of this influence in evidence surrounding the multiple lawmaking episodes that produced private enforcement regimes scrutinized in part II of this book. The next three chapters consider the same set of hypotheses—treated above with statistical hypothesis testing—from the concrete historical perspective of actors constructing the statutory enforcement framework governing federal civil rights laws. The combination of statistical and qualitative historical evidence will provide a firmer basis from which to draw conclusions.
82
CHAPTER 3
APPENDIX
Collection of the Dependent Variable In order to identify all fee shifts and damages enhancements currently in existence, I ran word searches in the electronic Westlaw database of the United States Code. I devised the searches after examining numerous statutes with fee shifts and damages enhancements over the course of the twentieth century in order to identify appropriate language for the search. To identify fee shifts I searched “attorney! or legal or litigation or suit /10 fee! or cost! or expense!” To identify damages enhancements I searched “punitive or exemplary or double or ‘two times’ or ‘two fold’ or (twice /5 amount or sum) or (equal /5 amount or sum) or triple or treble or ‘three times’ or ‘three fold.’ ” All of the code sections returned from the searches were inspected to determine whether they were, in fact, plaintiffs’ fee shifts and damages enhancements. Only provisions that applied to court actions by private plaintiffs, and explicitly provided for recovery of attorneys’ fees and multiple or punitive damages were included in the dataset. With respect to fee shifts, in addition to provisions providing for an award of fees to a winning plaintiff, provisions providing for attorneys’ fees to a “prevailing party” or to “any party” were also included. While, read literally, these provisions could allow for fees to a defendant, courts have consistently adopted a “dual” (and asymmetric) interpretation of such provisions under which fees are generally awarded to prevailing plaintiffs as a matter of course, but are only awarded to prevailing defendants in the rare cases in which it is established that the plaintiff’s action was brought in bad faith, and was clearly frivolous or brought for purposes of harassment.57 Courts have concluded that this dual standard is grounded upon congressional intent. Given this legislative intent, and given that Congress legislates against this firm and long-standing interpretive background, it is appropriate to include such fee shift provisions, which asymmetrically benefit plaintiffs, in the analysis. In order to determine the enactment date, I examined a combination of Westlaw’s and Lexis’s annotated legislative history of each law to ascertain whether the provisions were part of the law as originally enacted, or were added subsequently by amendment, which was critical to assigning the correct date. In order to identify laws that had been enacted but exited the United States Code because they expired, were repealed, or were struck down by the Supreme Court, I ran the same searches in the Westlaw Statutes
EMPIRICAL MODEL OF ENACTMENT
83
at Large database from 1973 to the present, and identified and coded provisions that had exited the United States Code. No word-searchable electronic database of the Statutes at Large exists prior to 1973. Laws expired, repealed, or struck by the Supreme Court from the period 1887 to 1972 were identified by reviewing lists of laws enacted annually published in the Statutes at Large (1900–1944) and in Congressional Quarterly (1945–1972) to identify statutes in which enforcement issues might allow the use of private litigation. Those statutes not already in the dataset were then examined to ascertain whether they contained fee shifts or damages enhancements, and if so they were added to the dataset. Statistical Methods The dependent variable—the sum of all fee shifts and damages enhancements enacted annually—is a series of annual counts. Because the distribution of event counts is discrete, not continuous, and is limited to nonnegative values, it is best modeled assuming that the errors follow a Poisson rather than a normal distribution.58 Because the count data is time series, in order to check for autocorrelation I examined a plot of the autocorrelation function of the residuals of each model,59 which revealed second-order autocorrelation. In order to account for autocorrelation in count data, Schwartz et al.60 and Katsouyanni et al.61 have developed an autoregressive Poisson model that includes a specification of the autocorrelation in the model. Their model, like a negative binomial count model, also allows for overdispersion of the dependent variable, which is the case with the data analyzed here. Overdispersion is present where the variance exceeds the mean, which violates an assumption of a standard Poisson model. Their approach fits a log-linear model using iterative weighted least squares. Starting values for the parameters are obtained by running a standard Poisson regression, and the residuals are saved and incorporated into subsequent iterations of the model as explanatory variables for the number of lags specified,62 in this case two. Plots of the autocorrelation function of the residuals in each model confirmed that this method eliminated the autocorrelation problem. Applying the Dickey-Fuller and Augmented Dickey-Fuller unit root tests to the raw series reveals evidence of nonstationarity in the dependent variable and several of the independent variables (Partisan Seat Share, Presidential Distance, and Congressional Ideology). Statistical analysis of the relationship between two nonstationary time series can yield spurious results. All of the nonstationary series analyzed here are integrated of order one (that is, they are rendered stationary by taking first differences). However, running regressions on period-to-period changes
84
CHAPTER 3
in variables (first differences), rather than levels of variables, changes the theoretical meaning of the hypotheses being tested and leads to loss of information, particularly regarding long-run relationships.63 The long-run relationship between levels of nonstationary variables can only be estimated if they are integrated of the same order (as they are here), and the series are cointegrated. The core idea of cointegration is that even if series are nonstationary, there may be a long run comovement between the variables such that a linear combination of them is stationary. The principal diagnostic technique for determining whether nonstationary series are cointegrated was developed by Engle and Granger64 and entails assessing whether the model residuals are stationary using the Augmented Dickey-Fuller test, but evaluating the test statistic against critical values computed by Engle and Granger. This test, known as the Augmented Engle-Granger test, shows that all models reported in table 3.2 are stationary (the test statistic is reported as “AEG”), such that t-statistics are valid and spurious regression results on Partisan Seat Share, Presidential Distance, and Congressional Ideology are avoided.65 The coefficients of a count model are not directly interpretable. In order to transform them into interpretable form, an x-unit increase in an independent variable translates into a factor change in the rate of the dependent variable given by exp(xiβi). For example, the coefficient for Divided Government in model 1 is .655. The factor change in the expected rate of enactments for a one-unit change in Divided Government (a change from 0 to 1) is given by exponentiating ((1)(.655)), which equals 1.93. This means that when Divided Government goes from 0 to 1, holding other variables constant, the expected number of enactments increases by a factor 1.93. This is the equivalent of saying that the expected number of enactments increases by 93 percent.
PA RT I I
Private Enforcement Regimes and Civil Rights Introduction Over the past decade or so, the number of privately prosecuted job discrimination lawsuits filed in federal court averaged about 20,000 per year.1 The foundational decision to make private litigation the central vehicle of implementation for federal job discrimination laws was made in Title VII of the landmark Civil Rights Act (CRA) of 1964. As observed in the introductory chapter, this decision was neither natural nor inevitable. It was, rather, a radical departure from the dominant enforcement model used in state fair employment practice laws: an administrative commission with adjudicatory and cease-and-desist powers.2 Not one state of the twentyeight with fair employment practice laws in 1964 provided statutorily for private civil litigation for enforcement. Only a single United States territory—Puerto Rico—used the approach that Congress would ultimately follow in the CRA of 1964’s job discrimination provisions: private civil actions in court, with economic damages and attorney’s fee awards for winning plaintiffs.3 Why did Congress reject the dominant administrative enforcement model in 1964 in favor of a private enforcement regime for the CRA of 1964’s job discrimination title, and what consequences did this decision have? While part I of this book focused on understanding and explaining private enforcement regimes in general, part II will focus closely on the area of civil rights, taking up the questions just posed. The purpose of this shift from the general to the particular, and from the macro to the micro, is to shed light on dimensions of private enforcement regimes that can only be discerned by closely scrutinizing particular policy episodes and contexts. Turning from the statistical evidence in chapter 3 on the causes of the congressional choice across all policy areas to enact private enforcement regimes, chapters 4 to 6 sift the qualitative historical record to uncover the causes of Congress’s choice to rely centrally on private litigation in the implementation of important federal civil rights laws, starting with the foundational CRA of 1964. These three chapters examine the origins and development, between 1963 and 1991 (when Title VII was last
86
PART II
significantly amended), of this approach to implementation so heavily dependent upon private litigation. In chapter 5 we will see that the explanation leads well beyond the domain of job discrimination litigation and into that of civil rights litigation more broadly. Because chapters 4 to 6 really present a single, lengthy, detailed narrative tracing the origins and evolution of the federal statutory framework allowing and incentivizing private lawsuits for implementation of federal civil rights laws, this introduction to part II sets the stage for all three chapters by (1) explaining the motivations behind the methods employed, (2) setting forth reasons for selecting the case of job discrimination for study, and (3) reviewing prior explanations for the outcome investigated. Readers not concerned with issues of method or alternative hypotheses can skip directly to chapter 4. Process Tracing It is important to be explicit about the methodological logic and function of the historical material to be presented. The historical materials are approached in the tradition of “within-case analysis,” which searches for multiple forms of evidence about what caused the overall outcome in a particular case, and which typically is guided by the goal of interrogating more general social scientific theories and explanations. Variants of this method have gone under the names “process tracing,” “discerning,” “causal narratives,” and “analytic narratives,” among others.4 For present purposes, Alexander George and Timothy McKeown’s characterization of what they call “process tracing” is useful. They explain that this approach to within-case analysis of causation attempts “to uncover what stimuli the actors attend to; the decision process that makes use of these stimuli to arrive at decisions; the actual behavior that then occurs; the effect of various institutional arrangements on attention, processing, and behavior; and the effect of other variables of interest on attention, processing, and behavior.”5 This entails identifying actors who made key choices, and seeking to pin down their preferences, perceptions, and expectations, their understanding of policy alternatives, their reasons for rejecting paths not taken, their assessment of strategic opportunities, and the way in which their choices were shaped and constrained by the structural environment in which they acted.6 The goal of process tracing is thus to illuminate the microfoundations of individual behavior that translate initial conditions into observed outcomes. Process tracing can provide evidence relevant to the argument of this book that statistical analysis of data cannot. The theoretical arguments elaborated in chapter 2 entail a series of claims about the intentions and
ENFORCEMENT REGIMES AND CIVIL RIGHTS
87
actions of legislators and the interest groups they represent. This book has argued that legislators purposefully mobilize private plaintiffs and their attorneys to enforce the law; that they self-consciously deploy statutorily constructed economic incentives to do so; that this intentionality is embedded in a larger institutional framework in which legislators explicitly recognize private enforcement regimes as an alternative to administrative power; that the choice between private enforcement regimes and administrative state-building is shaped by ideological conflict among state actors and the groups they represent, mediated through the terrain of America’s fragmented state structures; and that legislative-executive polarization is a particularly important cause in the modern American state. The findings in chapter 3 provide important evidence in support of this intentional and causal account. They show, among other things, a strong correlation over time between Congress’s ideological distance from the president—which past research demonstrates is associated with the delegation of less power to the executive—and its enactment of private enforcement regimes. This aggregate-level data analysis, however, is not sufficient by itself to decisively establish the intentional and causal account of private enforcement regimes just outlined, for the observed relationship may be susceptible to alternative explanations. This is particularly true in light of the fact that polarization between the president and Congress opened up significantly at the end of the 1960s, and was largely sustained for the balance of the century, during which time Congress became much more likely to enact private enforcement regimes. One might reasonably wonder whether, for example, (1) it was simply a coincidence that policy activists learned about and became more cognizant of the potential efficacy of private enforcement regimes, leading to their increased use, at the same time that legislative-executive conflict was escalating; or (2) growing legislative-executive conflict and growing enactment of private enforcement regimes were both artifacts of some other durable change (perhaps in American political culture) that occurred at the end of the 1960s, but are not themselves causally linked; or (3) legislativeexecutive conflict in the late 1960s did cause congressional enactment of private enforcement regimes, and then this form of implementation became an entrenched, routine, boilerplate approach to regulating and subsequently was used by legislators in an unreflective manner, independent of the legislative-executive conflict that grew alongside it. The model in chapter 3 sought to incorporate independent variables into the model to account for all potential causes suggested by relevant literatures and by common sense, and statistical techniques were employed that are designed to guard against spurious statistical findings of a relationship be-
88
PART II
tween unrelated variables that merely happen to be drifting alongside one another over time. Nevertheless, the basis for causal inference could be substantially strengthened with historical evidence. Qualitative historical evidence is necessary to evaluate whether the intentional and causal account actually links the initial condition of a planned regulatory intervention to an observed outcome of private enforcement litigation as the method of implementation. If legislators’ and interest groups’ explicit and contemporaneous self-understanding of their reasons for supporting or opposing enactment of private enforcement regimes comport with the intentional and causal theory developed in chapter 2, and with the aggregate-level statistical findings reported in chapter 3, then the overall strength of causal inference will be enhanced. This will be especially so to the extent that the historical evidence shows a causal mechanism to be operative repeatedly, in multiple historical contexts over a long stretch of time. While the book considers an array of historical evidence, given that it is the choices of legislators that this book seeks to explain, it relies extensively on legislative historical evidence. After being introduced as a bill, an important legislative proposal, before passage, is typically subject to committee hearings in which both government officials and interest groups testify about its merits or deficiencies; is amended in committee; is referred to the full body for consideration with committee reports explaining, justifying, and sometimes attacking its provisions; is debated on the floor of each chamber; is often amended through recorded roll call votes; and may be further revised in conference committee if the two chambers passed different versions of a bill. Much can be learned about the sources of a law’s ultimate shape by tracing this journey. With respect to the interpretation of the historical evidence that will be presented, particularly from committee hearings and floor debates, a few words are in order about the meaning of rhetoric in the course of ideological clashes over civil rights policy. When engaged in public struggles over civil rights policy, legislators and interest group representatives now and again incline toward particularly vitriolic rhetoric, at times to express deeply held beliefs, and at times to strategically appeal to voters, donors, or other audiences. In presenting key themes, language, and arguments enunciated by players in the struggles over civil rights policy that are chronicled—such as civil rights advocates’ charges that Nixon and Reagan were enemies of civil rights who used government to subvert them—it is important to be clear about what the evidence is intended to show, and what it is not intended to show. It is presented as evidence that serious and sharp interbranch conflicts over civil rights policy were an important force driving the passage and
ENFORCEMENT REGIMES AND CIVIL RIGHTS
89
content of important civil rights laws. It is not meant as evidence of the actual merits of one side or the other’s positions, a debate that is beyond the scope of this book. Case Selection The case of federal job discrimination laws is a fertile site on which to investigate the origins and consequences of private enforcement regimes, satisfying numerous important considerations of case selection. As a case, it (1) has extreme values on the variable being studied; (2) is data-rich, so as to facilitate process tracing; (3) contains within it multiple events for study, distributed over time; (4) is of intrinsic importance; and (5) has been the subject of alternative hypotheses. After discussing each of these issues, I address the question of possible selection bias. Job discrimination laws present a case “extreme on pertinent measures,” which can be desirable for a case study because high values of the variable being studied will likely be accompanied by high values of the factors causing it, rendering them visible against the background of the case.7 Job discrimination litigation has been the most common type of suit in federal court next to prisoner petitions over the past decade or so,8 and it is commonly cited as an example of the high degree of litigiousness in the implementation of policy in the United States.9 Because it is a domain dominated by private enforcement litigation, it provides an opportunity to trace the causal processes that produced this outcome. Job discrimination laws are a data-rich case. While data-richness generally favors case selection because it allows the examination of more questions than data-thin cases, it is especially important to a research design that will evaluate theories using process tracing, as this book does, because that method demands extensive data.10 Although important legislative decisions are sometimes made outside the public view, leaving little or no historical record to trace, this is not so for federal job discrimination laws. In the legislative events examined, the role of private litigation versus administrative power in implementation was vigorously contested. These legislative conflicts produced an ample historical record, spanning nearly three decades, with which to investigate causal mechanisms driving key legislative choices. Job discrimination laws comprise an “embedded case,” having multiple units of analysis contained within it, as contrasted with a “holistic” case containing only a single unit for study. The legislative creation of the framework governing implementation of federal job discrimination laws principally occurred in four distinct legislative enactments (in 1964, 1972, 1976, and 1991), and in the course of studying these legislative
90
PART II
episodes we will encounter legislative decision-making in the domain of civil rights more broadly that also sheds light on the causes of private enforcement regimes. As contrasted with a holistic case constituted by only one unit of analysis (such as a single legislative enactment), an embedded case containing multiple units can enhance insights by providing more extensive opportunities for investigation.11 In a study covering a considerable stretch of time, as the present one does, the existence of multiple units within the case can provide “within-case” variation in important explanatory variables over time, which helps to guard against erroneous inferences based upon idiosyncratic relationships at a particular historical moment.12 Here, multiple units (legislative enactments affecting private enforcement) within the case distributed over time allow variation, for example, in interbranch conflict, partisan control of the different branches of government, and interest group mobilization. “Intrinsic human or historical importance” of cases, or “the magnitude of their human consequences,” is also weighed by social scientists in favor of studying them.13 Federal prohibition of employment discrimination in the private sector started with Title VII of the landmark Civil Rights Act of 1964, widely regarded as among the most significant legislative enactments of the twentieth century. Federal job discrimination laws now govern the workplaces of the overwhelming majority of public and private employees in the United States, commanding nondiscrimination on the basis of race, gender, national origin, religion, age, and disability. Over about the past decade these federal laws have generated an annual average of approximately 80,000 administrative charges of discrimination before the Equal Employment Opportunity Commission, and 20,000 private enforcement lawsuits in federal court.14 While it may be hyperbolic to suggest that job discrimination lawsuits are a “way of life in the United States,” as some have,15 civil rights regulation of the workplace most definitely is, and lawsuits are the dominant implementation vehicle. Even aside from the value of job discrimination litigation as a means to interrogate more general causal theories, the size and significance of this domain make it well worth understanding in itself. Finally, it is a virtue of a potential case when alternative theories make different predictions about it so that the case will allow evaluating the relative power of competing theories.16 In addition to investigating the effects of fragmented state structures on enactment of private enforcement regimes, this book is also interested in assessing evidence with respect to other causal accounts as well, four of which—the rent-seeking lawyer, issue group, party alignment, and budget constraint hypotheses—were set out in chapter 3 and incorporated in the empirical model there. As discussed in more detail shortly, the rent-seeking lawyer and budget constraint hypotheses have been asserted by scholars as explanations for
ENFORCEMENT REGIMES AND CIVIL RIGHTS
91
the legislative choice of private lawsuits to implement federal job discrimination laws. Thus, the case of job discrimination laws has the virtue of allowing assessment of rival theoretical accounts of the observed outcome—the dominance of private lawsuits as the implementation vehicle. Anytime a researcher seeks to draw general inferences from a case, the problem of selection bias must be considered, particularly when the researcher selects a case that possesses extreme values of the attribute studied. How do we know that regulating job discrimination is not uniquely prone, for example, to triggering legislative-executive conflict in a manner that leads to reliance on private lawsuits for enforcement, or that there is not something distinctive about regulating job discrimination that lends itself to regulation through private lawsuits? It bears emphasis, first, that we know that at the state level in 1964 the dominant outcome was an administrative cease-and-desist framework, and thus there is nothing in the nature of job discrimination that dictates private litigation for enforcement. Second, it is critical to keep in mind what role the process-tracing evidence is playing in this book’s research design. The empirical model in chapter 3 established that some explanatory variables are significantly associated with patterns of legislative enactment of private enforcement regimes, while others are not. The process tracing undertaken here is principally concerned with evaluating whether the hypothesized mechanisms linking explanatory variables to enactments are visible in the historical evidence. Because the larger population of cases is well understood through the cross-case analysis undertaken in the aggregate-level statistical model (constituted by many, many cases), and the process-tracing evidence from the civil rights case is being evaluated alongside that crosscase evidence, the potential problem of selection bias is substantially mitigated.17 Finally, with respect to the effects of the explanatory variable of central interest (legislative-executive conflict), the book’s conclusion preents corroborating evidence from two additional cases, one in the domain of labor policy and the other in environmental policy. Existing Accounts of Congress’s Choice to Incentivize Private Title VII Lawsuits Scholars attempting to explain why Congress relied so overwhelmingly on private litigation for implementation of Title VII have emphasized not just its private right of action, but also its provision allowing winning plaintiffs to recover attorney’s fees, which was clearly intended to facilitate, rather than merely permit, private enforcement litigation. As will be discussed in the next chapter, earlier civil rights laws allowing for private
92
PART II
enforcement and civil liability, but lacking fee recovery for winning plaintiffs, had gone severely underenforced despite widespread discrimination, and Title VII’s fee-recovery provision provided an important mechanism to actually generate substantial levels of private enforcement. One line of argument about why Congress included the fee-shifting provision in Title VII tracks the budget constraint hypothesis, and another follows the rent-seeking lawyer hypothesis, both of which were discussed in chapter 3, and neither of which was validated by the empirical model presented there. Of course, despite lack of validation in the empirical model spanning all policy areas and a wide stretch of time, one or both of these explanations could be important to the particular case of job discrimination litigation. On the account of events reflecting the budget constraint hypothesis, Congress’s inclusion of the fee-shifting provision expressed a “realiz[ation] that the federal government would be unable to handle all of the cases involving this type of discrimination,”18 and represented, according to Lee Epstein and Karen O’Connor, a “[r]ecognition of the fact that the resources of the federal government would be inadequate to enforce fully” Title VII’s promise of equality.19 The dominance of private litigation in Title VII implementation, then, resulted from Congress’s judgment that the national government lacked the resources to handle enforcement. As discussed in the last chapter, lack of state resources to enforce, or the desire to avoid raising revenue for this purpose, has been suggested as an explanation for legislators’ choice of private enforcement regimes across the whole range of policy domains.20 In contrast, Paul Frymer maintains that rent-seeking “trial lawyers” associations, which had “concentrated interests” in obtaining “particularized benefits” in the form of monetary gains from litigating employment discrimination claims, acted “under the radar,” scarcely noticed by civil rights opponents, to persuade legislators to adopt enforcement provisions using private litigation and attorney’s fees for winning plaintiffs.21 The legislators who crafted and passed the CRA of 1964 were primarily concerned with taking credit for popular and high-sounding civil rights policies, but not with the technical details of enforcement provisions. On Frymer’s account, Title VII’s private right of action, with fee shifting for winning plaintiffs, was not motivated by judgment about policy at all. Rather, “non-ideological factors”—the pursuit of money by lawyers— “explain why lawyers and courts became the chief enforcers of labor civil rights.”22 To the extent that there was any political calculation by legislators, the Democratic party, which controlled Congress and the presidency in 1964, preferred court-centered enforcement because it was a “quieter” and “politically safer” implementation vehicle, thus posing a lesser threat of rupturing a party coalition so deeply divided over race.23
ENFORCEMENT REGIMES AND CIVIL RIGHTS
93
As discussed in chapter 3, the contention that legislators, particularly congressional Democrats, write regulatory laws in a manner promoting private litigation in response to demand by rent-seeking lawyers has been broadly asserted across the entire range of policy domains.24 The next three chapters offer a fresh policy history that differs markedly from these accounts.
Chapter 4 FOUNDATIONS The Civil Rights Act of 1964
It is often forgotten that the famous March on Washington in August 1963, where Martin Luther King delivered his “I Have a Dream” speech, was billed as the “March on Washington for Jobs and Freedom.” Denial of equality in the workplace had been a fundamental concern of the civil rights movement that culminated in passage of the Civil Rights Act of 1964. Title VII of that act, barring job discrimination in the private sector based upon race, gender, national origin, or religion, was among the most radical socioeconomic interventions by the national government in American history. The question of how the job discrimination prohibitions would be implemented, and specifically the role of administrative power versus private lawsuits, was the subject of focused conflict as the bill that became Title VII of the CRA of 1964 worked its way through the legislative process. It is abundantly clear that those who did battle over the framing of Title VII’s enforcement provisions in 1963–64 and ultimately determined their content were goal-oriented, purposive, and strategic actors self-consciously attempting to maximize on policy objectives. The ultimate outcome was a compromise between conservative Republicans and liberal Democrats. In 1963–64, liberal Democrats and civil rights groups wanted an agency modeled on the National Labor Relations Board with adjudicatory and cease-and-desist powers for Title VII enforcement, without private litigation. They were skeptical about the capacity of private litigation, even with fee shifting, to overcome obstacles faced by private enforcers in the job discrimination context. Conservative Republicans, made pivotal to the CRA of 1964’s passage both by southern Democrats’ vehement opposition to civil rights legislation and by the filibuster in the Senate, authored Title VII’s private enforcement regime. They were natively circumspect toward bureaucratic regulation of their business constituents in general. Operating within constitutional separation of powers arrangements, they were also acutely fearful, with ample historical precedent, that they would not be able to control aggressive enforcement by an NLRB-style civil rights agency under the ideological influence of interventionist Democratic presidents. At the same time, given the explosive salience of civil rights in the elec-
CIVIL RIGHTS ACT OF 1964
95
toral environment of 1963–64, they had to offer a meaningful enforcement alternative to civil rights advocates or risk killing the bill and being blamed for its death. Private litigation was that alternative. Civil rights advocates insisted that if private civil rights enforcement was the best they could do, provisions aimed at overcoming economic obstacles to private enforcement, such as allowing winning plaintiffs to recover attorney’s fees, had to be included. Ultimately, Republicans wielded their veto powers to beat back bureaucratic state-building by liberal civil rights advocates, offering in its stead private litigation with attorney’s fees for winning plaintiffs. Institutionally speaking, Title VII’s private enforcement regime was a “common carrier”1 of pivotal Republicans’ desire to defeat bureaucratic state-building, and civil rights liberals’ desire to maximize enforcement if private litigation was the best they could do. Institutional and Political Strategic Context for Passage of Title VII Understanding the legislative outcome on fair employment practice enforcement in 1964 requires bearing in mind the institutional and electoral strategic context that produced it. In the two postwar decades leading to passage of the CRA of 1964, the most vociferous opponents of civil rights legislation, overwhelmingly southern Democrats, regularly exploited the veto points riddling Congress to kill civil rights bills that could have commanded a simple majority in a floor vote.2 The committee system in which seniority was paramount, coupled with the longevity of southern legislators in what was effectively a one-party regime in the region, guaranteed to Southerners gatekeeping control of key committees, which they used to quash civil rights bills time and again. The Senate Judiciary Committee, chaired from the Eighty-fourth to the Ninety-fifth Congresses by Mississippi Democrat and fervent civil rights foe James Eastland, was widely known as a “graveyard of civil rights legislation.”3 The most significant obstacle, however, was the Senate filibuster rule that allowed a bill to be defeated by thirty-four votes against ending debate. The southern Democratic anti–civil rights bloc needed a relatively modest number of conservative Republicans, with antiregulation or states rights leanings, to join them in order to defeat cloture, something they consistently achieved to repulse race legislation that would threaten Jim Crow. As an institutional terrain that civil rights legislation had to circumnavigate, Congress was a countermajoritarian minefield. In the years leading up to the battle over the CRA of 1964, however, both parties had electoral incentives to try to pass such legislation. Between the mid-1950s and the early 1960s, the historic, moving, and to some frightening achievements of the African American civil rights move-
96
CHAPTER 4
ment in Montgomery, Little Rock, Greensboro, and Birmingham, and the often brutal and sometimes deadly white southern reactions to them, caused dramatic change in the climate of opinion in the United States in favor of civil rights legislation. Civil rights were thereby forced onto the federal legislative agenda and became an important locus of party competition.4 Although the African American vote was solidly Democratic in the presidential elections of 1956 and 1960, it was not overwhelmingly so, as it would become and remain for decades after the bid for the presidency in 1964 by Goldwater, who had voted against the act.5 In view of their southern wing, Democrats were no more identified in the mind of the American voter as the party of African American civil rights than were Republicans during this period, and Republicans had underscored southern Democratic obstruction of civil rights legislation in an effort to make inroads among African American voters.6 Following introduction of the Kennedy administration’s omnibus civil rights bill in 1963, although northern Democrats were certainly its keenest advocates, a majority of both parties, with an uneasy gaze fixed on the 1964 elections, would want to claim credit for its passage. Perhaps even more, they were anxious to avoid having responsibility for its failure laid at their feet. In 1963–64, members of Congress could be broadly grouped into three types on civil rights issues.7 Northern liberals, predominantly but not exclusively Democrats, were ardent civil right advocates. Nearly all southern Democrats and the most staunchly conservative Republicans were devoted and predictable opponents. Members from the eleven former confederate states constituted about one-third of Democratic seats in both chambers (31 percent in the Senate and 36 percent in the House), and without these southern votes Democrats in both chambers lacked a simple majority.8 Between the poles of ardent supporters and devoted opponents, moderate and many conservative Republicans, largely from the Northeast and Midwest, were possible supporters of civil rights legislation, though there were definite limits to the level of intervention in the economy and encroachment on state prerogative that they would tolerate. A substantial share of these Republicans, neither certain supporters nor certain opponents of civil rights legislation, would be needed to pass the Kennedy administration’s bill, particularly in the Senate to break the filibuster. These Republicans were the pivotal voters in the legislative process that produced the Civil Rights Act of 1964.9 Conflicts over Public Civil Rights Enforcement in the Late 1950s Growing demand for civil rights legislation in the second half of the 1950s produced the Civil Rights Acts of 1957 and 1960, the first federal civil rights laws to be enacted since 1875. The 1957 act created the U.S.
CIVIL RIGHTS ACT OF 1964
97
Commission on Civil Rights to study violations of voting and other civil rights and to recommend remedial legislation; created the Civil Rights Division in the Justice Department; and empowered the attorney general to file civil actions and obtain injunctions against interference with the right to vote based upon race.10 The 1960 act empowered federal judges, upon finding a “pattern or practice” of voting discrimination, to appoint federal referees authorized to register voters, laid out standards for doing so, and required the maintenance of certain voting records, to which the attorney general would have access.11 Most salient to the present discussion, however, were provisions of both bills struck as a condition of passage in the Senate. The essential legislative conflicts in both 1957 and 1960 were focused overwhelmingly on the allocation of powers to enforce civil rights laws, not on substantive rights. Civil rights advocates in Congress proposed inclusion in both laws of provisions authorizing the attorney general to prosecute civil actions and obtain injunctions on behalf of aggrieved parties under all federal civil rights laws. This would have included, perhaps most prominently, the Civil Rights Act of 1871, which prohibits the deprivation of any federal right by a state actor, including the Fourteenth Amendment right to equal treatment without respect to race, in such domains as schooling, voting, employment, and public facilities of any kind. The prospect of across-the-board civil rights enforcement by national authorities provoked fierce opposition by Southerners in both 1957 and 1960. In order to overcome opposition and pass some civil rights legislation, the across-the-board federal enforcement powers were dropped in both years.12 The Reconstruction civil rights laws that would have been covered by the defeated federal enforcement authority already contained private rights of action, as did many state civil rights laws. However, civil rights proponents regarded federal enforcement authority as critical because private civil suits had proven an ineffective enforcement mechanism. They were, in practice, much too infrequent, despite widespread discrimination, to meet even minimal enforcement requirements.13 Civil rights advocates diagnosed this underenforcement as resulting from a number of sources, the most important of which were economic. Civil rights plaintiffs were typically poor and thus unable to pay the considerable legal costs of federal litigation.14 Under the standard “American rule” on legal fees, each side pays its own expenses, whether it wins or loses.15 Although counsel could take civil rights cases based upon a contingency agreement to keep a share of damages recovered, economic damages in most civil rights cases were both small and highly speculative.16 Consequently, Jack Greenberg, an NAACP lawyer who was soon to succeed Thurgood Marshall as director-counsel, explained in 1959 that “these cases involve . . .
98
CHAPTER 4
the expense of engaging counsel, whose fee is unlikely to be recompensed by the amount of damages awarded, even if they won.”17 Civil rights advocates well understood, according to Greenberg, that without “counsel fees for prevailing plaintiffs . . . virtually no lawyers could undertake difficult time consuming cases.”18 In economic terms, most civil rights claims were not just low value; they were negative value. In addition to this dire economic calculus, there were several other factors that led to the paucity of litigation under civil rights statutes providing private rights of action. Greenberg reported that in some parts of the South it was not possible for African Americans to retain counsel in civil rights cases even if they could pay. “The Southern white attorney who will handle a desegregation case is almost unique,” Greenberg wrote in 1959, and “many large Negro centers and even some important cities in the South still have no Negro lawyer or but a few.”19 A plaintiff who did sue could expect business and governmental defendants with bountiful legal resources to cause extensive delay and drive up costs through the use of elaborate opportunities for procedural gamesmanship in court proceedings.20 In the South, where enforcement was most needed, prospective plaintiffs had good reason to fear harassment, possibly violent, for filing suit,21 and to anticipate biased juries and judges if they had the courage to file.22 Private litigation was inherently limited as a civil rights enforcement mechanism, thought civil rights advocates, and thus it was absolutely imperative that national authorities be given strong enforcement powers. Developments on Title VII in the House: From Administrative to Judicial Power The job discrimination bill that ultimately became Title VII of the CRA of 1964, after the considerable amendment process traced below, was initially introduced in the House as H.R. 405 in January 1963 by Representative James Roosevelt of California, the son of Franklin Delano Roosevelt and a liberal Democrat with a record of devoted advocacy for strong job discrimination legislation.23 The bill was referred to the House Committee on Education and Labor, chaired by another vigorous civil rights advocate, Adam Clayton Powell (D-NY) of Harlem, and in July 1963 the committee reported the bill and recommended passage.24 Struggles over the bill reflect that by 1963, for legislators outside the South, the central field of battle on job discrimination legislation did not concern what rights should be enacted, but rather how and by whom they should be enforced. H.R. 405 framed a robust administrative enforcement machinery closely modeled on the NLRB. It proposed to create the Equal Employ-
CIVIL RIGHTS ACT OF 1964
99
ment Opportunity Commission, to be composed of a board and an Office of the Administrator. The board would be a quasi-judicial body, constituted by five members appointed by the president with confirmation by the Senate and serving staggered seven-year terms, with authority to hold administrative hearings on complaints and issue enforceable orders to cease-and-desist, reinstate, hire, and award back pay.25 The administrator, also appointed by the president with confirmation by the Senate and serving a four-year term, was empowered to conduct investigations, including the authority to demand and inspect records and compel testimony, and would prosecute cases before the board.26 Judicial review would be directly to the court of appeals on the record developed by the agency, and agency findings of fact were deemed conclusive provided they were supported by substantial evidence.27 The judicial review framework, governed by the Administrative Procedures Act, was highly deferential to the agency.28 Complaint processing within this framework would proceed as follows: The administrator would investigate all charges; if it was determined upon investigation that no “probable cause” existed to credit the charge, it would be dismissed. If, upon investigation, it was determined that probable cause did exist to credit the charge, the administrator would attempt voluntary conciliation, and failing that, would conduct an evidentiary hearing and present the hearing record together with a recommended decision to the board, which in turn would render a final decision.29 This initial formulation of what would become Title VII was like the National Labor Relations Act in one other respect: the administrative venue was exclusive, and there would be no private right to prosecute enforcement actions in court.30 H.R. 405, with its bureaucracy-centered enforcement framework—having no trial court proceeding and no private right of action—embodied the enforcement preferences of civil rights advocates among legislators and activists.31 Democrats had consistently advocated this administrative enforcement framework, with no private right to litigate, in job discrimination bills since 1944,32 and, as already noted, it was the dominant model at the state level.33 Civil rights advocates regarded the administrative cease-and-desist model as superior to court-based implementation for a number of reasons. Probably most important, it was relatively fast and cheap. As compared with procedurally complex, temporally protracted, and financially draining litigation, the administrative process—being more flexible and less rigidly encumbered by formal legal procedures—would be a more expeditious venue in which individual workers could enforce their rights, and would be less costly, with the investigation and prosecution conducted and paid for by the government. Whereas federal judges are generalists, an administrative approach would better leverage needed
100
CHAPTER 4
administrative expertise. Whereas courts passively wait for often fearful and vulnerable individuals to step forward and initiate civil rights suits, an administrative organ of government could proactively pursue a systematic enforcement program. As compared with a multitude of trial judges making civil rights policy in a piecemeal and decentralized manner in response to individual complaints, a centralized agency could create policy that was unified, coherent, and more predicable.34 Bureaucracy, civil rights advocates thought, would be better than lawsuits—far better. Of the CRA of 1964’s two radical new interventions in the private sector—employment (Title VII) and public accommodations (Title II)— civil rights advocates anticipated that the real enforcement battles would be fought out in the job discrimination context, and thus they were far more concerned with securing administrative enforcement of that title. Although their first preference for public accommodations enforcement was also administrative, they anticipated ready compliance with the public accommodations provisions, which they believed would actually serve business proprietors’ economic interests, and the violation of which would be quite simple and straightforward to prove. Thus, the judicial enforcement venue in the initial formulation of the public accommodations title in the bipartisan bill worked out between the Kennedy administration and Republican leadership was never contested by civil rights groups. However, civil rights advocates anticipated fierce resistance to Title VII in some sectors of labor markets, and they regarded the prospect of proving intentional discrimination, which could entail proving decision makers’ subjective state of mind, to be far more challenging. From civil rights groups’ perspective, Title VII presented a monumental enforcement challenge, and thus it was particularly important to secure an administrative enforcement framework.35 Seven Republicans on the Education and Labor Committee strenuously dissented in the report on H.R. 405. Of the seven, five articulated clear support for the substantive antidiscrimination prohibitions in the bill. However, all of the dissenters, in three separate opinions, trained their twofold attack on the enforcement provisions. First, they claimed that the proposed agency mirrored the excessive bureaucratic regulation of business that was characteristic of the NLRB, and of so much administrative state-building since the New Deal. It epitomized the “everincreasing encroachment on the part of administrative tribunals,”36 which resembled “the arbitrary fiat of king or baron,”37 and would “involve the federal government in the most intimate details of the operation of every business enterprise” in the nation.38 Second, they stressed that these agencies, whose leadership had been appointed by Democratic presidents in all but eight of the last thirtyone years, were ideologically slanted against business and susceptible to
CIVIL RIGHTS ACT OF 1964
101
the president’s political influence in ways that Congress could not control. On this point, the analogy of the proposed EEOC to the NLRB, on which it was so clearly modeled, was repeatedly invoked by Republicans as emblematic of the political mischief that could emanate from strong bureaucratic powers placed in the hands of overzealous administrators appointed by liberal interventionist presidents. Committee members Dave Martin (R-NE) and Paul Findley (R-IL) warned in their dissent from the committee report: “It is a major mistake to model legislation in this field on the National Labor Relations Board, which has one of the sorriest records of all the Federal agencies for political involvement.”39 This concern was echoed by fellow dissenters Peter Frelinghuysen (R-NJ) and Robert Griffin (R-MI), who also likened the proposed EEOC to the NLRB, complaining that “administrative tribunals . . . too often operate in an atmosphere of political and emotional pressures.”40 Roosevelt’s and Truman’s NLRB had vigorously pursued labor policies abhorred by the conservative coalition of Republicans and southern Democrats—policies that that coalition regarded as grounded more in executive ideology than legislative intent, and that produced perennial, intense, and memorable legislative–executive conflict over control of the agency.41 Similar executive politicization could surely be expected at Kennedy’s EEOC. In sum, the Republican dissenters—who would live to fight another day, as we shall see—opposed an NLRB-like enforcement regime for job discrimination because they disfavored bureaucratic regulation of business in general, and saw bureaucracy as susceptible to ideological bias and to political influence beyond their control, particularly by the proregulatory Democratic presidents who had dominated the executive branch for decades and would appoint EEOC leadership at least in the short run and possibly in the long run. The Republican dissenters also converged on what the enforcement scheme should be: they demanded the “historic safeguard of trial before an impartial judiciary.”42 Rejecting the NLRB-like administrative adjudication model, they called for a prosecutorial one. The agency should be stripped of the power to hold hearings and issue cease-and-desist orders, and instead be limited to conducting investigations and prosecuting actions in “impartial” federal trial courts, which would render fact-finding, liability, and remedy determinations. Republicans had supported the committee in reporting a fair employment practices bill in 1962 with such court-centered enforcement provisions and the administrator acting as prosecutor.43 They had long complained that the informality of administrative proceedings against business in ideologically biased New Deal agencies such as the NLRB, in the hands of overzealous and interventionist enforcers, resulted in relaxing the burden of proving violations, and in loosening the standard of what evidence could be introduced to establish
102
CHAPTER 4
liability, to the point of stacking the deck against business.44 Dissenters from the House report chided that the “NLRB-type administrative tribunal” that H.R. 405 proposed to create to enforce Title VII “ha[s] acquired a well-deserved reputation for ignoring the rules of evidence,” and it is one before which “the accused . . . must bear the burden of proving his freedom from guilt.”45 As compared to such proceedings, Republicans concerned with protecting business interests believed that it would be more difficult to establish the liability of defendant-employers in federal courts, which they regarded as more ideologically hospitable, and in which more extensive and formalized procedural rules, such as the plaintiff bearing the burden of proof by a preponderance of the evidence, and the constraints imposed by the federal rules of evidence, were institutionalized. A judicial venue had other potential benefits, from defendant-employers’ point of view, which were the flip side of reasons that civil rights groups wanted an administrative venue. The lawyer-dominated procedural formality and complexity of court would allow employers with deep pockets, and concomitant access to legal resources, to cause enforcement actions to be protracted and very costly, thereby discouraging them. Further, and perhaps most important, it was absolutely beyond question that the EEOC would be capable of adjudicating vastly more administrative complaints than it would be capable of prosecuting federal lawsuits, and thus far fewer claims would actually be adjudicated in a court-centered framework. At this stage, though, the Republican dissenters had lost the fight. The next move for proponents of H.R. 405 was to get it incorporated into the Kennedy administration’s civil rights bill (H.R. 7152), then being considered by Subcommittee No. 5 of the House Judiciary Committee, which held hearings on numerous comprehensive civil rights bills during the summer of 1963.46 In crafting H.R. 7152 the administration’s strategy was to collaborate closely with Republican leadership in order to produce a bipartisan bill that would attract sufficient Republican support to overcome the inevitable southern Democratic opposition, particularly the filibuster in the Senate. William McCulloch of Ohio, the ranking Republican on the House Judiciary Committee, insisted on two things as a condition of cooperation with the administration: that the administration publicly share credit with Republicans for a successful civil rights law, and that any bipartisan compromise bill passed by the House be capable of commanding a cloture vote in the Senate.47 With respect to the latter condition, House Republican leadership wanted to avoid pressing members from more conservative districts to cast risky votes for an overly ambitious bill that would subsequently be cut back in any event. They also wanted to avoid presenting conservative Republicans in the Senate, who would occupy the filibuster pivot, with a more sweeping bill
CIVIL RIGHTS ACT OF 1964
103
than they could stomach, thereby cornering their party brethren into the role of pruning it back and handing Democrats an electoral club with which to beat them in 1964.48 The bipartisan compromise reached—H.R. 7152—had no provisions prohibiting discrimination in private employment.49 The president’s view was that prohibiting discrimination in private employment was among the most controversial items on the civil rights agenda; including it would provoke such fierce opposition by Republican opponents of business regulation that it had no chance of survival and might kill any bill to which it was attached.50 However, leading civil rights groups and their liberal Democratic allies in Subcommittee No. 5 proved unwilling to acquiesce to the administration’s conservatism on this issue. They pressed energetically for incorporation of H.R. 405 as reported by the Education and Labor Committee, with its compulsory employment discrimination provisions governing private employment and its strong administrative enforcement machinery, into the administration’s civil rights bill. They succeeded, and along with a number of other important strengthening measures, Subcommittee No. 5 voted to incorporate H.R. 405 before referring the administration’s bill to the full Judiciary Committee.51 The Republican leadership with whom the administration had fashioned the bipartisan compromise bill cried foul and threatened that if reported by the full committee, this stronger version of H.R. 7152 would be cut to pieces via amendments on the House floor.52 The administration intervened aggressively and, after intense negotiations, managed to restore enough of the compromise embodied in the original H.R. 7152 to retain the support of Republican House leadership for a bill that liberal Democrats on the Judiciary Committee would also vote for.53 When the controversy blew up over the expansiveness of Subcommittee No. 5’s amendments to the compromise bill, the adoption of the strong employment discrimination provisions (H.R. 405) was arguably the single most controversial change.54 However, like the views of the Republican dissenters from the Education and Labor Committee report on H.R. 405, Republican grievances were focused on the proposed administrative enforcement machinery rather than on the substantive rights themselves. The compulsory job discrimination prohibitions governing the private sector were retained in the new compromise bill reported by the Judiciary Committee—as Title VII—but the provisions for enforcing the prohibitions were radically altered. These new Title VII enforcement provisions were drafted by Republicans, to meet Republican preferences, as a condition of Republican support.55 Robert Griffin (R-MI), Charles Goodell (R-NY), and Albert Quie (R-MN) were the chief architects of the new enforcement provisions.56 They precisely reflected the preferences of the Republican dissenters in
104
CHAPTER 4
the House Report on H.R. 405 (of whom Griffin had been one), outlined above, which were grounded in grave concern about an overzealous agency politicized through executive appointments. The enforcement regime was transformed from an administrative adjudicatory model to a prosecutorial model. Agency powers to hold hearings and issue orders to cease-and-desist, reinstate, hire, and for back pay were deleted. In their place was substituted agency authority to act as prosecutor in federal district courts. The new language provided that if two or more members of the five-person commission found, after investigation, reasonable cause to believe that discrimination had occurred, the commission “shall” (mandatorily) bring a civil action.57 It would be up to courts to find facts and elaborate the concrete meaning of Title VII’s broadly formulated prohibitions. Vote counts in the House indicated that there was majority support for giving the EEOC cease-and-desist powers.58 A number of observers close to the negotiations, including Leadership Conference general counsel Joseph Rauh, stated that the elimination of cease-and-desist powers was aimed at the Senate, where it was believed that they could not make it past the filibuster pivot, which, it was widely recognized, would be controlled by Senate minority leader Everett Dirksen (R-IL).59 This move was consistent with the House Republican leadership’s strategy that any bipartisan compromise bill agreed to in the House be capable of commanding a cloture vote in the Senate. Cease-and-desist powers did, however, face opposition by important conservative House Republicans as well. William McCulloch, ranking minority member of the House Judiciary Committee and a key Republican with whom the administration negotiated the version of H.R. 7152 that was reported, included in the committee report a statement of additional views on behalf of himself and six other committee Republicans that laid out their reasons for supporting the switch from administrative cease-and-desist to court-based implementation. Their reasons closely paralleled those of the dissenters in the earlier House Report on H.R. 405. Politicization of the agency by an ideologically distant executive branch was again a central concern. These Republicans, facing a Kennedy White House and writing at a time when six of the last eight presidential administrations were Democratic, were strongly disinclined to fashion new and potent administrative powers and deliver them into the hands of their Democratic opponents. They expressed anxiety about the likelihood of an ideologically liberal commission, to be appointed, they thought, by Kennedy, hell bent on an agenda of “forced racial balance” with “mathematical certainty” in employment. An “impartial” federal judiciary, with its more extensive and formalized procedural regimen, would be far more conducive to employers “estab-
CIVIL RIGHTS ACT OF 1964
105
lishing innocence,” and, as already noted, it would very likely lead to adjudication of many fewer claims than a commission that could investigate and adjudicate complaints at no cost to the claimant. According to the committee Republicans: As the title was originally worded, the Commission would have had authority to not only conduct investigations, but also institute hearing procedures and issue orders of a cease-and-desist nature. A substantial number of committee members, however, preferred that the ultimate determination of discrimination rest with the Federal judiciary. Through this requirement . . . we believe that the employer . . . will have a fairer forum to establish innocence since a trial de novo is required in district court proceedings together with the necessity of the Commission proving discrimination by a preponderance of the evidence. It must also be stressed that the Commission must confine its activities to correcting abuse, not promoting equality with mathematical certainty. . . . [T]he Commission will only jeopardize its continued existence if it seeks to impose forced racial balance upon employers.60 The suspicion that the Kennedy administration, and EEOC leadership to be appointed by Kennedy, would pursue an aggressive policy of racial preferences, if given the power to do so, was expressed more sharply in the minority report, which claimed that such practices were already being carried out in Kennedy’s Department of Labor.61 The proposed civil rights bill “vests . . . almost unlimited authority in the President and his appointees to do whatever they desire,” complained the dissenters, and this authority would be used to force an employer “to hire according to race, to ‘racially balance’ those who work for him in every job classification or be in violation of Federal law.”62 It is in the House Republican amendments, diminishing administrative power and enhancing judicial power, that the prospect of private enforcement litigation makes its first appearance in Title VII’s legislative history, though in an extremely qualified sense. The bill provided that if the EEOC failed or declined to bring a civil action, “[T]he person claiming to be aggrieved may, if one member of the Commission gives permission in writing, bring a civil action to obtain relief.”63 There was no private right to sue, for private suit was wholly dependent on administrative authorization. Further, this version of Title VII does not include a fee-shifting provision allowing a winning plaintiff to recover attorney’s fees, a tool legislators were certainly aware of because one was included in the public accommodations title in the same bill.64 Given that the only economic damages provided for under Title VII of the bill were back pay,65 the need
106
CHAPTER 4
to deduct attorney’s fees from that award would have rendered a rather modest, and in many cases negative, economic incentive for private enforcement, even in the event of administrative authorization to proceed. It bears emphasis that, although weakening and circumscribing the agency’s role, in the House-passed Title VII the agency was still given solid enforcement powers of a kind that had largely eluded civil rights proponents in the acts of 1957 and 1960, and the power to issue binding orders had not been eliminated from the enforcement regime, but rather had been transferred from the agency to courts. The House-passed bill created a mission agency, charged it with the initial investigation and evaluation of complaints, empowered it as prosecutor to set and execute a federal enforcement agenda, mandated that it prosecute meritorious claims, and allowed it to use, or decline to use, private enforcement litigation. Developments on Title VII in the Senate: Privatizing the Prosecutorial Function It was a foregone conclusion that there would be a southern Democratic filibuster in the Senate. It was also clear that nonsouthern Democrats would have to attract the support of at least twenty Republicans to secure cloture. Although there were approximately a dozen moderate or liberal Republicans who very likely could be recruited without much difficulty, there would be no cloture without a sizable share of the conservative Republicans.66 As has been widely observed in the historical literature on the legislative events leading to passage of the CRA of 1964, Everett Dirksen, minority leader and ranking Republican on the Judiciary Committee, would play a decisive role in securing, or not securing, the votes of the conservative Republicans needed for cloture. Although Dirksen faced a political climate in which appearing opposed to civil rights could cost Republicans in the upcoming election of 1964, his pivotal role gave him leverage to shape the legislation to suit his and his constituents’ preferences better than the House-passed version.67 Whereas the key move of House Republicans on the fair employment provisions had been to judicialize the enforcement forum, relying upon bureaucratic authority to execute the prosecutorial function, the key move of Republicans in the Senate, led by Dirksen, was to substantially privatize the prosecutorial function. They made private lawsuits the dominant mode of Title VII enforcement, creating an engine that would, in the years to come, produce levels of private enforcement litigation beyond their imagining. Although a number of important changes to Title VII occurred in the Senate,68 the discussion here is restricted to modifications concerning who would have the power to enforce the law and under what circumstances.
CIVIL RIGHTS ACT OF 1964
107
Dirksen was particularly influential among the Republican wing representing the strongly antiregulatory preferences of American business. As negotiations in the Senate over the House-passed H.R. 7152 were unfolding, Murray Kempton observed in the New Republic regarding this wing of the party: [L]ike Dirksen, they feel their kinship with those businessmen who dislike fair employment practice laws if only because they mean keeping even more records and entertaining even more federal agents. Dirksen is the senator most trusted to represent the American businessman in his distaste for any visitor from government. . . . There are no less than a dozen Republican Senators whose inclinations to sympathy, like Dirksen’s, are vulnerable to no appeal so much as the one which comes from a chamber of commerce.69 According to participants in the negotiations over the civil rights bill in the Senate, during their course Dirksen was in communication with representatives of the National Association of Manufacturers and the Chamber of Commerce, and was “very nearly the sole point of contact for the business lobbyists.”70 Business groups had been indecisive and, despite some serious misgivings about Title VII, had taken no position during the House proceedings. By the time the bill reached the Senate, they regarded a job discrimination title in some form as inevitable. When they did mobilize during the Senate proceedings to minimize Title VII’s encroachment on employer prerogatives, their main focus, like that of Republican leadership, was to curtail the enforcement provisions.71 The U.S. Chamber of Commerce wanted to weaken both the public and the private enforcement provisions in the House-passed bill. Their principal recommendations were these: (1) the EEOC should be limited to “conciliation and persuasion” and stripped of the right to sue; (2) whereas the House-passed bill allowed a private suit with the permission of one commission member, private suits should only be allowed with the permission of two commission members; (3) whereas the House-passed bill allowed a third party (like a civil rights group) to file a claim “on behalf of” an aggrieved individual, this provision should be stricken so that only aggrieved persons could file one; and (4) whereas the House-passed bill contained no statute of limitations for filing a claim, an aggrieved person should be limited to doing so within ninety days of the alleged discriminatory act (a very short limitations period compared to other federal laws).72 In laying out on the Senate floor what he regarded as deficiencies in the House-passed version of Title VII, Dirksen forcefully struck the same two linked notes that House Republicans had: it created excessive bureau-
108
CHAPTER 4
cratic regulation of business, and the overzealous liberals who would be chosen to wield the proposed new instruments of administrative power would abuse them. He exhorted that the EEOC’s proposed investigatory powers were boundless, affording no protection whatsoever “to an employer from fishing expeditions by investigators, in their zeal to enforce title VII.”73 He complained of “layering upon layer of enforcement,” which would be used by ideologically biased administrators to “draw and quarter the victim,” by which he meant American business.74 In pursuing amendments to Title VII’s enforcement provisions, Dirksen sought to minimize its “infringement on business freedom”75 and “harassment of businessmen,”76 particularly as it affected big business in the North, a critical Republican constituency.77 The compromise that was ultimately achieved, securing sufficient Republican votes for cloture and opening the way to passage of the landmark legislation, was principally negotiated off the Senate floor among Dirksen, the Kennedy Justice Department, and Democratic and Republican party leaders in Congress. The details and language of the compromise were mainly hammered out in the minority leader’s conference room by several conservative Republican lawyers on Dirksen’s Judiciary Committee staff, working together with Justice Department lawyers, including Nicholas Katzenbach, deputy attorney general; Burke Marshall, assistant attorney general heading the Civil Rights Division; David Filvaroff, special assistant in the Department of Justice; and Harold Green, chief of appeals and research, Civil Rights Division.78 John G. Stewart, legislative assistant to Minnesota’s Hubert Humphrey—Democratic floor manager in the Senate for the CRA of 1964— recounts that while the substantive employment discrimination prohibitions in Title VII, as passed by the House, were not seriously contested by Dirksen, how those prohibitions would be enforced was among the most difficult issues on which to achieve compromise in the entire bill.79 Dirksen was adamant from the outset, and ultimately proved impossible to move, in his insistence that the central burden of enforcing Title VII lay with private plaintiffs. In the first circulated draft of Dirksen’s proposed amendments to Title VII, “Dirksen proposed shifting the burden of enforcing the equal employment standards in Title VII from the . . . [EEOC] to the individual complainant,”80 eliminating the EEOC’s right to sue and reducing its role to nothing more than investigator and supervisor of voluntary conciliation efforts, leaving the national government with no enforcement authority whatsoever. The Kennedy Justice Department and Democratic leadership in Congress would not agree to elimination of all federal enforcement authority. After extensive negotiations, a compromise was reached in which Title VII would have a blend of public and private enforcement, with private
CIVIL RIGHTS ACT OF 1964
109
enforcement clearly predominating. Dirksen secured Democratic agreement to strip the EEOC of all enforcement powers. The Department of Justice, rather than the EEOC, would be allowed to prosecute Title VII claims, and it would be authorized, relative to the EEOC’s right to sue in the House-passed Title VII, to prosecute a much narrower range of cases. The attorney general could only initiate suits based upon a “pattern or practice” of discrimination. With this language, Dirksen sought to limit federal enforcement to circumstances in which regular, repeated, consistent, and officially sanctioned discrimination could be documented. Its purpose was to focus federal enforcement on employers in the South, where systematic policies of discrimination were widespread, and minimize it in the North, where employment discrimination was overwhelmingly de facto rather than de jure.81 Further, the House-passed version had stated mandatorily that the EEOC “shall” file suit upon a finding of discrimination that it could not conciliate, whereas the Dirksen substitute stated that the Justice Department “may” file suit, in its discretion, upon finding a pattern or practice of discrimination.82 It would be under no obligation to do so. Although the Justice Department could not initiate suits without evidence of a pattern or practice of discrimination, it was given authority, only at a court’s discretion, and only upon a finding that the case implicated issues of “general public importance,” to intervene if an individual had already elected to file suit.83 To the extent that it had the resources and commitment, the Justice Department could only set an affirmative and coordinated enforcement agenda in the pattern or practice context, and even in that limited context, it was not required to prosecute claims even if they were manifestly meritorious. The shift in enforcement authority from the EEOC to the Department of Justice was highly significant to conservative Republicans. Many administrative law scholars have suggested that “single-mission” agencies like the EEOC tend to have organizational norms that encourage zealous and single-minded pursuit of the agency’s mission.84 Whether or not this is true in general, it was perceived by conservative Republicans in 1963–64 to be true of the EEOC proposed by H.R. 405. According to Hugh Davis Graham, Dirksen well understood “that the Justice Department was a relatively small, elite cabinet agency . . . and so prided itself on enforcement through key case selection rather than through massive litigation, . . . pos[ing] a smaller threat of potential harassment to employers than would a new mission agency like the EEOC, which reminded Dirksen and his more conservative colleagues uncomfortably of its crusading early model: the NLRB.”85 Alfred Blumrosen, a job discrimination law expert who would become the EEOC’s first chief of conciliations in 1965, and who was a contemporaneous observer of the legislative proceedings
110
CHAPTER 4
in 1963–64, concurs with this view: “The conservative Republicans did not want another single-mission agency to put pressure on business.”86 Dirksen regarded the net effects of these concessions on executive enforcement powers by supporters of the Kennedy administration’s bill to be extremely significant. He trumpeted them to conservative Republicans whose support for his compromise bill he sought to win.87 These changes were also straightforwardly responsive to business groups’ demand that the commission’s enforcement powers be stripped. Whereas Republicans’ evisceration of the EEOC has been widely recognized, the private enforcement alternative proffered by the Dirksen compromise is either ignored or receives only passing mention, with its significance unrecognized in contemporary treatments of Title VII’s legislative history. This was the critical moment in which the foundation was laid for an approach to enforcing job discrimination laws dominated by private litigation—one that would, in time, produce more federal lawsuits than any other issue in the federal system, save for prisoners petitioning to be set free. What Dirksen took away in governmental enforcement power he gave back in private enforcement power. Under the amendments to Title VII introduced by Dirksen in the Senate, if the EEOC were unable to facilitate a voluntary settlement, the person claiming discrimination would have a right, regardless of the EEOC’s assessment of the merits of the case, and without the need for authorization by a commission member, to proceed with a private lawsuit in federal district court.88 The lawsuit would be de novo—meaning a completely fresh proceeding, as distinguished from appellate review—with no presumptions in favor of or deference to the agency’s determination. Further, the court was granted authority to waive filing fees and appoint an attorney for plaintiffs, and to award winning plaintiffs attorney’s fees and other costs of the suit to be paid by defendants.89 The amendments to Title VII’s enforcement regime introduced by Dirksen were ultimately adopted, and their exact language was retained in the law as passed.90 It bears emphasis that these provisions regarding appointment of counsel, filing fees, and attorney’s fees are departures from the default rules in the American legal system that would be controlling if Congress did not specifically override them. In federal civil actions, absent an explicit statutory directive, counsel is appointed only in “exceptional circumstances,” which in practice has meant almost never.91 Likewise, unless Congress specifies in a statute that a winning plaintiff can recover attorney’s fees from the defendant, the “American rule” applies, and each side is generally responsible for its own attorney’s fees, regardless of the outcome of the case. When Congress has wished to proactively facilitate private enforcement of statutes, it has developed a practice of including a provision expressly overriding the American rule and allowing winning plaintiffs to
CIVIL RIGHTS ACT OF 1964
111
recover attorney’s fees.92 This practice was not new in 1964. It was inaugurated in federal statutory law with the Civil Rights Act 1870 and was used in landmark regulatory laws in the Progressive and the New Deal eras.93 Economic damages under Title VII were limited to back pay, which could be quite insubstantial and easily less than the costs of prosecuting an action, especially for workers in low-wage labor pools.94 Accordingly, the appointment of counsel and attorney’s fee provisions would facilitate the litigation of claims that, under default rules, would have produced a net loss, even to winning plaintiffs. Why were these provisions included in the amendments introduced by Dirksen, who, it seems, would not have wanted to foment private lawsuits against his business constituents, who were pushing amendments to limit such suits? The answer is that civil rights advocates were concerned about lack of enforcement through private lawsuits in court, and they advocated provisions to encourage it. Earlier in the legislative process, the NAACP, the ACLU, and the National Lawyers Guild had successfully urged the inclusion of a fee-shifting provision in the public accommodations title of the Kennedy administration’s original bill.95 According to the NAACP Legal Defense Fund (LDF) director-counsel Jack Greenberg, whereas the first preference of civil rights groups was for strong administrative enforcement of civil rights, if private litigation was the best they could do, “[W]e always supported counsel fees for prevailing plaintiffs as the only way to make private enforcement feasible. Otherwise, virtually no lawyers could undertake difficult time-consuming cases.”96 Alfred Blumrosen, a pioneering expert on job discrimination legislation and a keen contemporaneous observer of the legislative process that produced Title VII in 1963–64, recounts that when the right to sue in ordinary individual claims was shifted from the EEOC to private plaintiffs in the Senate, civil rights advocates “argued that the private ‘right to sue’ would be hollow because impecunious plaintiffs could not obtain counsel, and, if they did, counsel fees would eat up the limited monetary recovery available under the act,” and thus they successfully pushed for the fee-shifting provision.97 Indeed, recall that it was the meagerness of private enforcement of Reconstruction and state civil rights laws with private rights of action, but without provisions providing for recovery of attorney’s fees or appointment of counsel, that had motivated the urgent but unsuccessful push by civil rights advocates in 1957 and 1960 to empower the attorney general to bring civil actions to enforce those laws. The potential underenforcement problem with Title VII was even greater than under Reconstruction civil rights laws, because those laws allowed recovery of more extensive economic damages than permitted under Title VII’s back-pay-only rule.98 Civil rights advocates clearly recognized major impediments to private
112
CHAPTER 4
civil rights enforcement. They worried that many potential plaintiffs were poor and thus unable to pay the costs of litigation; the costs of prosecution often exceeded small and uncertain damages in civil rights cases, so attorneys would not litigate them on a contingency basis; and in many parts of the South there were no attorneys willing to take civil rights cases.99 According to David Filvaroff, a Justice Department lawyer who participated in negotiating and drafting the details of the Dirksen compromise on behalf of the department, the Civil Rights Division’s leadership “thought the fee provision was very important.”100 Justice Department lawyers negotiating with Dirksen and his staff were regularly apprised of civil rights groups’ views on the issues under consideration, and they understood that fee recovery could provide a very important benefit to would-be plaintiffs lacking the resources to pay an attorney.101 High-level Justice Department lawyers like Burke Marshall and Nicholas Katzenbach, and Democratic leaders such as Hubert Humphrey, also wanted any compromise bill to retain the support of civil rights groups. They were acutely sensitive to the prospect of being accused in a press conference by NAACP lead lobbyist Clarence Mitchell, or Leadership Conference general counsel Joseph Rauh, of having caved in to pressures from Dirksen to gut the bill.102 If they were to trade public enforcement for private enforcement to secure Dirksen’s help in overcoming the southern Democratic filibuster, then the private enforcement framework had to be made meaningful. Democratic Senate floor manager Hubert Humphrey, advocating the fee-shifting and appointment-of-counsel provisions, was the only legislator to explain in the record why they were included in the law. Humphrey was well positioned to know. Of Democratic Party leaders who negotiated the compromise with Dirksen, he was among those in most regular communication with both Dirksen and civil rights groups as the process developed.103 For a decade and a half Humphrey had been a primary proponent of civil rights legislation among members of Congress and had worked closely with civil rights groups in crafting civil rights bills and pressing them in the legislative process.104 As a result, he had strong and long-standing relationships with Mitchell and Rauh, and he worked closely with both, meeting with them frequently as the legislative battle transpired.105 According to Humphrey, the Title VII amendments introduced by Dirksen regarding appointment of counsel and waiver of the filing fee were included because “it is recognized that the maintenance of a suit may impose a great burden on a poor individual complainant”;106 the attorney’s fee provision “should make it easier for a plaintiff of limited
CIVIL RIGHTS ACT OF 1964
113
means to bring a meritorious suit”;107 and the appointment of counsel provision was also included “because Negroes in some areas are unable to obtain legal representation to institute civil rights suits.”108 These explanations given by Humphrey were straightforwardly responsive to the reasons previously identified by civil rights leaders—important Humphrey constituents, with whom he was in close contact—for the dearth of private civil actions, despite widespread discrimination, under Reconstruction and state civil rights laws. Dirksen, in negotiating agreement to a predominantly private enforcement framework as an alternative to strong governmental enforcement powers—thereby stripping the law of all federal authority to initiate enforcement in ordinary individual discrimination claims—agreed to address the well-known impediments to private civil rights enforcement and thus included the attorney’s fee and appointment-of-counsel provisions desired by civil rights advocates. Dirksen was not oblivious to the possible problem of private lawsuits for his business constituents. It was a concern that, as discussed above, they had raised with him directly. He included several provisions in his amendments that appeared calculated to limit the use of the private right of action that he created, changes that were responsive to the preferences of business groups. First, as requested by the Chamber of Commerce, Dirksen’s amendments specified that only aggrieved claimants, and not third parties such as the NAACP—the most prominent civil rights litigator in the United States—would have standing to initiate claims.109 His goal with this provision was to contain the volume of litigation.110 The amendment overrode language in the House-passed version of Title VII, which had allowed groups to initiate claims “on behalf of” an employee.111 However, this provision ultimately had no bearing upon the ability of a third party, such as the NAACP, to serve as counsel to an individual claimant, which, as discussed in the next chapter, would be a frequent occurrence in Title VII’s early years. Second, Dirksen’s amendments stipulated that, in states with fair employment practice laws, before claimants could initiate charges with the EEOC, they would be required to exhaust state remedies, though claimants would generally be free to exit state systems for the federal process sixty days after initiating a state claim.112 The House-passed version had contained no such deferral to state procedures.113 Although this provision would delay plaintiffs in states with fair employment practice laws in pursuing their federal claims, it was only mandatory for two months, and once a claimant initiated the federal process, there would be no deference owed to state administrative determinations. Third, as requested by the Chamber of Commerce, Dirksen’s amendments added a remarkably short statute of limitations: victims of discrimination would have only
114
CHAPTER 4
ninety days from the discriminatory act to file federal claims, or 210 days if they resided in states with fair employment practice procedures that had to be exhausted, after which the claim would expire.114 The Housepassed version had contained no limitations period.115 That public enforcement powers had been traded for private litigation in the Dirksen compromise, and that the details of Title VII’s private enforcement regime were a critical part of the bargain, was well understood by observers and legislators. One contemporaneous commentator explained that “[t]he system of individual enforcement was the result of a conscious, explicit rejection of a system of administrative enforcement.”116 Another characterized the provisions regarding appointment of counsel and attorney’s fees as “the price that had to be paid to secure bipartisan agreement on striking out the power of the EEOC to enforce Title VII by court action.”117 Humphrey characterized the transfer of authority to bring suit from the commission to the individual claimant, and the allowance of appointment of counsel and recovery of attorney’s fees, as “the most significant change . . . in Title VII” to occur in the Senate,118 a sentiment echoed by Leverett Saltonstall (R-MA).119 Voting Alignments and Roll Calls on Private Enforcement There were roll calls on numerous dimensions of the CRA of 1964’s private enforcement provisions, demonstrating that legislators regarded the details of these provisions, such as rules governing fee shifting and appointment of counsel, to be extremely important. Such details were the focus of intense scrutiny, deliberation, and conflict among lawmakers. Saltonstall remarked, concerning the task of nailing down the details of the private right of action in Title VII in the ultimate compromise, “We struggled over this section for a long time.”120 The public accommodations title in the original Senate version of the bill (S. 1732), considered in the Commerce Committee, had contained an attorney fee-shifting provision, which, according to Strom Thurmond, the committee deleted “after much careful deliberation.”121 When faced with the same fee-shifting provision in the House-passed version of the public accommodations title that made it to the Senate floor (Title II), Thurmond offered a roll call to strike it out, urging his colleagues to uphold the earlier deal made in the Commerce Committee, but they rejected his pleas and defeated the amendment.122 The next day, undeterred by this defeat, Senator Sam Ervin (D-NC) offered the same amendment to strike out the fee-shifting provision, warning that it would unleash a stampede of “ambulance chasing” lawyers “chosen and controlled by the NAACP and its subsidiary corporations.”123 Jack
CIVIL RIGHTS ACT OF 1964
115
Miller (R-IA) countered that since plaintiffs could recover fees only if they prevailed, “the ambulance chaser—if we wish to use that term—will be on his guard not to go after anything except meritorious cases.” Ervin shot back that “the attorney who had a meritorious case would probably be crushed to death in the rush of attorneys seeking cases regardless of their merits.”124 In support of Ervin’s amendment, Senator Bourke Hickenlooper of Iowa, a conservative Republican opponent of the bill, in perhaps the most ominous admonition ever uttered about a fee-shifting provision, offered biblical counsel: “He that passeth by, and meddleth with strife belonging not to him, is like one that taketh a dog by the ears.” Humphrey replied, “That is a partisan remark.”125 The roll call was defeated and the fee-shifting provision was retained.126 In addition to these two roll calls, Southerners in the Senate offered an additional six aimed at diminishing private enforcement under Title II or VII or both: the Thurmond amendment to delete any private right of action under Title VII, substituting criminal prosecutions with attendant proof beyond a reasonable doubt and trial by jury;127 the Thurmond amendment to delete court authority to appoint counsel under Title VII;128 the Ervin amendment providing that an attorney would have to consent to be appointed under Titles II and VII;129 the Thurmond amendment to delete any private right to sue under Title II, substituting criminal prosecutions;130 the Long amendment to stipulate that an individual could recover attorney’s fees under Title II only on a “showing of good cause”;131 and the Thurmond amendment requiring exhaustion of all administrative remedies before commencing private litigation under Title II.132 Southerners offered two more amendments aimed at giving the very same benefits to civil rights defendants that they sought to take away from civil rights plaintiffs: the Stennis amendment to authorize the court to appoint counsel and award attorney’s fees for a defendant sued by the attorney general under any part of the act, including Title VII;133 and the Thurmond amendment to allow the court to appoint an attorney for defendants under Title II.134 All eight roll calls to limit private enforcement were defeated, as were the two to aid civil rights defendants. Legislators from the eleven former Confederate states voted in favor of restricting enforcement by private plaintiffs at a rate of 91 percent, whereas Democrats outside the South voted against restricting enforcement by private plaintiffs at a rate of 98 percent, and Republicans outside the South (all but one) voted against at a rate of 79 percent. Bivariate logistic regression indicates that the odds of voting in support of private enforcement incentives were ninety-two times larger among legislators who voted in favor of the act as compared to legislators who voted against it. These roll calls show that legislators
116
CHAPTER 4
were abundantly aware of the details of Title VII’s private enforcement provisions. Lawmakers opposed to the law sought to minimize their litigant-mobilizing effect, and nonsouthern Democrats and Republicans who supported the law voted together to sustain the details of the private enforcement regime—a cornerstone of the Dirksen compromise—against southern attack. Interest Groups on the Scene A brief mention is in order of the interest group presence in the halls of Congress during the legislative struggle over the CRA of 1964. It provides an interesting point of contrast with later lawmaking episodes addressing civil rights enforcement, taken up in the next two chapters. A massive interest group effort lay behind the watershed legislation. A Congressional Quarterly Weekly Report article titled “Intensive Lobbying Marked House Civil Rights Debate” observed that the CRA of 1964 “was the subject of some of the most intensive and effective behind-thescenes lobbying in modern legislative history,” and provided an inventory of all organizations that lobbied for the bill.135 The forty-four organizations listed as lobbying in favor of the act were predominantly civil rights groups, labor organizations, and church groups. Only a single lawyers’ association—the American Bar Association—is listed, though it must be acknowledged that the ABA was surely the most powerful lawyers’ association in the United States. The nongovernmental witnesses to appear in the hearings before Subcommittee No. 5 were likewise dominated by civil rights and other issue groups, labor organizations, and religious groups. Of forty-one nongovernmental witnesses invited to testify, none represented lawyers’ associations (the ABA was not invited), and only two were from specifically legal advocacy organizations: the American Civil Liberties Union and the Lawyers’ Committee for Civil Rights Under Law.136 No representatives of lawyers’ associations were invited to testify in the House Committee on Education and Labor hearings on H.R. 405.137 Lawyers’ associations representing the for-profit bar were simply not on the scene. This would soon change. Private Enforcement as Compromise The Dirksen compromise on Title VII enforcement was exactly that. Although Southerners and the most conservative Republicans worried about “ambulance-chasing” lawyers “chosen and controlled by the NAACP,” devoted civil rights advocates were also disappointed by the settlement.
CIVIL RIGHTS ACT OF 1964
117
They denounced the evisceration of the EEOC and regarded the civil action provisions as clearly weaker than the administrative cease-anddesist powers they actually thought necessary, worrying that potential civil rights plaintiffs, often poor and vulnerable to retaliation, would not be in a position to take on American business in the courtroom, particularly in the South.138 Richard Berg, who worked in the Office of Legal Counsel in the Kennedy Justice Department when the deal was struck and then became deputy general counsel of the EEOC shortly afterward, expressed liberals’ low expectations of the effects of the law after it was passed, but before it came into effect: The enforcement procedures of the title . . . bear only too visibly the marks of compromise, and seem to me to contain serious deficiencies. It seems questionable that much can be accomplished through suits in federal court by persons aggrieved by acts of discrimination. The practical advantages will lie heavily with the defendants, and even where the evidence of discrimination is overwhelming, it cannot be expected that many complainants will undertake the burden of an individual suit.139 The Leadership Conference concluded, no doubt in exaggerated rhetoric, that the changes in Title VII’s enforcement provisions in the Dirksen compromise rendered it “virtually meaningless.”140 Writing in 1966, law professor and civil rights advocate Michael Sovern agreed: “Letting the complainant sue was one of the original models of anti-discrimination enforcement . . . and it has never worked.”141 Between southern fears and liberal disappointment, it is important to stress that the enforcement provisions of Title VII were meaningful and important. They embodied a compromise on enforcement, not an abdication. With respect to Title VII, the attorney general received some, though not all, of the type of enforcement powers long sought by civil rights groups and defeated in 1957 and 1960. However, the lion’s share of enforcement responsibility would lie with private litigants. In light of the well-recognized scarcity of private enforcement of existing civil rights laws with private rights of action, the compromise, to succeed, had to go beyond merely allowing aggrieved individuals to sue. It had to overcome a key obstacle to private enforcement—the difficulty of retaining and paying counsel—and thereby encourage meritorious actions. The compromise sought to address this issue and facilitate litigation, as compared with existing civil rights laws that merely allowed it, with its provisions concerning appointment of counsel and attorney’s fee awards. To be sure, the effects of this compromise were uncertain.
118
CHAPTER 4
Pivotal conservative Republicans, empowered by a profoundly divided Democratic Party and the filibuster in the Senate, authored the compromise. Instinctively wary of bureaucratic regulation of business, and deeply apprehensive, based upon long experience, that an NLRB-style civil rights agency in the hands of their political adversaries would be extremely difficult to hold sway over, they would not countenance H.R. 405’s EEOC. However, facing impending elections in which civil rights would be a central issue, Republicans had to propose an alternative implementation approach. Attempting to steer a middle course, House Republicans judicialized the enforcement venue, striking Title VII’s ceaseand-desist powers, substituting an EEOC right to sue in federal court, and introducing for the first time the prospect of private enforcement suits, though only with the agency’s permission and without any effort to surmount long-standing and well-recognized economic obstacles to private civil rights litigation. Conservative Republicans in the Senate then largely privatized the prosecutorial function. They stripped the EEOC of the obligation to sue in meritorious cases in which conciliation failed, instead conferring upon the Justice Department, a weaker organ of administrative power, the discretionary right to initiate suit, but even then only in the limited pattern-or-practice context. There would be no governmental authority to initiate enforcement of ordinary individual claims under Title VII. In exchange for these reductions in executive enforcement power, the Dirksen compromise created an unconditional private right of action under Title VII, including the appointment of counsel and attorney’s fee provisions, which were clearly calculated to address the economic impediments already inhibiting private enforcement of Reconstruction and state civil rights laws. In sum, conservative Republicans, whose support for civil rights legislation could not be expected unconditionally, exercised their pivotal legislative powers to derail liberal efforts at bureaucratic state-building, imposing private litigation as an alternative instrument of regulation. This legislative move would have profound long-run consequences for the implementation of job discrimination prohibitions under Title VII and beyond. As new forms of job discrimination were prohibited under federal law, Congress, not surprisingly, simply inserted the new prohibitions into the existing enforcement model founded upon private litigation. The Age Discrimination in Employment Act of 1967 followed the enforcement model of private litigation with no administrative ceaseand-desist powers.142 The Rehabilitation Act of 1973,143 prohibiting employment discrimination based upon disability by the federal government and federal contractors, and the Americans with Disabilities Act of 1990,144 prohibiting employment discrimination based upon disability in the private sector and by states, simply incorporated Title VII’s enforce-
CIVIL RIGHTS ACT OF 1964
119
ment provisions by reference. Thirty years on, this job discrimination enforcement framework would produce more nonprisoner civil litigation in the federal system than any other. A Note on the Fair Housing Act: Replaying the Story of Title VII The story of the Civil Rights Act of 1968, also known as the Fair Housing Act (FHA) of 1968, is remarkably similar to that of Title VII. It provides powerful confirmation of the core dynamics that drove the dominance of private litigation in the Title VII enforcement compromise: pivotal conservative Republicans, empowered by a divided Democratic Party and supermajoritarian legislative institutions, imposed private litigation, with mobilizing incentives such as fee shifting and appointment of counsel, in order to defeat bureaucratic state-building by liberal civil rights advocates. It is also significant because it demonstrates the contagiousness of the settlement on civil rights enforcement authored by Dirksen in his amendments to Title VII in 1964, a subject taken up extensively in the next chapter. The FHA prohibited discrimination in the sale or rental of residential housing on the basis of race, color, national origin, religion, or sex, and was introduced in 1967 in the Senate as S. 1358 by Walter Mondale (D-MN).145 It delegated authority to the secretary of the Department of Housing and Urban Development to investigate complaints, and upon a finding of reasonable cause, to attempt voluntary conciliation, and failing that, to hold hearings and issue cease-and-desist orders when appropriate. If the secretary dismissed the complaint, without an evidentiary hearing, based upon a finding of no reasonable cause to believe that discrimination had occurred, or if she failed to render a reasonable cause determination within thirty days, the complainant would be free to proceed with a private lawsuit.146 The bill did not specify what damages were available, nor did it provide for appointment of counsel, waiver of filing fees, or awards of attorney’s fees for winning plaintiffs. Therefore, plaintiffs were required to bear all the costs of federal court litigation when economic recoveries would be highly speculative, if available at all. Mondale’s proposed FHA, with strong administrative enforcement and a weak private enforcement regime, was energetically supported by the NAACP and the Leadership Conference, whose lead lobbyists, Clarence Mitchell and Joseph Rauh, worked hard to marshal support.147 It was energetically opposed by Southerners, who mounted a filibuster in the Senate. They were joined by conservative Republicans, including Dirksen, in defeating cloture by a vote of fifty-five in favor and thirty-seven against, with most Republicans voting for cloture.148 Though Mondale’s FHA, with its administrative cease-and-desist authority, commanded a solid majority, the bill’s proponents recognized that they lacked the two-
120
CHAPTER 4
thirds necessary to end debate, and they were prepared to make concessions, including significant reductions in the scope of the housing market covered—possibly even the exclusion of single-family dwellings.149 As in 1964, minority leader Dirksen spearheaded off-the-floor negotiations over compromise provisions that could garner enough conservative Republican support to overcome the filibuster.150 A Washington Post editorial titled “Bow to Minority Rule” stated that it would be “disastrous to the Republican cause to defeat the civil rights bill,” which commanded majority support, and that Dirksen was acting as “chief architect” of a compromise because he was “convinced that he must strive for the enactment of a civil rights bill of some kind for the sake of improving the Republican image in the election next November.”151 President Nixon, too, urged that passage of the Fair Housing Act would serve Republican interests in the upcoming election.152 “And so we labor together precisely as we did in 1964,” Dirksen remarked when introducing his substitute amendment, “because I am in almost the identical position.”153 As in 1964, Dirksen’s changes focused almost entirely upon the enforcement provisions, seeking to restrict the substantive coverage of the law to a considerably lesser degree than the liberals anticipated and were prepared to concede.154 Dirksen’s changes in the enforcement provisions mirrored the 1964 compromise: public enforcement authority was traded for private litigation. HUD’s adjudicatory and cease-and-desist powers were eliminated, and the secretary, like the EEOC, was relegated to supervising voluntary conciliation efforts.155 To compensate for the elimination of these administrative enforcement powers, the Dirksen substitute fortified the private enforcement regime relative to Mondale’s bill. Dirksen proposed to allow winning plaintiffs to recover any economic damages suffered, plus up to $1,000 in punitive damages. He further proposed that, as with the CRA of 1964, the FHA allow the court to appoint counsel, waive the filing fee, and award attorney’s fees to winning plaintiffs.156 As in 1964, Dirksen’s enforcement formula attracted sufficient support from conservative Republicans to secure cloture.157 Private litigation was again offered by conservative Republicans as a substitute for bureaucratic state-building, and it again commanded broader consensus than the administrative power sought by liberal civil rights advocates. The FHA of 1968 was passed with Dirksen’s enforcement provisions, but with one notable caveat that speaks to legislators’ continuing attention to detail when crafting private enforcement regimes. Robert Byrd (D-WV), who had voted against the CRA of 1964 and the confirmation of Thurgood Marshall to the Supreme Court in 1967, offered an amendment specifying that a plaintiff only be awarded fees if “not financially
CIVIL RIGHTS ACT OF 1964
121
able to assume said attorney’s fees.”158 Byrd argued that, in contrast with Title VII’s back-pay-only rule, FHA plaintiffs could recover all actual economic damages, as well as punitive damages, thus making the recovery of attorney’s fees less imperative. Mondale conceded, saying that the amendment did not interfere with the fee-shifting provision’s purpose, which was to allow “an indigent or impoverished plaintiff . . . to go to an attorney and say, ‘If we are successful, the court may award fees.’ ”159 The Byrd amendment was adopted by unanimous consent. Such details matter, and subsequently, until Byrd’s language was stricken in 1988 amendments to the FHA,160 courts not infrequently denied winning FHA plaintiffs attorney’s fees because they were “financially able to assume” them.161 Implications of the Process-Tracing Evidence In conclusion, I turn to how the process-tracing evidence in this chapter fits with the theories and evidence presented previously. Within the book’s research design, the process-tracing evidence provides an opportunity to observe the causal processes that produced the outcome of an enforcement framework heavily dependent upon private litigation; to assess how well the observed processes support the multiple theories (presented in chapters 2 and 3) of why Congress enacts private enforcement regimes; and to evaluate how consistent the observed processes are with the correlative relationships established in the empirical model presented in chapter 3. Legislative-Executive Conflict. The evidence supports the legislativeexecutive conflict hypothesis, which holds that conflict between Congress and the president over control of the bureaucracy can create incentives for Congress to enact private enforcement regimes, a hypothesis that was supported by the significance of the legislative-executive conflict variables in the empirical models in chapter 3. Conservative Republican legislators explicitly articulated apprehension about ideologically biased abuse of administrative power as a reason to prefer private lawsuits over administrative implementation. They feared that bureaucrats, if given the power that liberals sought in H.R. 405, would be “political” and “arbitrary” in their bias against business, to the point of presuming employers’ guilt; would try to “impose forced racial balance” in the workplace with “mathematical certainty”; and would, “in their zeal to enforce Title VII,” “draw and quarter the victim” (American business). Why did conservative Republicans believe this, when no administrators had yet been nominated? The answer is that they well understood that all five members of
122
CHAPTER 4
the proposed board with final administrative decision-making authority, as well as the head of the Office of the Administrator that would investigate and prosecute, were to be selected by Kennedy (when the bill was in the House) or Johnson (when the bill was in the Senate), and they had little reason to believe that the executive branch would change partisan hands anytime soon. Putting aside their hyperbolic descriptions of the administrators they feared, it was quite reasonable for conservative antiregulation Republicans to believe that the EEOC would be dominated by leadership well to their left. Legislative-executive conflict as a source of private enforcement regimes is not only about divided government, but also about institutional conflict more generally. While Democrats held the presidency in 1964 and controlled both chambers of Congress by wide margins, the unique and profound split within the party over African American civil rights effectively induced simulated conditions of divided government. Members from the eleven former Confederate states constituted about one-third of Democratic seats in both chambers, and without these southern votes Democrats in both chambers lacked a simple majority. This made a sizable share of Republican legislators indispensable to passage, and these Republicans faced a White House long dominated by Democrats, at the time controlled by Kennedy and Johnson, whom they were not eager to arm with potent new administrative powers to regulate American business. Veto Points and Compromise. The evidence also supports the veto points and compromise hypothesis, which maintains that the many veto points in the American lawmaking process can encourage private enforcement regimes as a compromise alternative to bureaucratic state-building. The veto-point-ridden character of the legislative process in Congress was entwined with legislative-executive conflict in contributing to the outcome. Whereas conflict between the policy goals of conservative Republicans and the Kennedy and Johnson administrations contributed to conservative Republicans’ desire for private litigation over bureaucracy, the filibuster in the Senate was significant in empowering them to enact their preferences into law. Dirksen and the wing of the Republican party he represented were gatekeepers of a key veto point in the legislative process—made pivotal by deep divisions within the Democratic party over race—empowering them to refashion the law to better suit their preferences, which they did by substituting private lawsuits for administrative power. The contemporaneous understanding of all parties to the legislative transaction was that this move was a compromise that winnowed down Title VII into a less ambitious and assertive state intervention. This
CIVIL RIGHTS ACT OF 1964
123
scenario, supporting the veto points and compromise hypothesis, was replicated four years later with the FHA of 1968. Judicial Ideology. Pivotal legislators’ assessment of the ideological position of the judiciary, and its attendant implications for regulatory implementation, also factored into their choice of private enforcement litigation. In chapter 2, it was observed that judicial friendliness to congressional preferences should naturally make private enforcement regimes more attractive to Congress, other things being equal (the judicial friendliness hypothesis). It was further observed that judicial hostility to congressional preferences can be counteracted, with respect to its effect on litigant mobilization, by enactment of ever more robust private enforcement regimes (the judicial hostility hypothesis). The judicial ideology variables proved insignificant in the empirical model in chapter 3, where it was concluded that it was not possible, based upon the statistical evidence alone, to judge whether legislators simply were not affected by judicial ideology when constructing regulatory enforcement regimes, or whether the effects of the two countervailing hypotheses were canceling one another out in the data. The evidence in this chapter supports the logic of the judicial friendliness hypothesis. In 1963–64, conservative Republican opponents of an administrative cease-and-desist approach explicitly based their opposition, in part, upon the belief that their business constituents would be treated more fairly by federal judges than by executive administrators. This contention was cast partially in terms of the heightened procedural due process of courts, but concerns about due process were manifestly rooted in fear of ideological bias by administrators insufficiently disciplined by procedural constraints. Conservative Republicans clearly anticipated greater ideological friendliness from federal judges, whom they characterized as “fair” and “impartial,” than from administrators, whom they characterized as “political” and “arbitrary,” and who, they charged, would actually presume the guilt of businesses accused of discrimination. In choosing courts over bureaucracy, pivotal Republicans were influenced by their view that courts would be a more ideologically hospitable environment. Policy Drift. The policy drift hypotheses proceed along two lines: the first holds that current majority coalitions in Congress enact private enforcement regimes to insulate enforcement of regulatory laws that they pass from subversion by future majority coalitions (coalition drift); the second holds that they do so to insulate from subversion by bureaucrats pursuing their own personal preferences (bureaucratic drift). The coali-
124
CHAPTER 4
tion drift hypothesis was supported by the empirical model in chapter 3, which found impending seat share losses by the majority party (electoral uncertainty) to be significantly associated with enactment of private enforcement regimes. The process-tracing evidence, however, provided no support for the coalition drift hypothesis. The historical evidence surrounding passage of Title VII did not reflect that any of the architects of Title VII’s private enforcement regime were motivated by concern about subversion of implementation by future legislative majority coalitions. The bureaucratic drift hypothesis was not susceptible to straightforward testing in the statistical model, though the model did provide qualified support for it. The presence of issue groups in hearings was associated with enactment of private enforcement regimes, and scholars have argued that such groups, beginning in the middle to late 1960s, have been motivated in their advocacy for private enforcement regimes by a fear of bureaucratic drift. As discussed in the next section, evidence of such issue group preferences in the civil rights context had not yet emerged by 1964. The historical evidence is, however, consistent with one strand of the bureaucratic drift hypothesis—the strand rooted in fear that bureaucrats will pursue their own ideological agenda. The evidence presented in the chapter makes abundantly clear that conservative Republicans feared ideologically liberal, expansive, and overzealous implementation of Title VII from H.R. 405’s EEOC. In evaluating the historical evidence, this has been weighed in support of the conclusion that fear of presidential subversion was an important cause of the legislative choice of private lawsuits for Title VII implementation, and presidential subversion is outside the definition of bureaucratic drift set forth in chapter 2. However, to the extent that conservative Republican fear of liberal bias in the agency went beyond expectations about the influence of Democratic presidents, and was based as well upon perceptions about other sources of liberal bias among bureaucrats, this would constitute some support for the bureaucratic drift hypothesis. While conservative Republican denunciations of would-be EEOC administrators were pitched at a rather high level of generality, without clear specification of the roots of the anticipated liberal bias, some of the evidence does point beyond presidential subversion. Republicans were deeply concerned not only about the existence of executive enforcement powers, but also about the location and structure of those powers. They believed that enforcement powers in the hands of a single-mission agency like the EEOC would lead to more assertive intervention into American business than enforcement powers in the hands of the Justice Department. The structural nature of a single-mission agency to regulate business would, they believed, promote organizational culture and behavior biased toward excessive regulation.
CIVIL RIGHTS ACT OF 1964
125
To conservative Republicans in 1963–64, independent of presidential influence, this structural logic in itself entailed a threat of bureaucratic drift in the liberal direction, and it contributed to their support for an enforcement alternative that was importantly dependent upon private lawsuits. Issue Groups. The historical evidence surrounding enactment of Title VII provides no affirmative support for the issue group hypothesis, though it also does not contradict it. That hypothesis holds that beginning in the middle to late 1960s, growth in the number and activism of issue groups, which distrusted bureaucracy and preferred litigation as a vehicle to implement their policy agendas, became an important source of demand driving legislative reliance upon private enforcement regimes. The empirical model in chapter 3 found that the extent of issue group presence in congressional hearings on regulatory legislation was significantly associated with Congress’s enactment of private enforcement regimes. The historical evidence reviewed in this chapter, in contrast, reflected that civil rights groups’ enforcement preferences were exactly the opposite of those emphasized by the issue group hypothesis. They emphatically wanted a muscular administrative apparatus, not private litigation, to implement Title VII, and they regarded Republicans’ imposition of private litigation over administrative machinery as a painful defeat. Groups’ cognizance of the economic obstacles to private enforcement did lead them to seek fee shifting and appointment-of-counsel provisions when it became clear that they largely would be relegated to private lawsuits for enforcement, but in doing so they were making the best of what they perceived to be a bad situation. This evidence does not contradict the issue group hypothesis, however, because the events traced in this chapter occurred on the eve of the period in which that hypothesis predicts a transformation in issue group preferences away from bureaucracy and toward private lawsuits. Subsequent chapters thus will provide more suitable process-tracing evidence with which to evaluate the issue group hypothesis. Rent-Seeking Lawyers. The historical evidence is at odds with the rentseeking lawyer hypothesis, which holds that demand for private enforcement regimes by for-profit lawyers, acting in their own financial interests, is an important reason that legislators enact them. This lack of support is consistent with the findings of the empirical model in chapter 3, which detected no relationship between the presence of lawyers’ association representatives in congressional hearings on regulatory legislation and Congress’s propensity to enact private enforcement regimes. In the extensive legislative historical record surrounding the enactment of Title VII, there is simply no evidence that lawyers’ associations were a
126
CHAPTER 4
mobilized interest involved in trying to shape its enforcement provisions. There was only one lawyers’ association (albeit the most important one, the ABA) among the forty-four groups lobbying in favor of the CRA of 1964, and none testified in the hearings on H.R. 405 or the Kennedy administration’s civil rights bill. Within the substantial secondary literature on the legislative history of the CRA of 1964 relied upon in this chapter, no evidence of rent-seeking lawyer activity is identified. It must be acknowledged, though, that unlike groups sincerely pursuing policy goals, rent-seeking lawyers’ associations and legislators serving them—since they would have been hijacking public polices for the personal enrichment of an unpopular group—may have self-consciously acted in secret. This would have been economically and politically rational behavior, and it would explain lack of affirmative support for the rent-seeking lawyer hypothesis in the historical record. However, the rent-seeking lawyer hypothesis, as applied to Title VII, seems quite doubtful given what is in the historical record. We know that antiregulation Republicans were the driving force behind inserting private lawsuits into Title VII. We also know that the reasons they openly gave for doing so were quite consistent with conservative Republican preferences: to avoid the creation of a bureaucratic apparatus that would be controlled by their ideological adversaries; to shift to an implementation venue that they believed would be more favorable to businesses (would not presume their guilt, or “draw and quarter” them); and generally to reduce the law’s “harassment” of business. While it is not impossible that conservative Republicans who gave these reasons for preferring private lawsuits over bureaucratic state-building were actually secretly motivated to enrich lawyers in general, or “trial lawyers” in particular, at the expense of business, it is not plausible. The historical record strongly suggests that the provision allowing successful plaintiffs to recover attorney’s fees was sought by civil rights groups once it became clear that private enforcement was the best they could do, and that the provision was obtained by Kennedy Justice Department (particularly Civil Rights Division) lawyers who negotiated the compromise on Title VII with Dirksen and his staff. The historical record also strongly suggests that civil rights advocates’ goal in seeking the fee-shifting and appointment-of-counsel provisions was to overcome well-known economic obstacles to private enforcement, as articulated by NAACP LDF director-counsel Jack Greenberg, and as stated in the legislative record by Humphrey. While it is possible that Civil Rights Division leadership and legislators such as long-standing civil rights liberal Hubert Humphrey were secretly motivated to enrich lawyers, rather than to foster prosecution of meritorious Title VII claims, and to respond to important civil rights groups’ demands in the context of an impending
CIVIL RIGHTS ACT OF 1964
127
election in which civil rights were sure to be an enormously salient issue, this motivation seems implausible. Party Alignments. The strong form of the party alignment hypothesis is not supported by the evidence surrounding passage of Title VII, but that evidence is compatible with the weak form of the hypothesis. The strong form holds that enactment of private enforcement regimes is a distinctively Democratic phenomenon, while the weak form holds only that Democrats are more likely to enact private enforcement regimes than Republicans. Two of the three empirical models in chapter 3 found the degree of Democratic control of Congress to be a significant variable predicting enactment of private enforcement regimes, though the magnitude of the effect was far from dominant, and the effects of no other explanatory variables were conditional upon it. The models thus supported the weak form and rejected the strong form of the party alignment hypothesis. In chapter 3 we observed some examples of Republican Congresses incentivizing private lawsuits to serve Republican constituents, such as business in conflict with unions, Cubans divested of property by the Castro regime, and abortion opponents. When shaping Title VII’s enforcement provisions, antiregulation Republicans sought, to an important extent, to act on behalf of their business-employer constituents, particularly big business in the North, which actively sought to diminish administrative enforcement. As contrasted with the above examples of Republicans enacting private enforcement regimes to achieve the affirmative enforcement goals of constituents, the legislative battle over Title VII enforcement introduces a different way that private enforcement regimes can serve some Republican constituencies: as a weapon to defeat bureaucratic state-building, which they may regard as worse than private lawsuits. It is certainly true that when Dirksen sought to insert a private right to sue in the Senate in exchange for curtailment of EEOC enforcement powers, it was liberal Democrats and civil rights groups that pushed Dirksen to include in his substitute bill incentives to ensure that the private right of action would be used. When Southerners offered amendments to sever these enforcement incentives from the private right of action and strike them from the law, Republicans and nonsouthern Democrats voted together by a wide margin to defend them. Budget Constraints. The budget constraint hypothesis—which holds that resistance to government spending motivates Congress to enact private enforcement regimes—is not affirmatively supported by the processtracing evidence. Lack of support for the budget constraint hypothesis is consistent with the findings of the empirical model in chapter 3, which
128
CHAPTER 4
detected no relationship between budget surpluses or deficits and Congress’s propensity to enact private enforcement regimes. In 1963–64, no participants in the legislative process expressed the view that a private enforcement regime was the preferred approach to Title VII implementation because it would reduce the need for governmental expenditures on bureaucratic machinery. While opposition to government spending can be a motive for opposing growth in the bureaucratic state, in the course of conservative Republicans’ repeated and extensive attacks on H.R. 405 as excessive bureaucratic state-building, not a word was uttered about monetary costs.162
Chapter 5 REVERBERATIONS 1965–1976
In the decade following the Civil Rights Act of 1964, private civil rights enforcement took firm root and began to flourish in a manner and to an extent that was not expected by Title VII’s architects on the left or right. Political, social, and economic reality is complex, and human beings are imperfect prognosticators of the long-run policy “feedback” effects of interventions in that reality.1 Among forms of unanticipated feedback, the self-reinforcing type is critical to the story of civil rights litigation following passage of the CRA of 1964. The long-run growth and entrenchment of an institution can turn upon whether it generates self-reinforcing dynamics. If it does so, a policy intervention can be transformed into one of far larger than anticipated magnitude.2 Two distinct self-reinforcing feedback processes are critical to the present discussion. First, an initial intervention can generate learning that transforms the preferences of important actors, altering their understanding of what is possible or desirable, thereby increasing demand to retain the policy and expand its application.3 Second, an initial intervention can create valuable goods—political, policy, or personal resources and opportunities— that social groups mobilize to defend and enlarge.4 While the founding in 1964 of Title VII’s private enforcement regime was rooted in ideological conflict on the terrain of America’s fragmented political institutions, self-reinforcing feedback from the 1964 act contributed importantly to its consolidation and expansion in the 1965–76 period. These feedback effects flowed through, and were shaped by, the same institutional fragmentation that gave rise to Title VII’s private enforcement regime, particularly legislative-executive conflict over control of the bureaucracy. Self-reinforcing effects were evident both in policy learning that shifted civil rights advocates’ preferences in favor of private enforcement, and in the resources and opportunities that the 1964 compromise created, which groups mobilized to defend and extend. This feedback led to the entrenchment and growth of private civil rights enforcement. By the late 1960s, during a campaign to amend Title VII to give the EEOC cease-and-desist powers, civil rights advocates refused to give up private lawsuits with fee shifting in exchange for the pure administrative cease-and-desist framework that they had unsuccessfully sought
130
CHAPTER 5
in 1963–64. They pushed instead for a hybrid model including both private litigation and cease-and-desist powers, in which private litigation would play a major role and would serve as a safety valve, providing a measure of insulation from deficiencies in the agency’s performance. In Title VII’s early years, civil rights advocates witnessed an executive enforcement machinery lacking both the material resources and the political will and commitment to carry out its mission, as they conceived it. This was true, in their view, even under Johnson, who had so forcefully pushed the CRA of 1964 through Congress. It became palpably more true when Nixon took office, and civil rights groups and civil rights liberals in Congress entered upon a period of open antagonism with the White House, claiming that it was consciously evading statutory mandates. At the same time, empowered by the private enforcement provisions offered by Republicans in 1963–64, civil rights groups found themselves at the leading edge of Title VII enforcement, wielding the weapon of private litigation to make what they judged to be new, meaningful, and gratifying inroads into labor markets previously foreclosed to African Americans. By the end of the 1960s, civil rights advocates had decided they could not afford to rely solely upon the beneficence of bureaucrats. Private litigation with fee shifting, as an insurance policy against underenforcement by the executive branch, was too valuable to give up. Additional self-reinforcing effects from Title VII’s private enforcement regime reverberated through the first half of the 1970s, ultimately contributing to the spread of fee-shifting across the entire field of civil rights. Congress’s reliance on private lawsuits with fee shifting in Titles II and VII of the CRA of 1964, and again in the CRA of 1968, fostered the growth of civil rights legal advocacy groups and spawned a for-profit civil rights bar, which together increased demand for fee shifting in the domain of civil rights. By the early to mid-1970s, the successful mobilization of considerable levels of private enforcement by for-profit attorneys had dispelled skepticism among civil rights advocates about the capacity of fee shifting to generate an adequate volume of private enforcement. Indeed, they came to regard the private for-profit bar as a powerful enforcement infrastructure to be cultivated. Continuing enmity between the Nixon administration and civil rights liberals, inside and outside of Congress, reinforced and deepened their commitment to private enforcement. They succeeded in securing legislative extension of the CRA of 1964’s fee-shifting approach first to school desegregation, and then to voting rights, and ultimately across the entire field of civil rights with the Civil Rights Attorney’s Fees Awards Act of 1976. To justify passage of that act, a broad and bipartisan coalition of legislators pointed to the success of fee shifting in mobilizing robust private enforcement in the recent civil rights laws; to the insufficiency of executive enforcement; and to the abil-
REVERBERATIONS: 1965–1976
131
ity of fee-shifting rules to provide needed enforcement without increasing bureaucracy or budgets. Thus was created the modern approach to civil rights enforcement.
The Road to the Equal Employment Opportunity Act of 1972 The worries of the liberal critics of the Dirksen compromise on enforcement appeared to be borne out almost immediately, soon giving rise to a bipartisan consensus on the need to shore up Title VII’s enforcement provisions. Shortly after Title VII became effective in July 1965, the EEOC was drowning in a volume of complaints far larger than anticipated, which it struggled to process with a modest budget. Its rapidly growing backlog created processing delays of fifteen months by 1967 and of two years by 1968 and 1969.5 By the end of 1966, the agency, having received nearly 15,000 complaints, had achieved conciliation—meaning some relief for the claimant voluntarily conceded by the employer—in only 330, or 2 percent of them.6 With the EEOC lacking any powers of coercion, employers were, by and large, simply unmoved by the agency’s suasion, making efficient and expeditious settlement of complaints impossible.7 As the House report on a bill to strengthen EEOC powers would later put it: “Instead of cooperating with the EEOC to resolve charges against them, employers ignored the conciliation process and relied, often successfully, on the assumption that aggrieved employees would fail to litigate their claims independently.”8 In a 1967 interview with the Wall Street Journal, EEOC chairman Stephen Shulman characterized the agency’s problem: “We’re out to kill an elephant with a fly gun.”9 This state of affairs was not a logical correlate of the lack of EEOC enforcement powers. Theoretically, in the absence of cease-and-desist powers or a right to sue vested in the EEOC, its ability to achieve voluntary conciliation where it found discrimination would depend on the threat of litigation posed by the Justice Department or private parties. However, by the end of 1967, when Title VII had been in effect for two and a half years, the Justice Department had filed only ten pattern-or-practice suits.10 The Civil Rights Division within the department was small, and it was responsible for covering the whole domain of civil rights, within which employment discrimination was a low priority, of only marginal concern relative to voting rights and school desegregation.11 Under the Dirksen compromise, recall, the Justice Department was not required to sue even in manifestly meritorious cases, and by and large it elected not to. Private lawsuits were also few in number during Title VII’s early years, notwithstanding the provisions meant to facilitate them. By September
132
CHAPTER 5
1967, more than two years after Title VII’s effective date, only fifty-six private suits had been filed, and of them at least thirty-seven were filed by the NAACP’s LDF, such that, in the entire country, at most nineteen cases were prosecuted by non-interest group private attorneys.12 Whereas the LDF and other nonprofit civil rights groups could play an important role in developing policy through impact litigation—one they had long played, most notably in the campaign culminating in Brown—they lacked the resources to provide day-to-day enforcement of ordinary claims for the entire national workforce. Moreover, the LDF’s focus was on race, and thus it appears that, in its first few years, Title VII was scarcely enforced in cases relating to gender, religion, and national origin. Testifying before a Senate committee in 1969, EEOC chairman William Brown reported that in cases in which the agency found that discrimination had occurred but conciliation efforts failed to secure relief for the claimant, he or she did not proceed with litigation over 90 percent of the time.13 It appeared to observers that because of the massive expense of litigation, the typically small and uncertain economic damages involved, the gross imbalance of resources and legal sophistication between employers and often poor and uneducated employees, the notorious slowness of the judicial process, employees’ fear of retaliation, and bigoted judges in the South, private litigation brought by for-profit attorneys simply would not produce an adequate volume of suits to address enforcement requirements.14 If the Dirksen compromise on enforcement had been engineered by its author to minimize “harassment” of business under Title VII, from the vantage point of the late 1960s it appeared to have worked. As a result of the insubstantial level of enforcement by private litigants and the Justice Department, and the consequent weakening of EEOC conciliation efforts, not long after the agency opened its doors, President Johnson, liberal legislators, and civil rights groups urged that Title VII be amended to give the EEOC cease-and-desist powers. Bills containing such powers, among other measures to strengthen Title VII, were introduced in every Congress between 1965 and passage of the Equal Employment Opportunity Act of 1972. Though a bipartisan consensus emerged that Title VII’s enforcement provisions were feeble and in need of strengthening, many Republicans who favored strengthening amendments ultimately rejected cease-and-desist, preferring instead to empower the EEOC to enforce through lawsuits, following the prosecutorial agency model in the version of Title VII initially passed by the House in 1963.15 In 1965, cease-and-desist powers for the EEOC passed the House but then died in the Senate; in 1970, they passed the Senate but were killed in the House Rules Committee. The Senate version of the bill that would become the Equal Employment Act of 1972 initially contained EEOC cease-
REVERBERATIONS: 1965–1976
133
and-desist powers, but after repeated motions for cloture in the Senate were defeated, a bargain was struck eliminating them in exchange for the cloture votes of conservative Republicans.16 When the Equal Employment Act of 1972 was ultimately passed, cease-and-desist was defeated. Instead, the EEOC was given the authority to prosecute both patternor-practice and ordinary individual suits.17 Given the agency’s resource starvation and immense backlogs, however, it was not clear how much additional enforcement this legal power would actually produce. The dynamics of legislative coalitions that produced this outcome were similar to those in 1964. Civil rights groups in alliance with liberal legislators, primarily nonsouthern Democrats, were vigorous supporters of ceaseand-desist, and southern opposition to any strengthening of Title VII’s enforcement provisions was vehement, producing a filibuster in the Senate. Between these two poles, the pivotal voters were again conservative Republicans, who rejected cease-and-desist in favor of the court-based prosecutorial agency model, once again deriding what they regarded as liberal efforts at excessive bureaucratic state-building and expressing fear that they would not be able to control an agency in the wrong political hands.18 The essential character of the Dirksen compromise on Title VII enforcement was confirmed and entrenched, mixing public and private enforcement lawsuits, with the lion’s share of responsibility for enforcement lying with private plaintiffs and their attorneys. In the course of the failed campaign for EEOC cease-and-desist powers between 1965 and 1972, there was a gradual and important shift in the perspective of civil rights groups on the value of private Title VII lawsuits. Although these groups had not sought private enforcement in 1963–64, by the end of the 1960s they guarded it fiercely. During the renewed push for cease-and-desist, some elements demanded striking the private right of action as the price to be paid for cease-and-desist authority.19 In 1966, the AFL-CIO, by far the most powerful labor organization within the Leadership Conference, conditioned its support for authority on eliminating the private right of action. A number of early and successful Title VII suits had targeted AFL-CIO unions, and federal court decisions had intruded into union affairs far more than had administrative enforcement under state fair employment practice commissions. The AFL-CIO thus would support cease-and-desist instead of, but not in addition to, private litigation.20 In 1967, civil rights groups within the Leadership Conference, following the lead of Jack Greenberg of the LDF, which had been involved in financing and prosecuting a large majority of the Title VII suits, refused to support a bill trading private enforcement for cease-and-desist, which would have effectively reproduced H.R. 405. They did, however, reluctantly agree to accept a significantly weakened private right of action
134
CHAPTER 5
in exchange for AFL-CIO support for cease-and-desist powers.21 A bill subsequently proposed by the Johnson administration in 1967, and introduced by liberal Democrats Emanuel Cellar of New York in the House and Philip Hart of Michigan in the Senate, embodied this compromise. The EEOC was given cease-and-desist powers, and a private right of action was retained under certain circumstances outlined below, but the provisions allowing appointment of counsel, waiver of the filing fee, and awards of attorney’s fees to winning plaintiffs were struck out.22 This hybrid enforcement regime, combining cease-and-desist with private litigation, would work as follows. If upon initial investigation the commission did not find “reasonable cause” to credit the charge of discrimination, then the complaint would be dismissed without a hearing, and the complainant would be free to proceed with private litigation de novo in district court. Under H.R. 405, if the complaint were dismissed at this stage, after investigation but without a hearing, the complainant would only have had access to appellate review, which is extremely deferential to the agency. Civil rights advocates were no longer willing to give the agency the power to effectively terminate a claim following investigation but without a full evidentiary hearing. As in H.R. 405, if the agency did find reasonable cause to credit the complaint, then it would attempt to conciliate a voluntary settlement, and failing that, it would hold an evidentiary hearing and render a decision, which would be subject only to deferential appellate review on the agency record. If the commission had not initiated the hearing procedure within 180 days from when the complaint was filed, the complainant would be free to exit the administrative process and file a lawsuit. Civil rights advocates were no longer willing to wait indefinitely for action by the agency. It is critical to stress that the agency’s powers under this framework were considerable, though less than in H.R. 405. If the agency elected to hold an evidentiary hearing and acted quickly, a genuine cease-and-desist model would prevail and courts would play only a limited role. However, if the agency elected not to hold an evidentiary hearing or moved too slowly, claimants would be free to litigate. The Ninety-ninth Congress failed to act on this bill. In 1968, civil rights groups within the Leadership Conference, again led by Jack Greenberg, reversed course, repudiated their deal with the AFL-CIO, and refused to agree to eliminate the provisions regarding appointment of counsel, waiver of the filing fee, and attorney’s fee awards, even if this position meant turning the most powerful labor organization in the United States against cease-and-desist powers for the EEOC.23 Instead, civil rights advocates wanted those provisions attached to the private right of action within the hybrid cease-and-desist / private litigation framework just out-
REVERBERATIONS: 1965–1976
135
lined. This is what they lobbied for between 1968 and passage of the Equal Employment Act of 1972.24 Why would civil rights groups no longer accept H.R. 405? Why did their preferences change? The dynamics of self-reinforcing policy feedback, in the form of policy learning, were clearly at play, a change that was importantly fostered by a rightward shift in the executive branch when Nixon took office. The logic of policy learning as a form of feedback is straightforward. Under constraints of bounded rationality, the human capacity to predict the effects of new policies, created in a complex world, is limited. Actors at or near the center of the policy process, however, are able to learn from the performance of past policies, and this learning guides future policy decisions.25 Since 1964, civil rights groups had learned painful lessons about limitations in executive enforcement of Title VII that went well beyond the EEOC’s lack of cease-and-desist powers, and these lessons had shaken their confidence in bureaucracy as the sole instrument of implementation. Over the same period, civil rights groups also learned gratifying lessons about what they could achieve through litigating Title VII claims. Although only negligible private enforcement of Title VII by the for-profit bar had materialized, the NAACP’s LDF, by far the largest prosecutor of Title VII actions in the country, was encountering a hospitable judiciary and achieving stunning successes. These two sets of lessons learned from experience with Title VII implementation since 1964—negative with respect to bureaucracy and positive with respect to private litigation—transformed the relative weighting of civil rights groups’ preferences for administrative adjudication versus litigation. This transformation was deepened and reinforced when Nixon took office and open conflict broke out between civil rights advocates and the new administration over control of the civil rights bureaucracy. Loss of Faith in Executive Power The Brookings Institution and the United States Commission on Civil Rights funded a study, conducted in 1967 and 1968, on the state of Title VII enforcement, and in the course of the study leaders of civil rights groups were surveyed on their views of existing enforcement.26 Although the EEOC was highly sympathetic toward and cooperative with civil rights groups, particularly the NAACP, in sharing information and developing legal arguments and strategy,27 civil rights groups were gravely concerned that the EEOC was so underfunded and understaffed relative to the task confronting it that it was incapable of adequately processing claims, even with the best of intentions.28 Presidential budget requests were inadequate, and congressional authorizations were worse, preclud-
136
CHAPTER 5
ing meaningful investigation and evaluation of claims within a reasonable time period. According to the study, “Civil rights leaders canvassed by questionnaire . . . came down hardest on delays in processing complaints,” reporting that years passed after complaints were filed, with no contact from the agency’s staff, leading from a “first flurry of hope” at the time of filing to “a feeling of complete hopelessness when the complainants never heard from the EEOC.”29 To civil rights advocates, allowing claimants to file a lawsuit, exiting the administrative process if it dragged on too long, would provide an “escape from the administrative quagmire” that the agency’s proceedings had become.30 Civil rights leaders also objected that Title VII enforcement was an extremely low executive priority, even within the Johnson administration. This lack of executive commitment was in evidence at both the EEOC and the Department of Justice. Johnson was slow to staff the new agency and left EEOC leadership posts vacant for extended periods.31 The agency’s case-by-case processing was far too conservative, civil rights advocates complained, urging instead that it pursue a more ambitious enforcement program targeting systemic discrimination.32 In Jack Greenberg’s estimation, James Roosevelt, the EEOC’s first chair, “did little or nothing, either out of torpor or for political reasons.”33 The agency’s political timidity caused Greenberg and other civil rights advocates to worry that giving up the private right of action for pure administrative enforcement could lead to agency capture. Greenberg later wrote, “I have no doubt that if there had been only administrative enforcement, employers and unions would have been all over the EEOC trying to shape what would happen.” He observed that the private right of action “kept the EEOC honest because it didn’t want to be shown up by private enforcers.”34 Alfred Blumrosen, who was chief of conciliations at the EEOC from 1965 to 1967, worried by 1970 that giving up the private right of action for cease-and-desist would render the EEOC “a captive of those interests which were to be regulated.”35 Testifying in 1967 during hearings on Title VII enforcement, Whitney Young, executive director of the National Urban League, summed up civil rights activists’ dejection over the agency: “The actual agency experience, which has demonstrated that the Commission cannot enforce compliance, has given rise to disillusionment and lack of confidence.”36 The Justice Department, which in Title VII’s early years meekly used its discretionary right to sue, did not inspire confidence in executive enforcement among civil rights groups either.37 Testifying in the 1967 congressional hearings on proposed Title VII amendments, Roy Wilkins, executive director of the NAACP, protested “the extremely selective and limited use” that the Justice Department had made of its right to sue.38 In 1969 hearings on the same subject, Greenberg, echoing his concerns about
REVERBERATIONS: 1965–1976
137
agency capture, suggested a more jaundiced perspective, noting that when the Justice Department fails to enforce the law against powerful unions and businesses, “suspicion arises that the Government did not act because some labor union or corporation used its political influence.”39 This, in Greenberg’s view, carried over an unfortunate practice from the Kennedy administration under which civil rights groups “had repeatedly seen civil rights enforcement compromised by political considerations.”40 Rightward Shift in Executive Enforcement Preferences Whereas civil rights advocates’ disillusionment over executive commitment to vigorous civil rights enforcement took root under the Johnson administration, when Nixon assumed office their relationship with the executive branch worsened palpably. Nixon had deployed his storied “southern strategy” in the 1968 campaign, endeavoring to win over white southern Democrats by convincing them that the Republican party better represented their preferences on a variety of issues, including civil rights.41 Once Nixon assumed office, his relations with civil rights liberals became overtly antagonistic, and he was attacked by NAACP leaders as “anti-Negro,” as the worst president on civil rights during the postwar period, and as engendering a “general uneasiness” among African Americans.42 While civil rights advocates had been disappointed by what they regarded as the Johnson administration’s lackluster performance on civil rights enforcement, they regarded Nixon as hostile and obstructionist, further reducing their expectations for executive enforcement of civil rights laws in general, and of job discrimination laws in particular. Within a few months of Nixon assuming office, conflict over Title VII enforcement erupted between civil rights advocates and the new administration. In March 1969, Senator Dirksen publicly called on the new president to oust Clifford Alexander, the still-serving EEOC chair appointed by Johnson, because of his “harassment” of business on the issue of job discrimination, and the White House drew fire from civil rights advocates by announcing the next day that Alexander would be replaced.43 In response Alexander resigned, attributing his decision to “a crippling lack of Administration support,” and complaining that the Justice Department refused to prosecute clear violations of Title VII uncovered by commission hearings and referred to the department for prosecution.44 In early May, the United States Commission on Civil Rights, on which Democratic-appointed commissioners still enjoyed a four-to two margin, issued a public letter attacking as “seriously deficient” the Nixon administration’s enforcement of federal laws and regulations prohibiting job discrimination, and accusing it of “subsidizing discrimination” with public money by contracting with firms known to discriminate.45
138
CHAPTER 5
The letter attacking the Nixon administration was written, on behalf of the commission, by Howard A. Glickstein, the Johnson-appointed staff director of the commission who had earlier helped write the CRA of 1964 while an attorney in the Civil Rights Division of the Kennedy Justice Department.46 In August, the EEOC chair appointed by Nixon to replace Alexander testified in congressional hearings, consistent with the administration’s position, against giving the EEOC cease-and-desist powers (in the hybrid regime discussed above), which had been sought by Johnson, Alexander, nonsouthern congressional Democrats, and civil rights groups.47 These loud public controversies over the Nixon administration’s enforcement of job discrimination laws occurred in the months leading up to the August 1969 congressional testimony of the LDF’s Jack Greenberg and other civil rights leaders (discussed in detail in the next section), imploring Congress to retain Title VII’s private enforcement regime when some suggested eliminating it in favor of a cease-and-desist framework. One AFL-CIO official testifying on the same day as Greenberg and the other civil rights leaders put his finger on the linkage between the rightward shift in executive preferences and civil rights groups’ insistence on retaining private enforcement, referring to the understandable “desire of minority workers, and the organizations which represent them, to retain private rights of action, independent of the vagaries of, and changes in, Government agencies.”48 By the first summer of Nixon’s presidency, private enforcement of Title VII looked even more critical to civil rights advocates than it had under Johnson. Moreover, from the perspective of civil rights liberals, the rightward shift on civil rights in the executive branch went well beyond job discrimination. It also included school desegregation and busing, civil rights in higher education, voting rights, housing, and Nixon’s nomination of southern conservatives to the Supreme Court.49 Clear evidence of this perception on the part of civil rights liberals can be seen in the formation of a new subcommittee in the House Judiciary Committee specifically for the purpose of overseeing executive implementation of civil rights laws. At least partly at the suggestion of the Leadership Conference on Civil Rights, which then represented over 125 national organizations, the Civil Rights Oversight Subcommittee of the House Judiciary Committee was formed in 1971.50 Its first hearings in June 1971 were convened to investigate, according to subcommittee chair Don Edwards (D-CA), widespread complaints by “individuals and groups of varied backgrounds and from all parts of the country . . . that the Attorney General has failed properly to enforce the Voting Rights Act, has failed to carry out the intent of Congress, and has disregarded recent Supreme Court decisions protecting the right of all Americans to exercise their right to vote.”51 Civil rights
REVERBERATIONS: 1965–1976
139
advocates in the hearings accused attorney general John Mitchell of willful nonenforcement with the motive of winning white political support in the South.52 At the inaugural hearings, Joseph Rauh, general counsel of the Leadership Conference, made abundantly clear why the organization had pressed for creation of the new civil rights oversight subcommittee. Said Rauh: “I wish to express the indignation of the Leadership Conference at the general failure to adequately enforce civil rights legislation, especially in the last two years. . . . [W]hile there has been evidence of apathy and lack of response in other Administrations, we detect in the present one a conscious effort to frustrate the enforcement of laws that Congress has enacted. . . . It is comforting to have a committee here for those of us who have worked on getting the laws passed and have seen them mangled.” Rauh gave examples of the Leadership Conference’s grievances with the Nixon administration in the areas of housing, employment, voting, and education, particularly school desegregation.53 Conflicts over enforcement of school desegregation mandates emanating from the Brown decision were acute, and powerfully illustrate how Nixon’s assumption of office fueled both civil rights groups’ and liberal legislators’ preference for private enforcement lawsuits as an alternative to executive power. In July 1969, Nixon came under sharp attack from civil rights leaders, including Roy Wilkins, for what they regarded as the administration’s relaxation of executive pressure on states to desegregate public schools, and in particular its strong disinclination cut off federal funds to schools failing to comply with Brown, a move Nixon thought would alienate southern supporters.54 This potent tool to punish noncompliant schools, exercised through the Department of Health, Education, and Welfare, was created in Title VI of the CRA of 1964—which bans race discrimination in state programs receiving federal money—and was favored by civil rights groups.55 In August, Nixon’s Justice Department switched sides, relative to the department’s position under Johnson, to oppose NAACP lawyers in a desegregation lawsuit seeking the cutoff of federal funds, placing the department on the opposite side of the NAACP on a major issue for the first time since the department began participating in civil rights litigation in the 1950s.56 By 1970, civil rights groups had launched a series of lawsuits against Nixon’s HEW, alleging that it was failing to enforce Title VI and thereby allowing unlawful discrimination in elementary and secondary schools, and in higher educational institutions, mainly in the South but also across the country.57 The controversy regarding enforcement of federal school desegregation mandates produced straightforward legislative-executive conflict over the administrative state—a phenomenon that was on the rise in the
140
CHAPTER 5
Nixon years across many areas of policy58—leading to congressional mobilization of private litigants as an alternative to administrative power. Nonsouthern congressional Democrats expressly linked their discontent with the administration’s performance on enforcement of school desegregation with the necessity of mobilizing private litigants as an alternative route to enforcement. In the Emergency School Aid Act of 1972, the main purpose of which was to provide financial assistance to schools undergoing desegregation in order to facilitate that process, Congress provided for attorney’s fees for winning plaintiffs in school desegregation cases.59 It cited the CRAs of 1964 and 1968 as precedent for this approach to enforcement.60 The 1971 report of the Democrat-led Senate Committee on Labor and Public Welfare, which reported the bill, stated that the fee awards were necessary because the executive branch “is devoting neither the time, effort, nor the financial resources necessary for adequate law enforcement,” and as a result school desegregation “laws are not being enforced throughout the nation.”61 Committee member and civil rights liberal Walter Mondale (D-MN) was the chief advocate on the Senate floor of the need to mobilize private enforcement lawsuits: The truth of it is that there is a major, enormous gaping law enforcement crisis in the field of civil rights. . . . I was appalled to find that the Department of Justice spends $1 million a year on the enforcement of school desegregation laws in this county. . . . [S]ection 11 providing for fee awards tries to deal nationally with the law enforcement crisis . . . by bringing to bear a private law-enforcement remedy in light of the apparent futility of prevailing upon the Department of Justice and the Department of Health, Education, and Welfare to do their jobs. Time and time again I urged them to pursue a national enforcement policy, but up to this point it has been in vain.62 In light of this, Mondale concluded, “One of the most effective ways to achieve this [national enforcement policy] . . . is to permit aggrieved persons to bring a lawsuit,” and to recover attorney’s fees if successful.63 Mondale and the Committee on Labor and Public Welfare were echoing complaints made by civil rights advocates among witnesses in the hearings on the School Aid Act, who urged mobilizing private litigants as an alternative to reliance upon the Nixon administration for enforcement of federal school desegregation mandates. Howard Glickstein, the Johnsonappointed staff director of the Commission on Civil Rights, whose attacks on the Nixon administration for underenforcing job discrimination laws were cited above, also claimed that Nixon had failed to seriously enforce school desegregation laws,64 a charge repeatedly leveled against the administration by the commission.65 In the hearings on the School Aid
REVERBERATIONS: 1965–1976
141
Act, Glickstein stated that federal desegregation mandates had “not been adequately enforced by the agencies charged with the responsibility,” and suggested the need for “a mechanism to by-pass unwilling agencies,” 66 such as by incentivizing private lawsuits. Marian Wright Edelman of the Washington Research Project, who also criticized Nixon for weak leadership on school desegregation, observed that “[r]egulations are meaningless if administering agencies do not adhere to them,” and testified in favor of fee shifting for private enforcement litigation as a necessary alternative to reliance upon the administration for enforcement.67 Clarence Mitchell of the NAACP also complained of lack of federal enforcement; he argued in favor of fee shifting to make private desegregation lawsuits economically feasible, and stated bluntly: “[B]lack people . . . don’t trust the Nixon administration.”68 Taking a chapter from Dirksen’s book, liberal Democrats and their civil rights group constituents were now the ones who distrusted administrative power under the executive supervision of their ideological adversaries, and who turned to private lawsuits as an alternative to administrative power. In contrast with Dirksen’s concern about the overzealous use of administrative power, civil rights liberals were concerned about the underzealous use of it. It is noteworthy that the bill, as originally reported by the Senate Committee, contained a very unusual provision: it allocated $15 million for the creation of a fund from which attorney’s fees would be paid to successful private litigators of school desegregation cases, rather than having the fees paid by school districts found to be discriminating.69 Mondale said that this approach was preferred by the committee over making losing school districts pay so that fee awards would not deprive schoolchildren of already scarce education dollars.70 In comparison to the proposed $15 million fund, the annual budget of the Education Section of the Civil Rights Division in the Justice Department was about $1 million, an amount that the committee and Mondale characterized both as grossly inadequate and as evidence of a lack of executive commitment to vigorous enforcement.71 The Democrat-led committee was prepared to devote significant resources to the enforcement of school desegregation laws, but it judged the resources as more effectively spent on mobilizing private lawsuits than on funding Nixon’s Justice Department. Regarding money expended by the Nixon administration in 1971—which had been appropriated by Congress to facilitate desegregation—Mondale alleged that “much of it actually went to support segregation, not desegregation practices.” He characterized the use of public money to pay private lawyers to prosecute desegregation suits, rather than to fund the Nixon bureaucracy, as “a new and exciting way to assure that what we want done is, in fact, done,” “to be sure that our will is carried out,” and to “make certain that the funds
142
CHAPTER 5
are used for … [desegregation] and not diverted into illegal, wasteful, or unconstitutional efforts.”72 This reinforces the point that the Committee on Labor and Public Welfare regarded the Nixon administration as willfully evading its enforcement responsibilities, rather than as suffering from lack of resources. Aside from the issue of cost, the committee thought that private lawsuits were simply a more effective policy tool than Nixon’s Justice Department. On the Senate floor, Republican Peter Dominick of Colorado successfully offered a roll call striking out the section creating the $15 million fund and authorizing fee awards from it.73 Two days later, Republican Marlow Cook of Kentucky proposed a new provision awarding attorney’s fees to prevailing plaintiffs to be paid by losing defendants (school districts), an alternative to the public fund approach that he had urged when arguing in favor of the Dominick amendment.74 Some Republicans opposed the $15 million fund on the ground that taxpayers across the country should not be required to pay for the misdeeds of lawless school districts, and at the same time supported fee recovery for plaintiffs to be paid by the school districts themselves.75 Nonsouthern Democrats voted in favor of the fund approach at a rate of 89 percent, and after it was defeated by a coalition of Republicans and southern Democrats, nonsouthern Democrats voted in favor of imposing fees on losing defendants at a rate of 80 percent. Legislators from the former Confederacy voted overwhelmingly against both approaches (84 percent against the former, and 100 percent against the latter). Only Republicans had materially different positions in the two votes. They opposed the fund at a rate of 73 percent, but supported imposing fees on discriminating school districts at a rate of 62 percent.76 Republicans supported more enforcement, but they did not want to spend federal dollars on it. As with job discrimination in 1964 and fair housing in 1968, with school desegregation in 1971 it was again a Republican who proposed an amendment adding fee shifting for winning plaintiffs, and Republicans again were pivotal in dictating that outcome. Judicial Hospitality and Litigation Successes In contrast with civil rights groups’ flagging confidence in executive enforcement, they had a buoyant sense of their own efficacy, through litigation, as enforcers in the domain of job discrimination. Whereas attorneys in the for-profit sector had shown almost no interest in Title VII claims, by August 1969 the NAACP’s LDF, by far the largest prosecutor of such claims, had filed more than seventy cases.77 In the 1969 Senate hearings on bills that led to the Equal Employment Act of 1972, testifying in favor of the hybrid cease-and-desist / private litigation approach, Greenberg
REVERBERATIONS: 1965–1976
143
expressed far greater concern for preserving private litigation, the virtues of which were the chief focus of his remarks, than for adding cease-anddesist. Said Greenberg: “[T]hat part of the proposed law which most interests us as a private agency deeply involved in the implementation of Title VII is the section which preserves the right of private parties to file suits on their own behalf.”78 He urged legislators to reject “a movement to strike the independent private action as a price for getting the Bill.”79 If civil rights groups were disappointed in executive performance, they most certainly were not disappointed in federal courts, which by the time of Greenberg’s testimony in 1969 had shown a strong proclivity toward expansive, proplaintiff interpretations of Title VII, issuing decisions that made civil rights leaders “jubilant.”80 With a judiciary clearly to the left of Congress on civil rights, it appeared that the federal courts were interpreting Title VII far more liberally than Congress had intended.81 Federal courts’ liberal interpretation of Title VII during this period culminated in, and was symbolized by, the Supreme Court’s controversial decision in Griggs v. Duke Power Co. in 1971,82 a case that powerfully illustrates the magnitude of the policymaking authority, exercised through the rule articulation function, that was largely shifted from the administrative state to courts in the Dirksen compromise. Griggs endorsed “disparate impact” theory under Title VII, which concerns a particular form of unintentional discrimination. It is undisputed that in passing Title VII Congress’s primary regulatory target was intentional discrimination. Title VII, however, was silent on the issue of whether it also prohibited employment practices that are “facially neutral”—do not use racial or gender classifications—and are adopted without any discriminatory intent, but that nevertheless disproportionately and negatively affect employment opportunities for racial minorities and women. Examples of such policies are ones requiring that an applicant have a high school diploma as a precondition for certain jobs, or the use of test scores to make promotion decisions. To the extent that such selection criteria are negatively correlated with racial minority or female populations, use of the criteria will disproportionately screen out those groups. Disparate impact theory holds that such negative impact on protected groups may render an employment policy, if it is not rationally justifiable, violative of Title VII even in the absence of discriminatory intent or explicit use of racial or gender classifications. Like its text, the legislative history of Title VII does not provide a clear answer to the question of whether disparate impact theory should be recognized.83 The dispute over disparate impact theory goes to the very core of the principle of equality underlying Title VII. The “equal treatment” view of Title VII looks only to whether the decision process treats different races and genders in the same way, while the “equal opportunity” view will
144
CHAPTER 5
sometimes also weigh outcomes—specifically, whether in practice racial minorities and women are less likely to succeed under the challenged facially neutral policy. The equal treatment view insists upon rigorous color and gender-blindness in making employment decisions and policies; the equal opportunity view sometimes insists upon the need for color and gender-consciousness of the effects of employment decisions and policies for racial minorities and women. The policy questions raised by the controversy over disparate impact theory are fundamental to the meaning of Title VII.84 Title VII’s private enforcement compromise self-consciously empowered courts, rather than the EEOC, as the central interpreters of the law. In Griggs, the Supreme Court exercised that power to resolve this fundamental policy choice in favor of recognizing disparate impact claims. The court held that where facially neutral selection criteria have a disparate impact on protected groups, the employer would have the burden of establishing that the criteria are “reasonably related” to performance of the job in question. Discriminatory intent would not be a necessary element for establishing Title VII liability.85 Another important legal development during this period, again highlighting judicial hospitality in Title VII’s early years, concerns rules governing class actions. In 1966, the Judicial Conference—the principal policymaking body with authority over administration of the federal courts—initiated revisions to Rule 23 of the Federal Rules of Civil Procedure, widening access to class actions, with cognizance of the important benefits these changes would provide to civil rights plaintiffs.86 Federal courts immediately applied the new Rule 23 permissively in Title VII cases, allowing extensive use of the class device.87 In individual actions, relief ordered by the court typically benefits only specifically named plaintiffs. In class actions, in contrast, named plaintiffs can seek more extensive relief to benefit entire classes of employees, potentially making litigation a more effective tool for institutional reform, and spreading the costs of litigation so as to make some claims economically feasible that otherwise would not be.88 The NAACP’s LDF used the class device to great effect in receptive courts.89 On the whole, the Republican vision of courts as bastions of restrained adjudication in which defendants would find shelter, in retrospect, looked off the mark. It appears to have been an artifact of conservative battles in the 1940s to contain New Deal administrative state-building using judicial power,90 or of conservatives’ historic view of courts as bulwarks against encroachments upon property rights.91 Moreover, civil rights advocates’ litigation successes were not mere paper victories—high-sounding judicial pronouncements that would gather dust in law libraries, with no material effects on the ground. To the contrary, in urging Congress in the 1969 Senate hearings to retain the private right of action in a hybrid enforcement regime, Greenberg
REVERBERATIONS: 1965–1976
145
recounted the facts of numerous Title VII cases in which the LDF had secured classes of African American employees broad-based access to employment opportunities previously denied them. They had secured agreement from the A&P store chain in North Carolina and South Carolina to open up previously white-only positions to African American workers; in the “much celebrated Newport News Shipbuilding case,” a large Virginia shipbuilding company submitted to a settlement in which “hundreds of Negro workers moved into craft and supervisory positions theretofore barred to them”; the Georgia State Employment Service “agreed to process the applications of Negro job seekers on the same basis as white applicants”; and these were but a few of their successes. In the Newport News Shipbuilding case, the EEOC and the Federal Office of Contract Compliance had investigated the company for years, but “there was no effective movement towards settlement of outstanding claims of racial discrimination,” explained Greenberg, “until after we filed the lawsuit.”92 Indeed, civil rights advocates believed that private lawsuits, which could shine the light on an agency’s inaction or ineffectiveness by publicly enforcing the law, might prod the agency into more effective action, thereby creating better public enforcement.93 “The entire history of the development of civil rights law is that private suits have led the way and government enforcement has followed,” Greenberg concluded, and Title VII has and should continue in that tradition.94 Greenberg spoke not just for the LDF, but for civil rights groups broadly. Joseph Rauh, appearing on behalf of the Leadership Conference and testifying immediately following Greenberg, opened by stating: “I agree without reservation . . . [with] Mr. Greenberg’s statement.”95 Representatives of the NAACP and the National Urban League also testified in favor of the hybrid private litigation / cease-and-desist model.96 Representatives of the ACLU, too, insisted that to “deprive private parties of the right to seek redress in the Federal Courts for employment discrimination under Title VII . . . would be extremely detrimental to progress in equal employment.”97 No civil rights group took a contrary position. It bears emphasis how different this enforcement vision is from the administrative state-centered model embodied in H.R. 405’s NLRB for civil rights, the onetime dream of ardent civil rights advocates, including the NAACP, the LDF, and the Leadership Conference. Their preferences would diverge yet more radically from the New Deal–style administrative vision in the years to come. This is not to say that civil rights groups had given up on effective bureaucratic implementation, for they continued to press legislators hard but unsuccessfully for EEOC cease-and-desist powers within the hybrid framework into the late 1960s and early 1970s. Whereas some individual civil rights advocates may have rejected the desirability of cease-and-
146
CHAPTER 5
desist authority by this time,98 representatives of the leading national civil rights organizations—including the Leadership Conference, the NAACP, the LDF, and the National Urban League—uniformly supported giving the EEOC cease-and-desist authority until they were finally defeated in the Equal Employment Act of 1972.99 As already noted, a for-profit bar to prosecute Title VII claims had not materialized, and nonprofit interest groups obviously lacked the capacity to prosecute ordinary claims across the nation. Civil rights groups thus wanted administrative ceaseand-desist to process the huge number of ordinary claims that interest groups could not litigate, and to retain the private right of action so as to preserve their ability to carry out impact litigation; as a potential prod to shame the agency into enforcing; and as an avenue of escape from the administrative quagmire if that should be necessary. Although they still supported cease-and-desist, their preferences regarding administrative implementation versus private litigation had been decisively transformed since 1963–64. They now wanted both and were unwilling to give up private enforcement for cease-and-desist powers. In Title VII’s early years, civil rights groups had observed an agency with neither the resources nor the inclination to aggressively pursue Title VII’s promise of fair employment. During the same period, propelled to the center of the enforcement stage by the Dirksen compromise, civil rights groups carried out the prosecutorial role to great effect, providing African Americans material job opportunities previously denied to them, including in the South. This autonomy from bureaucrats appealed to civil rights groups. “[W]ith private enforcement we were the captain of our own ship,” explained Greenberg, and “we took initiatives that more cautious government agencies wouldn’t.”100 Civil rights advocates could not, they had now decided, afford to rely solely upon the beneficence of bureaucrats, who in turn depended on elected officials for resources and power, to enforce fair employment practices. It bears noting that although conservative Republicans were pivotal in rejecting cease-and-desist and retaining a litigation-dominated framework to enforce Title VII, some had misgivings that “the class action device enabled [plaintiff-litigants] to enforce it too effectively, by greatly increasing the exposure of employers to liability for back-pay.”101 The version of the Equal Employment Act of 1972 that initially passed the House in that year, sponsored by Republican congressman John Erlenborn of Illinois, contained a provision specifically excluding Title VII plaintiffs from the right to file class actions under Rule 23 of the Federal Rules of Civil Procedure, which had proven so valuable to plaintiffs in Title VII litigation. The Senate version of the bill lacked such a provision, and in conference committee the House’s class action exclusion for Title VII cases was dropped.102 This episode reflects that, whereas antiregula-
REVERBERATIONS: 1965–1976
147
tion Republicans still preferred litigation over bureaucracy as the central implementation vehicle, some of their business constituents, like some important labor unions, were most definitely feeling its effects. The ultimate solution to the EEOC’s weakness which was adopted in the Equal Employment Act of 1972—giving the agency the power to prosecute actions in court—appears ironic when viewed from the vantage point of 1964. Stripping the agency of prosecutorial powers was the prize that Dirksen secured in the 1964 compromise; the price he paid was to create Title VII’s private enforcement regime, which would become, unbeknownst to him at the time, the single largest source of nonprisoner litigation in the federal system. In the Equal Employment Act of 1972, the EEOC got back the power that Dirksen had taken away from it, and Title VII’s private enforcement provisions, fiercely defended by civil rights groups, stayed in the law. It appears that if Dirksen had simply accepted the House-passed enforcement provisions—EEOC prosecutorial powers, private actions only with agency permission, and no recovery of attorney’s fees—the long-run outcome may have been a far weaker enforcement regime. I return to this issue at the end of the next chapter, after tracing out more of the consequences of the 1964 compromise.
The Road to the Civil Rights Attorney’s Fees Awards Act of 1976 After the 1972 amendments to Title VII, the next important development in the statutory framework governing enforcement of job discrimination laws, among other important civil rights laws, was the Civil Rights Attorney’s Fees Awards Act of 1976. A few civil rights laws had been individually amended in the early to mid-1970s to include a fee-shifting provision modeled on Titles II and VII of the 1964 act. As discussed above, the School Aid Act provided for attorney’s fees for winning plaintiffs in school desegregation cases. In 1975, a fee-shifting provision was added to the Voting Rights Act of 1965, a legislative episode discussed later in this chapter. The omnibus Civil Rights Fees Act of 1976 extended the CRA of 1964’s fee-shifting language, to cover all other existing civil rights laws that permitted private enforcement but did not allow winning plaintiffs to recover fees.103 Two Reconstruction civil rights laws thereby amended would become important sources of job discrimination litigation in particular, and were already important sources of civil rights litigation in general (and would become more so in the years to come). The Civil Rights Act of 1866 guaranteed to all persons the right that “is enjoyed by white citizens,” among others, to make and enforce contracts, which was interpreted by courts
148
CHAPTER 5
beginning around 1970 to prohibit race (and only race) discrimination in private employment.104 The Civil Rights Act of 1871, providing a cause of action against state actors for the violation of any federal rights, encompasses within its vast scope suits by state employees against state governments for violation of constitutional rights, including discrimination in the terms and conditions of employment in violation of the equal protection clause.105 Economic remedies for successful plaintiffs, in the form of uncapped compensatory and punitive damages, are more extensive under both Reconstruction laws than under Title VII’s back-pay-only rule; the Reconstruction laws have significantly longer statute of limitations periods; while Title VII only covers discrimination against “employees,” the Reconstructions laws also cover discrimination against independent contractors; and, critical to private sector employment, Title VII does not cover employers with fewer than fifteen employees, whereas the CRA of 1866 has no such threshold.106 The Reconstruction laws thus provide significant additional coverage and benefits to job discrimination plaintiffs. Other important civil rights laws covered by the Civil Rights Fees Act were Title VI of the CRA of 1964, prohibiting discrimination in any program or activity that receives federal assistance, and Title IX of the Education Amendments of 1972, prohibiting gender discrimination in any educational program or activity receiving federal assistance. In later years the Civil Rights Fees Act of 1976 would be amended to apply to subsequently passed civil rights laws, such as the Religious Freedom Restoration Act of 1993 and the civil action provisions of the Violence Against Women Act of 1994 (subsequently ruled unconstitutional).107 In contrast with debates over Title VII enforcement between 1963 and 1972, no one advocated increased executive powers or resources to administer the civil rights laws in question, and thus it was not conservative Republicans who proposed litigant mobilization to stave off bureaucratic state-building. Although civil rights advocates were disappointed and reluctant recipients in 1964 of a job discrimination enforcement regime founded on private litigation, it was they who became catalysts for building upon the private enforcement foundation that conservative Republicans had laid. Whereas by the late 1960s, civil rights groups had become steadfastly attached to a private right of action with fee shifting to enforce Title VII, by the early 1970s they mobilized to extend the range of this enforcement approach across the entire field of civil rights. We saw the first instance of this trend in the last section when civil rights liberals secured fee shifting for winning plaintiffs in the School Aid Act of 1972, which marked the beginning of a movement that would culminate in passage of the Civil Rights Fees Act of 1976.
REVERBERATIONS: 1965–1976
149
Self-reinforcing policy feedback effects emanating from the CRA of 1964 were again at play. That law profoundly transformed the interest group environment by contributing resources to civil rights groups and stimulating the growth of a for-profit civil rights bar, and together civil rights groups and the for-profit bar mobilized to expand fee shifting in the domain of civil rights. It also generated policy-learning effects concerning the efficacy of fee shifting in the civil rights context, which contributed to demand by civil rights groups for the spread of fee shifting, and to Congress’s willingness to pass new fee-shifting legislation. These feedback processes unfolded within a context of, and were propelled by, continuing conflict over civil rights policy between the Nixon administration and civil rights liberals, both in Congress and among interest groups. Growth and Mobilization of the Civil Rights Bar Political scientists have emphasized that, as Paul Pierson put it, once they are enacted policies “create resources and incentives that influence the formation and activity of social groups . . . [and] create ‘spoils’ that provide a strong motivation for beneficiaries to mobilize in favor of programmatic maintenance or expansion.”108 Attorney’s fee awards contributed resources to existing civil rights groups that prosecuted lawsuits under the new civil rights laws, such as the LDF and the Lawyers’ Committee for Civil Rights Under Law, adding to their enforcement capacity.109 The class action cases that these groups often litigated could lead to greater attorney’s fee awards than individual claims.110 By 1973, the Lawyers’ Committee was devoting half of its staff to its job discrimination litigation unit, which had become nearly self-supporting through attorney’s fee awards.111 By 1975, $550,000 of the LDF’s operating budget of $3 million (over 18 percent) came from attorney’s fee awards.112 The availability of fee awards also contributed to the formation of significant new enforcement groups, with foundation seed money, such as the Native American Rights Fund in 1970 and the Women’s Law Fund in 1972, both still in existence today, on the expectation that they would be able to draw continuing operating funds from attorney’s fees awards.113 The Lawyers’ Committee formed an Attorney’s Fees Project in the early 1970s for the purpose of compiling and disseminating to civil rights litigators information on developing case law governing fee awards, seeking to aid attorneys in recovering fees and to move the law in a direction favorable to fee awards.114 In addition to increasing enforcement resources available to civil rights groups, the private enforcement regimes of Titles II and VII of the CRA of 1964, and the numerous civil rights laws to follow that model in the ensuing decade,115 had the feedback effect of fostering the growth of a
150
CHAPTER 5
private bar to litigate civil rights claims in general and job discrimination claims in particular. By the early 1970s, it appeared that the initial impression that Title VII would not generate much private litigation was premature. From the modest 56 private suits filed in Title VII’s first two years,116 the number climbed approximately to 100 in 1968,117 to 429 in 1970, to 1,802 in 1973, and to 4,002 in 1975, where it plateaued (ranging between about 4,000 and 4,600) for the balance of the decade.118 There seems to have been a lag of about five years before for-profit lawyers began to learn about, develop skills in, and practice in this new and unknown area of law, but once they did, encouraged by proplaintiff court decisions and statutory developments in the Equal Employment Act of 1972, the growth in Title VII filings during the first half of the 1970s was dramatic, exploding by a factor of nearly ten between 1970 and 1975.119 Moreover, contrary to fears among some civil right advocates that private enforcement would rarely be initiated by racial minorities in the South because of unique obstacles there, it was disproportionately concentrated in the South relative to workforce population, and roughly proportionately concentrated there relative to nonwhite workforce population. In 1975, the eleven former Confederate states accounted for approximately 25 percent of the total American workforce, and 38 percent of the nonwhite workforce.120 They accounted for 30 percent of all job discrimination charges filed with the EEOC, 36 percent of those based upon race, and 37 percent of all privately prosecuted job discrimination lawsuits filed in federal court.121 Immediately following passage of the Civil Rights Fees Act of 1976, Mary Derfner, director of the Lawyers’ Committee’s Attorney’s Fees Project, who testified in favor of the law before Congress, observed that during the first half of the 1970s the fee-shifting provisions in recent civil rights laws helped “public interest law firms burgeon,” and that “[p]rivate practitioners and individual members of smaller commercial law firms began to undertake civil rights cases in addition to their other work.”122 “Fee awards,” she observed, “made civil rights law a financially viable practice.”123 Title VII’s fee provision in particular, explained Armand Derfner, another Lawyers’ Committee attorney who testified in favor of the Civil Rights Fees Act, had “led to the development of a highly skilled group of specialist lawyers” to enforce it.124 A 1977 report of the Ford Foundation on civil rights litigation observed that “[u]ntil at least the mid-1960s the NAACP Legal Defense and Education Fund stood almost alone” as a prosecutor of civil rights suits, but by the mid-1970s “feegenerating private practice has in many areas of the South enabled an indigenous bar, engaged in litigating cases of racial discrimination, to survive.”125 In the same vein, an April 1976 Washington Post article titled
REVERBERATIONS: 1965–1976
151
“Civil Rights Turns to Gold Lode for Southern Lawyers” declared: “The lure of legal fees, paid by the loser, is fertilizing a whole new practice in civil rights disputes. . . . Congress intended it that way in passing laws specifying that legal fees be awarded in such cases.”126 Although ascertaining what “Congress intended” is often complex and elusive, it seems doubtful that Senator Everett Dirksen intended to create a bar of lawyers who would earn their living suing employers under civil rights laws, at the expense of employers, while Humphrey probably would have been pleased. Thus far this section has been arguing that fee shifting in the new civil rights laws transformed the interest group environment by contributing monetary resources to interest groups that they had an incentive to retain and expand, and by creating a private bar with an incentive to retain and expand this new practice area, which contributed to its livelihood. The incentives faced by interest groups changed in another very important respect. From civil rights groups’ perspective, the new and growing forprofit civil rights bar represented an enforcement infrastructure of great policy value that civil rights groups wanted to retain and expand. The utility of this private enforcement infrastructure, civil rights groups believed, was rooted importantly in the failure of executive enforcement, at least partly due to willful evasion of congressional mandates. In the mid-1970s, nonprofit and for-profit civil rights lawyers together complained to congressional committees that federal agencies were failing to enforce congressional mandates. They asserted that “increasingly in recent years” there was a “wide gap between the statutes adopted by Congress and their enforcement by the Executive Branch”; that agencies had displayed “flagrant refusal to comply with the clear language and intent of Congress”;127 that “[a]gencies are not doing their jobs”;128 that the Justice Department had decided “to generally not enforce” important civil rights laws;129 and that as a result private lawyers were needed to do “the work the government should do.”130 Representatives of leading national civil rights and other law reform organizations—including the Lawyers’ Committee, the ACLU, the National Organization for Women’s Legal Defense Fund, the Council for Public Interest Law, and the Center for Law and Social Policy—urged Congress that fee shifting was a solution to this enforcement problem. Allowing civil rights plaintiffs to recover fees, they maintained, was a way to create a well-trained bar of private for-profit lawyers that would be ready, willing, and able to execute the enforcement function.131 Based upon recent experience with civil rights laws that allowed fee recovery for successful plaintiffs, civil rights groups had come to see the private for-profit bar as a resource to be leveraged, through fee-shifting rules, to
152
CHAPTER 5
supply a core law enforcement infrastructure that the state either would not or could not provide directly. During this period legal advocacy organizations argued for general fee-shifting legislation to cover not only civil rights cases, but also environmental, consumer protection, and other “public interest” litigation in general, frequently citing the CRA of 1964 as the exemplar and as the justificatory precedent for this approach to regulation.132 It should be noted that this discussion of policy feedback has been focused on fee-shifting provisions and not appointment of counsel. Title VII’s appointment-of-counsel provision, although controversial when enacted, proved to be a dog that didn’t bark. Congress never appropriated funds to pay appointed counsel and cover the extensive outlay of costs that often attend litigation, and courts proved unwilling to make compulsory appointments that would entail requiring attorneys to provide free legal services and bear litigation costs.133 One Alabama district judge concluded, more than a little ironically, that Title VII’s provision authorizing appointment of counsel, apparently without compensation, violated the Thirteenth Amendment’s prohibition of involuntary servitude.134 Courts would only seek to coordinate voluntary free representation, which had obvious limits. The fee-shifting provision, on the other hand, operating as a source of funding for civil rights litigation independent of congressional budget allocations, was proving to be a reliable mechanism through which civil rights counsel would be paid if the plaintiff prevailed. The burgeoning civil rights bar—both nonprofit and for-profit— fertilized by private rights of action coupled with fee-shifting rules in the new civil rights laws, sought to extend the CRA of 1964’s fee-shifting approach to cover all existing civil rights laws that allowed private enforcement but did not provide for fee recovery. In the early to mid-1970s, the civil rights bar pursued its goal of across-the-board fee shifting in civil rights cases on both judicial and legislative tracks. Basing their arguments explicitly on the precedent of the 1964 act, both nonprofit and for-profit civil rights lawyers pressed courts to infer the availability of attorney’s fees under civil rights statutes lacking fee-shifting provisions, including the CRAs of 1866 and 1871.135 Although this court-based campaign met with some success in the early 1970s,136 it was shut down in 1975 by the Supreme Court in Alyeska Pipeline Service Co. v. Wilderness Society,137 where the Court held that any departures from the American rule on fees would have to take the form of an explicit statutory directive from Congress. During the same period, civil rights groups had been lobbying Congress for omnibus civil rights fee-shifting legislation to statutorily extend the CRA of 1964’s fee-shifting provision to other civil rights laws lacking one.138 In response to the Alyeska decision, civil rights groups,
REVERBERATIONS: 1965–1976
153
organized and represented by NAACP lead lobbyist Clarence Mitchell, lobbied Democrats in Congress and succeeded in securing passage of the Civil Rights Fees Act of 1976.139 Lessons Learned about Fee Shifting In conjunction with the feedback effects that increased demand for civil rights fee shifting by fostering the growth of the civil rights bar and its dependence on fee awards, self-reinforcing feedback from the CRA of 1964 also operated through the mechanism of policy learning. Recall that when the Dirksen compromise was struck, liberal civil rights advocates were disappointed at the loss of an administrative framework partly because they were concerned that the fee-shifting provision would not overcome impediments to private enforcement litigation in the job discrimination context and thus would not generate a sufficient volume of suits to meet day-to-day enforcement requirements. Immediately following Title VII’s passage, for example, Richard Berg, a civil rights proponent high in the Kennedy Justice Department and soon to be EEOC general counsel, warned that private litigation was limited as an enforcement strategy. Because litigation was costly and time-consuming, and the pool of potential plaintiffs was poor, vulnerable to retaliation, and facing an adversary with far more material and legal resources, said Berg, “[E]ven where the evidence of discrimination is overwhelming, it cannot be expected that many complainants will undertake the burden of an individual suit.” Thus, he continued, “[I]t seems questionable that much can be accomplished through suits in federal court by persons aggrieved by acts of discrimination.”140 This view was shared by the Leadership Conference, the NAACP, the LDF, and the civil rights community broadly.141 Of course, civil rights advocates were well aware of civil rights groups’ capacity to shape the law through impact litigation, as the NAACP had done in the campaign leading to the momentous Brown decision, but they lacked the resources to provide day-to-day enforcement of ordinary claims across the nation. This concern, widely shared among civil rights activists, was not unfounded. In 1963, the year in which Title VII was introduced, there had been only 424 private civil rights enforcement actions across the country, under all federal civil rights statutes that allowed (or had been interpreted to allow) private enforcement, including the CRAs of 1866, 1870, 1871, 1957, and 1960, and under state civil rights laws brought in federal court via diversity jurisdiction.142 The lion’s share of these suits were brought under the Civil Rights Act of 1871’s sweeping cause of action for the violation of any federal right by a state actor.143 By 1975, as we have seen, there were about ten times that number (4,002) of job discrimination
154
CHAPTER 5
suits alone. Title VII’s private enforcement regime had succeeded in generating levels of private civil rights enforcement unprecedented in American history. The genesis and growth of the private civil rights bar, in addition to creating demand to expand fee shifting across the entire field of civil rights, also stood as evidence weighing against initial skepticism among civil rights advocates about the efficacy of fee shifting as a mechanism for generating a significant volume of private enforcement. It is this process of policy learning that led civil rights groups to regard the burgeoning for-profit civil rights bar as a new infrastructural resource to defend and expand. On this particular point the two pathways of policy feedback— the transformation of incentives facing interest groups, and policy learning—were intertwined: civil rights groups learned about the utility of a resource and then wanted to cultivate it. Legislators, like civil rights groups, after witnessing the historically unprecedented levels of private civil rights enforcement under Titles II and VII of the CRA of 1964, concluded that fee shifting had proven decisive in mobilizing private civil rights enforcers, and they weighed this result in favor of expanding the policy. According to the Senate report for the Civil Rights Fees Act of 1976: “In 1964, seeking to assure full compliance with the Civil Rights Act of that year, we authorized fee shifting for private suits establishing violations of the public accommodations and equal employment provisions. . . . These fee shifting provisions have been successful in enabling vigorous enforcement of modern civil rights legislation.”144 This policy-learning account was repeated time and again in floor statements. Senate minority leader Hugh Scott (R-PA) stated that, based upon experience with Title VII, “[a]ttorney’s fees have proven to be a singularly effective and flexible way to encourage private enforcement of public rights.”145 Senator John Tunney (D-CA), who sponsored the bill, remarked that “[a]ttorneys’ fees have proved one extremely effective way to provide . . . equal legal resources” in civil rights cases.146 Senator James Abourezk (D-SD) observed, “The fee-shifting mechanism has proved a particularly equitable and efficient means of enforcing the [civil rights] laws by enlisting private citizens as law enforcement officials.”147 The characterizations of fee shifting as “singularly flexible” and “particularly efficient” are meant in comparison to bureaucracy, a comparison that was an important theme in the debate. There was widespread agreement that the administrative state, as it was currently constituted, lacked the capacity to adequately enforce the civil rights laws in question. According to the Senate report, private enforcement of these laws was necessary because “[a]lthough some agencies of the United States have civil rights responsibilities, their authority and resources are limited.”148 “Long experience has demonstrated,” Senator Edward Kennedy (D-MA) counseled, “that government enforcement alone cannot accomplish [en-
REVERBERATIONS: 1965–1976
155
forcement goals],”149 a view echoed by sponsor Tunney, who observed, “Although some of these laws can be enforced by the Justice Department or other Federal agencies, most of the responsibility for enforcement has to rest upon private citizens.”150 The main reason was simple: money. “The Justice Department does not have the resources to bring suit for every civil rights violation,” explained Representative Elizabeth Holtzman (D-NY).151 “The Government obviously does not have the resources to investigate all possible violations of the Constitution,” concurred Representative John Seiberling (D-OH), “so a great burden falls directly on the victims to enforce their own rights.”152 Although it may have been obvious, as Seiberling suggests, that the government had not dedicated significant resources to enforcing the civil rights laws in question, it was actually not at all obvious that it was incapable of extracting and deploying those resources, particularly if one considered the possibility of far less costly administrative implementation. Rather, legislators found fee shifting to be a more attractive alternative. Fee shifting represented a way of overcoming weak administrative state capacity without engaging in conventional forms of state building. As the Senate report put it, fee shifting provided for enforcement “while at the same time limiting the growth of the enforcement bureaucracy.”153 This had been a key part of Dirksen’s motivation, and it continued to appeal to many Republicans. Successfully mobilizing private litigants, explained Hugh Scott, who succeeded Dirksen as Senate minority leader, “would make the civil rights laws almost self-enforcing, . . . would cost the government nothing,” and would advance civil rights enforcement “without an increase . . . in the bureaucracy.”154 In the same vein, Senator Charles McCurdy Mathias (R-MD), who favored passing the act, cited a Congressional Budget Office finding that the law would produce “no additional costs to the government.”155 In viewing fee shifting as a source of costless and bureaucracy-free enforcement, Democrats joined Republicans. “Fee shifting provides a mechanism which can give full effect to our civil rights laws,” explained Senator Kennedy, “at no added cost to Government.”156 “The fee-shifting mechanism,” agreed Senator Abourezk, “increases law enforcement without increasing the Federal budget or bureaucracy.”157 Although there is no evidence that budget concerns were an important cause of the private enforcement settlement of 1964, it emerged as a justification for spreading fee shifting across the field of civil rights by the mid-1970s— a justification that braided naturally with antibureaucracy sentiments. A Note on the Voting Rights Act In this discussion of the Civil Rights Fees Act, an interjection on the Voting Rights Act Amendments of 1975, whose journey through the legislative
156
CHAPTER 5
process overlapped and was entangled with that of the Civil Rights Fees Act, is both necessary and illuminating. As civil rights groups pursued omnibus fee-shifting legislation to cover existing civil rights laws that lacked such a rule, the Voting Rights Act of 1965 was one of the laws that they wished to have covered by it. However, a few months before the 1975 hearings that led to the Civil Rights Fees Act of 1976, Congress specifically addressed the issue of attorney’s fees in voting rights cases in the Voting Rights Amendments of 1975, which renewed sections of the Voting Rights Act of 1965 set to expire and added some new provisions. Because both bills were before Congress at the same time, debate over attorney’s fees in voting rights cases, important in itself, also adds insight into Congress’s motivations in passing the more general Civil Rights Fees Act.158 Though Congress had not provided an express private right of action in the Voting Rights Act of 1965, in 1969 the Supreme Court in Allen v. State Board of Elections construed the law liberally to find an “implied” private right of action for at least some kinds of significant claims under the act.159 In the Voting Rights Amendments of 1975, Congress expanded the private right to sue under the law, expressly authorizing private lawsuits under important parts of the act not addressed in Allen, and provided for attorney’s fee awards for winning plaintiffs in all voting rights cases.160 The bill leading to the law was reported out of the House Judiciary Committee, where the fee shift and expanded private right of action were added via amendments sponsored by liberal Democrat Robert Drinan of Massachusetts.161 A group of six (out of eleven) committee Republicans offered an amendment in committee to strike the provision expanding opportunities for private lawsuits—though not the attorney’s fee provision—and the proposal was defeated.162 An attempt at omnibus civil rights attorney’s fee legislation occurred in the Senate version of the Voting Rights Amendments of 1975, but it did not succeed. When the Supreme Court handed down Alyeska—the decision holding that departures from the American rule on fees must come from Congress—the bill that became the Voting Rights Amendments of 1975 already had a section providing for attorney’s fees in Voting Rights Act cases.163 In response to Alyeska, the Senate Judiciary Committee added general language providing for attorney’s fees in most civil rights cases, much like what was enacted the following year as the Civil Rights Fees Act of 1976.164 However, because of time pressures to renew provisions of the Voting Rights Act that were due to expire, these last-minute changes to the Voting Rights Amendments of 1975 were abandoned when the Senate acted on the House version of the bill, which lacked an omnibus civil rights fee-shifting provision.165 Both the House and Senate committee reports on the 1975 bills to amend the Voting Rights Act stated that the mobilization of private en-
REVERBERATIONS: 1965–1976
157
forcers through fee shifting was indispensable to achieving congressional mandates expressed in that law.166 According to the hearing testimony of a number of proponents of the amendments, this need arose partly from inadequate Justice Department resources.167 However, echoing a now familiar theme, in the hearings Democrats and civil rights groups argued that the mobilization of private enforcers was also, and more fundamentally, needed because the Nixon/Ford administration’s enforcement of the Voting Rights Act had been willfully weak. This charge was not new. It had been the sole focus of the inaugural hearings of the House Civil Rights Oversight Subcommittee in 1971, where Attorney General John Mitchell was accused of seeking to “emasculate” and “de facto repeal” key provisions of the Voting Rights Act through crabbed interpretation and nonenforcement, with the motive of “appeasement of racist elements” and “seeking to woo white political support.”168 Conflict over voting rights enforcement had been brewing through the first half of the 1970s. Testifying in the 1975 hearings, Howard Glickstein, who by then had become director of the nonprofit Center for Civil Rights, saw deep irony in claims that the Justice Department lacked the resources to vigorously enforce voting rights laws, and in the transformation over the last two decades in expectations about executive power over voting rights enforcement. He had seen the arc of this transformation, from witnessing passage of the Civil Rights Act of 1957 as a young lawyer, through his years in the Civil Rights Division under Kennedy and Johnson, and then while working for the U.S. Commission on Civil Rights from 1965 to 1972, first as general counsel and then as staff director appointed by Johnson.169 Said Glickstein in his congressional testimony: It is interesting the way history has been turned upside down. In 1957, when Congress passed the 1957 Civil Rights Act, and the Attorney General was allowed to bring voting rights suits, everyone involved thought a millennium had arrived. Finally, the massive weight of the Federal Government was going to be injected in the picture, and we could all sit back and be confident that voting rights would be protected. Now, some 20 years later, we are saying that the Federal Government is so poor and so impoverished that it needs the assistance of private individuals to insure that voting rights are protected. . . . I think that this has a lot to say about the resources that our Government is devoting to the enforcement of the very fine civil rights laws that Congress has enacted.170 The last sentence makes evident that, in Glickstein’s view, the American state’s weak enforcement of voting rights laws had less to do with lack of resources than with the disinclination of pivotal state actors to mobilize
158
CHAPTER 5
state resources to that end. The problem, in his estimation, lay in executive choices about the allocation and deployment of resources: he stated that Nixon’s Department of Justice was characterized by “lack of enthusiasm” and “reluctance” as an enforcer of the Voting Rights Act, and had been for years, and he supported fee shifting to mobilize needed private enforcement.171 John V. Tunney, chair of the Senate Judiciary subcommittee that held hearings on the bill—and the senator who would introduce the Civil Rights Fees Act less than four months later172—made this claim repeatedly. He suggested that private enforcement with fee shifting was necessary “because the Attorney General has not . . . exercised the diligence that he should have in enforcing the provisions of the Voting Rights Act.”173 Citing findings by the U.S. Commission on Civil Rights, which recommended an expanded private right of action and fee shifting in Voting Rights Act cases, he stated that “the Department of Justice is not doing an adequate job in attacking discriminatory practices” that violate the act.174 He clearly did not believe that the problem was simply one of insufficient Justice Department resources, stating that he “was not satisfied with the Justice Department’s excuses” for its enforcement failures.175 In the subsequent floor debates liberal Democrats likewise argued that the expanded private right of action and fee shift were imperative because the existing approach to enforcement of voting rights was inadequate. Important sections of the law were not being enforced by the attorney general, and thus the amendments were “intended to beef up enforcement.”176 In urging attorney’s fee awards for winning plaintiffs in voting rights cases, renowned southern ACLU lawyer Charles Morgan, Jr., who had been plaintiffs’ counsel in the landmark Reynolds v. Sims voting rights case (a cornerstone of the “one person, one vote” doctrine), proffered an even harsher assessment of the Nixon administration’s enforcement record in his committee hearing testimony. Echoing charges leveled in the 1971 House oversight hearings on enforcement of the Voting Rights Act, he claimed that as part of Nixon’s “Southern strategy” his Justice Department had ratified “discriminatory” voting rules for southern states and had decided “to generally not enforce the law.” He continued: “The Justice Department has lawyers that run around over there working on things that do not guarantee citizens’ constitutional rights. It is an obscenity.” Morgan looked back wistfully on the Kennedy administration’s Civil Rights Division, which, he said, despite its modest resources, was “fighting the crusades” alongside civil rights groups, a time he characterized as the “good old days.”177 The Lawyers’ Committee for Civil Rights Under Law had three representatives testify in support of private lawsuits with fee shifting to ob-
REVERBERATIONS: 1965–1976
159
viate dependence on executive enforcement. These witnesses were leading southern civil rights lawyers Frank Parker and Armand Derfner, and Nicolas Katzenbach, deputy attorney general under Kennedy and attorney general under Johnson who helped draft the CRA of 1964 and the Voting Rights Act of 1965. He had become a member of the Executive Committee of the Lawyers’ Committee.178 Like the ACLU, the Lawyers’ Committee was deeply dubious of the Nixon administration as a civil rights enforcer. As the hearings transpired, the Lawyers’ Committee was among counsel prosecuting an ongoing lawsuit against Nixon’s attorney general for failure to enforce the Voting Rights Act (the attorney general position passed from Richard Kleindienst to Elliot Richardson to William Saxbe to Edward Levi during the course of the litigation). Derfner, a stalwart champion of private civil rights lawsuits with fee shifting starting in the early 1970s, was counsel of record on behalf of the Lawyers’ Committee.179 The suit alleged that the attorney general was failing to perform mandatory duties under the Voting Rights Act when he formally declined to issue a ruling, in response to a petition by a plaintiff-class of African American voters, on whether a redistricting plan recently adopted in South Carolina was racially discriminatory in violation of the act. By declining to rule, the attorney general let stand a decision from a South Carolina district judge affirming the apportionment plan alleged to be discriminatory by the plaintiffs. The plaintiff-class of African American voters won before the district court in 1973 and before the U.S. Court of Appeals for the District of Columbia in 1975, both of which held that the attorney general’s inaction was illegal and ordered that he act upon the petition.180 Thus, at the same time that Lawyers’ Committee attorneys testified in favor of incentives for private enforcement, they were suing the Nixon administration for non-enforcement. The litigation makes abundantly clear that, once again, some key supporters of expanding private civil rights lawsuits with fee shifting profoundly distrusted the Nixon administration as a civil rights enforcer, and regarded it as openly hostile to their agenda. In the hearings, the ACLU’s Morgan argued that Congress must find ways to undertake serious oversight of the Civil Rights Division’s enforcement of the Voting Rights Act, such as by legislatively requiring that it submit reports to Congress every six months detailing its Voting Rights Act enforcement activities. But oversight alone would not be sufficient. He offered the following counsel about civil rights enforcement: “Citizens should rely upon public officials to do the job for them,” but “[w]hen they do not, citizens should do it for themselves,”
160
CHAPTER 5
and attorney’s fee awards are a “mighty good incentive” to assure that the for-profit bar will spring into action when the latter course becomes necessary.181 Roll Calls and Voting Alignments on Private Enforcement Bills leading to the Civil Rights Fees Act of 1976 were introduced by civil rights liberals and Democrats in both chambers: John Tunney of California introduced it in the Senate in August 1975 (the week before the Voting Rights Amendments were passed), and Richard Bolling of Missouri did so in the House in October 1976.182 The only dissenting voices among those who spoke on the floor of Congress when the bill was debated were a few Southerners. Senator James Allen (D-AL) led a seven-day filibuster. His principal argument against the bill was that it was “stirring up litigation” and was a “bonanza for the legal profession . . . [that has] grown fat on litigation of this sort.”183 Senator Jesse Helms (R-NC) also urged his colleagues to vote against the bill. Relying upon statistical data from the federal court system, he complained that civil rights litigation in general, and job discrimination litigation in particular, had swelled to an excessive share of the federal court caseload, which he urged should not be increased even further with the proposed Civil Rights Fees Act of 1976.184 Notwithstanding Allen’s and Helm’s objections, partisan and sectional divisions over the Civil Rights Fees Act, which actually commanded a majority of southern votes, were muted relative to the partisan rancor that would infuse debates over civil rights litigation in the 1980s, taken up in the next chapter. The voting alignments looked much as they had in the roll calls on the private enforcement provisions in the CRA of 1964, with the exception that southern opposition was greatly mitigated. In addition to the House and Senate roll calls on passage of the Civil Rights Fees Act,185 there were seven roll calls on amendments offered to reduce the private litigant-mobilizing effects of the law (all of which were offered by Southerners): the Helms amendments to exempt state and local governments from the bill,186 to provide that attorney’s fees could only be awarded against a party that behaved with bad faith,187 and to eliminate the Civil Rights Act of 1866 from coverage by the bill;188 the Thurmond (R-SC) amendments to prohibit application of the act to busing cases,189 and to provide for payment of attorney’s fees by plaintiffs to defendants if the court found the plaintiff’s action to be frivolous;190 and the Allen amendments to restrict fee awards to circumstances in which the defendant acted in a “contumacious or vexatious manner,”191 and to eliminate the Civil Rights Act of 1866 from coverage by the bill.192 There were an additional three roll calls offered to reduce, retrospectively, the litigant-mobilizing effects of existing civil rights laws (all offered
REVERBERATIONS: 1965–1976
161
by Southerners): the Allen amendment to bar the award of attorney’s fees under the Equal Employment Act of 1972;193 the Scott amendment to bar the award of attorney’s fees against federal or state governments under Titles II or VII of the CRA of 1964, or the Voting Rights Act of 1965;194 and the Helms amendment to create a mandatory award of attorney’s fees to prevailing defendants under Titles II and VII of the CRA of 1964, the Voting Rights Act of 1965, and the Reconstruction civil rights laws.195 Finally, there was one roll call offered to expand the scope of the Civil Rights Fees Act’s application: the Kennedy amendment to expand application of the act to sex or blindness discrimination in educational institutions receiving federal funds.196 All proposed amendments to reduce litigant mobilization under the Civil Rights Fees Act of 1976 were defeated. The roll calls—many quite similar to ones unsuccessfully offered by Southerners in the Senate in 1964—show that, as with debates over enforcement of the CRA of 1964, the Fair Housing Act of 1968, the proposed amendments to Title VII leading to the Equal Employment Act of 1972, the School Aid Act of 1972, and the Voting Rights Amendments of 1975, once again in 1976 the use of statutory provisions calculated to mobilize private enforcement litigation was neither unreflectively routine nor uncontested. It was a selfconscious choice, motivated by policy goals, and subjected to contestation in the legislative process. If we aggregate all votes cast on the Civil Rights Fees Act of 1976 and proposed amendments to it across both chambers, nonsouthern Democrats voted in a direction favorable to mobilizing private litigants at a rate of 97 percent, nonsouthern Republicans did so at a rate of 68 percent, and representatives from the former Confederacy did so at a rate of 40 percent. On passage of the Civil Rights Fees Act of 1976, nonsouthern Democrats voted in favor at a rate of 96 percent, nonsouthern Republicans did so at a rate of 73 percent, and legislators from the former Confederacy did so at a rate of 54 percent (a majority of Southerners in both chambers). In the 1976 legislative proceedings in the Senate, as noted above, Southerners offered two amendments to the Civil Rights Fees Act to retrospectively amend Titles II and VII of the CRA of 1964 to limit plaintiffs’ ability to recover attorney’s fees and to provide for mandatory fee awards for winning defendants, giving legislators an opportunity to revisit some dimensions of the 1964 fee-shifting provisions to make them less favorable to plaintiffs. Southerners voted in favor of these changes to the CRA of 1964 at a rate of 60 percent, whereas nonsouthern Democrats voted against them at a rate of 100 percent, and nonsouthern Republicans voted against them at a rate of 72 percent.197 The Civil Rights Fees Act of 1976, extending the CRA of 1964’s fee-shifting language to other existing civil rights laws that allowed private enforcement but lacked a
162
CHAPTER 5
fee-shifting provision, passed overwhelmingly, by a vote of 306 to 68 in the House and 57 to 15 in the Senate.198 Although civil rights groups and liberal Democrats in Congress were unquestionably the impetus behind getting the bill on the agenda, it encountered little opposition outside the South and passed by a very wide margin. It bears emphasis here that, alongside the dramatic changes in civil rights groups’ preferences in favor of private enforcement since 1964, the Republican party’s position appears to have changed little. In the 1964 roll call votes on southern amendments offered to cut back the private enforcement provisions of Titles II and VII, nonsouthern Republicans had voted to sustain the private enforcement provisions at a rate of 79 percent, and in 1976 they voted to reaffirm, ratify, and extend this private enforcement bargain at a rate of about 73 percent. As traced in this chapter, by 1976 a fairly clear picture had emerged, from civil rights liberals’ point of view, of the relative efficacy of Title VII’s private enforcement regime. Private lawsuits under Title VII had proven to be an important weapon against discriminating employers, and had increased about tenfold in the first half of the 1970s. The targets of Title VII were beginning to feel the negative effects, from their standpoint, of private lawsuits. In 1972, House Republicans sought unsuccessfully to eliminate class actions under Title VII so as to curtail what they regarded as excessive back-pay exposure for employers. In the floor debates on the Civil Rights Fees Act, Southerners not only complained of excessive civil rights litigation but also cited hard statistics demonstrating its steep pattern of growth on the federal civil docket. This mobilization of private civil rights enforcers, and the pain they were causing defendants, of course, was an important motivation behind civil rights groups’ decision to lobby for the Civil Rights Fees Act of 1976. Despite all this, nearly three quarters of nonsouthern Republicans voted to protect the CRA of 1964’s private enforcement provisions, and to extend them to other civil rights laws, vigorously echoing Dirksen’s preference for private lawsuits over administrative state-building. Ironically, on this point liberal and conservative preferences had converged, each side deeply distrustful of administrative power, though for very different reasons. Interest Groups on the Scene In congressional hearings leading to the extension of the CRA of 1964’s fee-shifting approach in the mid-1970s, the relative presence of lawyers increased sharply as compared to 1963, reflecting the growth of civil rights groups specifically dedicated to legal advocacy as well as the forprofit civil rights bar, and the support of both for fee shifting in civil rights litigation. In the 1971 hearings on the School Aid Act, however,
REVERBERATIONS: 1965–1976
163
no lawyers’ associations were invited to testify, nor were any private forprofit legal practitioners, nor were any representatives of specifically legal advocacy organizations, though the ACLU, the National Legal Aid and Defender Association, and several law school deans did submit letters to committee members supporting fee shifting in school desegregation cases. Among nongovernmental witnesses in the hearings, civil rights and education groups (with general, rather than legal advocacy, missions) were the dominant presence and the main advocates of fee shifting.199 In the 1975 hearings leading to an expansion of opportunities for private litigation and the introduction of fee shifting in voting rights cases, among nongovernmental witnesses civil rights groups were again dominant, but this time a significant share of these groups were specifically legal advocacy organizations, many of recent vintage. Of twenty-five issue group appearances, nine were by legal advocacy organizations, four of which had been founded since the CRA of 1964, and these groups supported fee shifting. In these hearings, like those on the School Aid Act, not a single lawyers’ association representative or for-profit sector attorney appeared.200 The 1973 and 1975 hearings on fee shifting leading to the Civil Rights Fees Act of 1976, unlike hearings on the School Aid Act and Voting Rights Amendments, focused centrally on the issue of fee shifting. The policy areas examined included civil rights, but also extended to consumer, environmental, and general “public interest” litigation. Of twentyfour nongovernmental witnesses to appear and testify on the issue, fourteen represented legal advocacy groups (e.g. the Lawyers’ Committee for Civil Rights Under Law), two were lawyers who represented more general citizen’s groups (e.g., Consumers Union), one represented a lawyers association (the Association of Trial Lawyers of America), and five were for-profit lawyers testifying as practitioner experts. They all testified in favor of a rule allowing winning plaintiffs to recover attorney’s fees. A representative of the National Association of Manufacturers was the lone witness against fee shifting, though he only opposed fee awards against private (as opposed to governmental) defendants.201 The one common denominator on the interest group front in hearings on the School Aid Act, the Voting Rights Amendments, and the Civil Rights Fees Act, is that civil rights groups were on the scene advocating for attorney’s fee awards for winning plaintiffs.
Implications of the Process-Tracing Evidence I next assess how the process-tracing evidence in this chapter fits with the theories and evidence presented previously in the book. The purpose of the process-tracing evidence is to provide an opportunity to scrutinize
164
CHAPTER 5
the causal processes that produced the outcome of an enforcement regime dominated by private litigation; to appraise how well those causal processes support the multiple theoretical accounts of legislative enactment of private enforcement regimes discussed in chapters 2 and 3; and to evaluate how consistent those causal processes are with the findings of the empirical model presented in chapter 3. Legislative-Executive Conflict. The process-tracing evidence in this chapter supports the legislative-executive conflict hypothesis, which maintains that conflict between Congress and the president over control of the administrative apparatus can create incentives for Congress to enact private enforcement regimes. Differences in congressional and presidential preferences on civil rights policy were again at play in encouraging legislative reliance upon private enforcement, consistent both with the significance of the legislative-executive conflict variables in the empirical models in chapter 3, and with the historical evidence in the last chapter. With the move from unified to divided government when Nixon replaced Johnson as president, partisan divisions between liberal Democrats and the Nixon administration fueled congressional Democrats’ commitment to a private enforcement alternative to executive power. Evidence of these partisan policy differences can be seen in repeated and explicit claims by Democrats that Nixon was willfully neglecting civil rights enforcement, dramatically illustrated by Mondale’s declaration that there existed an “enormous gaping law enforcement crisis” that congressional oversight efforts had failed to correct, and that private lawsuits were needed to rectify. It can be seen in repeated and explicit charges by civil rights groups—key Democratic party constituents invited by Democrats to testify in hearings—that the executive apparatus under Nixon was neglecting and “mangling” civil rights laws to the point that it was “an obscenity,” and in such groups’ multiple lawsuits against that apparatus asserting failure to carry out enforcement responsibilities mandated by Congress. It can be seen in Johnson-appointed EEOC chair Clifford Alexander’s denunciation of the Nixon administration’s “crippling lack of support” for the agency; and in the majority-Democratic appointee Civil Rights Commission’s harsh attacks on Nixon’s civil rights enforcement record, along with its staff director Howard Glickstein’s recommendation that Congress find “a mechanism to by-pass unwilling agencies” in order to achieve enforcement goals. And so Congress did, extending fee awards for winning plaintiffs first to desegregation cases in 1972, and then to voting rights cases in 1975, and then across the board in the Civil Rights Fees Act of 1976. Whereas in the last chapter we observed congressional Republicans turning to private lawsuits motivated partly by
REVERBERATIONS: 1965–1976
165
distrust of executive power under Democratic supervision, once Nixon assumed office it was congressional Democrats, wary of a Republican executive, who turned to private lawsuits as an alternative. Judicial Ideology. Judicial ideology, while insignificant in the empirical model in chapter 3, appears important in the process-tracing evidence in this chapter, as it did in the last. In chapter 2 it was observed that, from the standpoint of theory, congruence of legislative and judicial policy preferences should make private enforcement regimes more attractive to Congress, other things being equal (the judicial friendliness hypothesis). It was also observed that increasing judicial hostility to congressional preferences can be counteracted, with respect to its effects on litigant mobilization, by enactment of ever more robust private enforcement regimes (the judicial hostility hypothesis). By the late 1960s, judicial friendliness, in the estimation of civil rights advocates, encouraged them to pursue the entrenchment and spread of opportunities and incentives for private enforcement litigation. At the very time that the executive branch moved rightward on civil rights with Nixon’s assumption of office, the federal courts were issuing far more liberal interpretations of Title VII—exercising their rule articulation function—than most observers had expected. Civil rights groups were jubilant at the judiciary’s substantive elaboration of Title VII doctrine, symbolized during this period by the Supreme Court’s endorsement of “disparate impact” theory in the 1971 Griggs decision. It is little wonder that they began to defend fiercely the right to litigate that had been imposed upon them, against their will, in 1964. In that year, antiregulation Republicans believed that courts would be a more ideologically friendly enforcement venue than an NLRB-style agency, encouraging their support for court-based implementation. By the late 1960s, civil rights advocates judged that federal courts were actually their ideological allies, which fueled their preference for implementation through private enforcement regimes. Veto Points and Compromise. Some of the evidence supports the veto points and compromise hypothesis, which maintains that the many veto points in the American lawmaking process can encourage private enforcement regimes as a compromise alternative to bureaucratic statebuilding. In the legislative events leading to passage of the Equal Employment Act of 1972, the veto-point-ridden character of the American lawmaking process again contributed to the defeat of a cease-and-desist adjudicatory agency in favor of an enforcement regime dominated by private litigation. In addition to the filibuster, which had been decisive
166
CHAPTER 5
in shaping Title VII’s enforcement provisions in 1964, also critical this time were bicameralism and the committee system. In 1965, EEOC ceaseand-desist powers passed the House but died in the Senate; in 1970, they passed the Senate but were extinguished in the House Rules Committee. The Senate bill that became the Equal Employment Act of 1972 had contained EEOC cease-and-desist powers, but they were eliminated in a bargain in exchange for the cloture votes of conservative Republicans needed to end a filibuster. The story of the School Aid Act also reinforces the compromise nature of fee shifting as a regulatory strategy. Democrats were prepared to allocate substantial public funds to pay private counsel to prosecute school desegregation cases. Republicans were unwilling to go along with this appropriation of funds. The parties compromised on fee awards for winning plaintiffs to be paid by school districts. Policy Drift. The coalition drift hypothesis holds that current majority coalitions in Congress enact private enforcement regimes to insulate enforcement of regulatory laws from subversion by future majority coalitions. The empirical model in chapter 3 supported this hypothesis, finding impending seat share losses by the majority party (electoral uncertainty) to be significantly associated with enactment of private enforcement regimes. The historical evidence in the 1965 to 1976 period, like the evidence surrounding passage of Title VII, did not support the coalition drift hypothesis. It did not reflect that civil rights advocates, when mobilizing to protect Title VII’s private enforcement regime and to extend it to other civil rights laws, were motivated by concern about insulating enforcement from more conservative and less interventionist future legislative majorities. The bureaucratic drift hypothesis holds that current majority coalitions in Congress enact private enforcement regimes to insulate policy from subversion by bureaucrats pursuing their own personal preferences, conceived broadly to include not only bureaucrats’ ideological preferences, but also timidity, apathy, shirking, careerism, and capture. While the bureaucratic drift hypothesis was not susceptible to straightforward testing in the statistical model, the model’s finding of a positive correlation between the presence of issue groups in congressional hearings and enactment of private enforcement regimes suggested the possible significance of bureaucratic drift. Whether this inference can be made is wholly dependent upon whether the claim of the issue group hypothesis about the motivations of issue groups is borne out. That hypothesis theorizes that issue groups in the middle to late 1960s began to advocate private enforcement regimes as an alternative to dependence on administrative power because of concern about bureaucratic drift. It is only with process-tracing evidence that we can assess whether this account of
REVERBERATIONS: 1965–1976
167
interest group preferences, perceptions, and intentions is in fact at play. The evidence in this chapter powerfully confirms that issue groups in the 1965–76 period were motivated, in episodes of successful advocacy for private enforcement regimes, partly by fear of bureaucratic drift. That evidence is discussed immediately below. Issue Groups. The evidence in this chapter repeatedly showed civil rights groups successfully lobbying and participating in the legislative process with the goal of entrenching and expanding the role of private lawsuits in civil rights enforcement. The evidence also showed that some of their motives mapped closely to those identified by the issue group hypothesis: such groups increasingly lost faith that the administrative state would prove an effective civil rights enforcer. To civil rights group leaders, Title VII enforcement appeared to be a very low executive priority even under Johnson, and the EEOC’s efforts appeared not only underfunded, but also torpid, lacking in ambition, and potentially compromised by the powerful interests that the law was supposed to regulate. In short, their vision of executive power as vigorous and potent—once animated by optimism about executive power with roots in the New Deal—had turned to dejection with their observation of bureaucratic drift toward passivity and stagnation. Civil rights groups had other important motives as well that were outside those contemplated by the issue group hypothesis, and that were significant to their advocacy for private enforcement regimes. It is important to emphasize that the issue group hypothesis concerns groups’ preference for private enforcement regimes due to a generalized distrust of bureaucracy as inherently limited as an implementation vehicle, regardless of who controls it. In this chapter we also observed civil rights groups’ particularized distrust of an existing executive administration as ideologically hostile to their policy agenda, a matter quite separate from inherent limits of bureaucracy. The evidence in the chapter repeatedly reflected civil rights activists’ complaints that the Nixon-Ford administration refused to use executive power to enforce civil rights because, following its own policy agenda, it did not want to. In this sense the operation of the issue group story was importantly braided with the operation of the legislative-executive conflict story. Civil rights groups were an important part of the Democratic party coalition that controlled Congress. Together with congressional Democrats, they faced a Nixon-Ford White House, and the coalition’s distrust and antipathy toward it fueled their collaboration to ratchet up private civil rights enforcement. Moreover, the historical evidence revealed how civil rights groups’ enforcement preferences developed. It showed that their transformation away from bureaucracy and toward private enforcement regimes flowed
168
CHAPTER 5
as self-reinforcing policy feedback effects from the 1964 compromise. They learned in the middle to late 1960s that enforcement vigor could not be expected from the executive branch, at least not to the extent that they hoped, for reasons both fiscal and political. They learned during the same period of their capacity to act as potent enforcers in ideologically hospitable courts. They came to recognize, in the early to mid-1970s, the ability of fee shifting to fund groups’ impact litigation, and, contrary to their firm expectations, to overcome many of the historic obstacles to achieving high levels of enforcement by a private, for-profit civil rights bar, which they now came to regard as an enforcement infrastructure of great value to be cultivated. The net result of these self-reinforcing feedback effects is that civil rights groups mobilized to entrench and extend this form of implementation across the entire field of civil rights. The evidence in this chapter suggests two amendments to the logic of the issue group hypothesis as applied to the civil rights context. First, groups’ growing disenchantment with bureaucracy during this period was rooted not only in convictions about bureaucracy’s inherent limitations as an implementation vehicle, but also in the emergence of conservative Republican control of the White House, which would be a relatively durable feature of American politics for decades to come. Second, while civil right groups’ growing distrust of bureaucracy was definitely an important source of their demand for private enforcement regimes, their affirmative optimism about private enforcement regimes as a vehicle to achieve policy goals was a very important source as well, both because the privatization and incentivization of the prosecutorial function had proven effective in mobilizing vigorous enforcement, and because courts were providing expansive interpretations of Title VII during this period. It is a story not only of interest group repulsion (from bureaucracy), but also one of interest group attraction (to litigation and courts). Rent-Seeking Lawyers. The rent-seeking lawyer hypothesis—which holds that demand by lawyers pursuing money is an important reason that Congress enacts private enforcement regimes—is not well supported by the historical evidence, though some of that evidence does reveal pursuit of fee-shifting rules in the civil rights context by for-profit lawyers. The empirical model in chapter 3 found no relationship between the presence of lawyers’ associations in congressional hearings and enactment of private enforcement regimes, and the historical evidence surrounding passage of Title VII likewise provided no support for the hypothesis. Of the three laws to expand civil rights fee shifting examined in this chapter, the for-profit civil rights bar was absent from the legislative process leading to the School Aid Act of 1972 and the Voting Rights Amend-
REVERBERATIONS: 1965–1976
169
ments of 1975, where civil rights and other issue groups appeared to be the key source of interest group support for fee shifting. In the mobilization leading to the Civil Rights Fees Act of 1976, however, the for-profit civil rights bar did enter the demand side of the picture. The availability of fee recovery for some kinds of civil rights claims starting with the CRA of 1964 clearly created the existence of a for-profit civil rights bar and its powerful growth in the first half of the 1970s. Acting in conjunction with civil rights groups, that bar was part of the mobilization before courts and Congress for across-the-board civil rights fee shifting, a campaign that ultimately failed before courts and succeeded legislatively in the Civil Rights Fees Act of 1976. While it is difficult to assess the causal importance of the role played by the for-profit civil rights bar in that legislative outcome, it is absolutely clear that nonprofit civil rights groups were the catalysts and agenda setters in defending Title VII’s private enforcement provisions in the late 1960s and early 1970s, and in securing legislative extension of fee shifting to other civil rights laws during the first half of the 1970s. Passage of the Civil Rights Fees Act of 1976 was identified as a top legislative agenda item by the NAACP, in collaboration with other civil rights groups, and the bill’s introduction in Congress was initiated by Clarence Mitchell, the NAACP’s renowned lead lobbyist. It may be that the for-profit civil rights bar added weight and power to civil rights groups’ lobbying efforts, and it may be that this proved decisive in securing passage of the Civil Rights Fees Act of 1976. Believing the latter to be true would entail the historical counterfactual claim that civil rights groups would have failed to obtain the legislation without the aid of the for-profit civil rights bar, a claim that strikes me as doubtful based upon the historical evidence, though one that cannot be decisively ruled out. Even if the for-profit civil rights bar was an indispensable component of the winning coalition, civil rights groups were undoubtedly the architects and leaders of the policy campaign. If rent-seeking activity by the for-profit civil rights bar—conjured into being by the CRA of 1964’s private enforcement model—did play a causal role, it was by bolstering the efficacy of civil rights groups in the legislative process. Party Alignments. The weak form of the party alignment hypothesis is supported by the historical evidence, and the strong form of it is rejected. This is consistent with findings presented in chapter 3, where the degree of Democratic control of Congress was significantly associated with enactment of private enforcement regimes in two of three empirical models, though far from a dominant factor, and none of the other independent variables was conditional upon it. The weak form of the hypothesis pre-
170
CHAPTER 5
dicts that Democrats will be more likely to support private enforcement regimes than Republicans, and the strong form predicts consistent Republican opposition. Democratically controlled committees reported the bills extending fee awards to winning plaintiffs in school desegregation cases, and then voting rights cases, and then civil rights cases across the board. In the School Aid Act, Republicans stripped out the public fund to pay the attorney’s fees of winning plaintiffs, and subsequently inserted the fee-shifting provision requiring defendants to pay them. In roll calls that isolated the choice to expand civil rights fee shifting in the School Aid Act (in the Senate) and the Civil Rights Fees Act (in both chambers), nonsouthern Democrats voted overwhelmingly in favor, and did so at a markedly higher rate than Republicans. Nonsouthern Republicans, though, also voted in favor at a rate of 62 percent and 73 percent, respectively. While Democrats proposed the expansions of attorney’s fee awards in civil rights cases, and were their keenest advocates, the new rules passed with broad Republican support. It is also quite clear why Democrats played the role of proposer and advocate: the laws were sought by leading national civil rights groups, which were long-standing and important Democratic party constituents. Budget Constraints. Evidence also emerged in support of the budget constraint hypothesis during debates over the School Aid Act of 1972 and the Civil Rights Fees Act of 1976, though in a qualified sense. The empirical model in chapter 3 did not discern any relationship between budget deficits versus surpluses and enactment of private enforcement regimes, and the historical evidence surrounding passage of Title VII also provided no support for the budget constraint hypothesis. In contrast, in floor debates over the School Aid Act of 1972 and the Civil Rights Fees Act of 1976, budgetary considerations were invoked by some supporters of fee shifting. In the case of the School Aid Act, in the Senate nonsouthern Democrats strongly supported creating a public fund to pay attorney’s fee awards for winning private prosecutors of school desegregation cases, an unusual proposal motivated by a desire to insulate school funds from compulsory court orders to pay fee awards. A solid majority of nonsouthern Senate Republicans opposed the appropriation of federal funds for this purpose, arguing that it would impose an unfair burden on taxpayers. At the same time, a solid majority of them supported attorney’s fee awards for winning plaintiffs in desegregation cases to be paid by discriminating school districts. These Republicans wanted to increase enforcement, but they did not want to spend federal dollars on it. Nonsouthern Democratic motivations to support fee shifting, though, were clearly not budgetary at all:
REVERBERATIONS: 1965–1976
171
they were willing to commit substantial federal dollars to enforcement, but thought that it would be more effectively spent on fee awards than on Nixon’s Justice Department. Thus, in the School Aid Act budgetary concerns had a partisan cast; they were visible only on the Republican side. Recall also that while courts ruled that they would not appoint counsel under Title VII’s appointment-of-counsel provision unless Congress appropriated funds to pay them, no public fund was created for this purpose. In debates over the Civil Rights Fees Act of 1976, there was a bipartisan chorus praising private lawsuits with attorney’s fee awards for winning plaintiffs as a highly efficient way to regulate without adding expenditures to the federal budget. One should take note, however, that the Democratic legislative catalysts, and their civil rights group constituents, behind spreading fee awards to school desegregation and voting rights leading up to the Civil Rights Fees Act, had clearly and repeatedly charged willful failure of executive enforcement in the Nixon and Ford years. This is not a problem that more funds would be likely to solve. From their point of view, emphasizing budgetary efficiency—even if it was not their main motivation—was an effective way to forge a coalition that included many conservative Republicans. It should further be stressed that in the Civil Rights Fees Act of 1976 budgetary considerations did not motivate the choice of private enforcement regimes instead of more expensive bureaucratic state-building. Unlike in debates leading to passage of the CRA of 1964, the FHA of 1968, and the Equal Employment Act of 1972, no one advocated bureaucratic state-building as a possible alternative to private litigation. Rather, given that the civil rights laws in question already allowed private enforcement litigation, the issue under consideration was whether to bolster their existing private enforcement regimes with fee shifting. It was in this context that some legislators identified lack of direct governmental expenditures—they ignored governmental expenditures on staffing the judiciary—as a virtue of fee shifting and as a reason to vote for it.
Chapter 6 ESCALATION The Civil Rights Act of 1991
On the face of it, the Civil Rights Act of 1991 was an odd law. It was passed in an era in which lawsuits were getting an especially bad reputation. It was the same year that Walter Olson published his first book-length polemic against lawsuits, The Litigation Explosion: What Happened When America Unleashed the Lawsuit.1 While the bills that would become the CRA of 1991 were being debated in Congress, Olsen’s book received a glowing review titled “Too Many Lawyers, Too Many Lawsuits,” by former chief justice of the Supreme Court Warren Burger, who had lately become an outspoken advocate for reforms to reduce the use of litigation in the implementation of policy.2 Despite this atmosphere of antipathy toward litigation, in the CRA of 1991 Congress increased the economic value of Title VII claims, making available extensive new monetary damages, manifestly for the purpose of increasing litigation under the law. Why would Congress purposefully increase Title VII lawsuits, already the target of complaints about excessive litigiousness? The answer is grounded in interbranch ideological conflict over civil rights throughout the 1980s. Mainly it occurred between a predominantly Democratic Congress and the Reagan administration over civil rights policy in general, and over control of the EEOC in particular. It also occurred, as the decade wore on, between Congress and a federal judiciary that increasingly, through the appointment of new judges, reflected the administration’s position on civil rights. Whereas the executive and judicial branches were clearly to the left of Congress on civil rights policy when the CRA of 1964 was passed, they migrated rightward in the 1970s, and then swung more sharply to Congress’s right on civil rights policy during the Reagan and Bush years.3 This polarizing interbranch realignment in the 1980s on civil rights policy produced extensive conflict between Congress and both the executive and judicial branches, conflict that grew more intense in the latter half of the decade. In the summer of 1989, a new conservative majority on the Supreme Court issued a series of five decisions leveling a frontal assault on Title VII’s private enforcement regime, and one clearly intended to curtail private enforcement levels. A swift and vigorous countermobilization by civil rights groups and their allies in Congress not only overrode most of
CIVIL RIGHTS ACT OF 1991
173
the Supreme Court decisions in the CRA of 1991, but, more significantly, seized the opportunity to add new monetary damages and jury trial provisions that substantially increased private enforcement. Pointing to the EEOC’s and the Reagan administration’s refusal to adequately enforce Title VII, and the failure of congressional oversight to remedy the situation, civil rights advocates argued for the necessity of mobilizing private litigants and their attorneys, using economic incentives, to do what the executive branch would not. And they succeeded.
Interbranch Conflict over Title VII Enforcement in the 1980s Civil Rights Enforcement in the 1980s Reagan ran on a platform of deregulation in general, and deregulation of civil rights in particular. He took aim, especially, at category-conscious, affirmatively oriented civil rights policy, condemning “bureaucratic regulations which rely on quotas, ratios, and numerical requirements,” opposing “reverse discrimination,” and advocating a color-blind and genderblind approach to civil rights.4 The Republican party platform on which Reagan ran in 1980 declared that “equal opportunity should not be jeopardized by bureaucratic regulations and decisions which . . . exclude some individuals in favor of others.”5 Once in office, Reagan curtailed administrative enforcement of civil rights by numerous objective measures. In a large-scale study surveying federal civil rights enforcement activity during the Reagan years across the fields of employment, housing, education, and voting, the Citizens’ Commission on Civil Rights reported “a dramatic decline in civil rights enforcement by the federal government.”6 During the Reagan years, the Justice Department and federal agencies, according to the report, prosecuted fewer enforcement lawsuits, obtained lesser redress for plaintiffs in conciliation and settlement, were slower in processing complaints, and initiated fewer and less aggressive investigations into patterns of discrimination.7 As one civil rights advocate and Reagan administration critic put it, the administration “began using executive powers to dismantle the governmental machinery” for protecting civil rights.8 While there was little doubt when Reagan took office that it would be rough sailing ahead for civil rights groups’ relationship with the executive branch, the already strained relations were exacerbated when Reagan appointed William Bradford Reynolds to head the Justice Department’s Civil Rights Division (CRD). An outspoken critic of race-conscious civil rights policy, Reynolds made clear in repeated public declarations that the CRD would be unwavering in its repudiation of any policies smack-
174
CHAPTER 6
ing of preferential treatment for racial minorities or women, in the employment context and otherwise. He attacked the approach to Title VII enforcement focusing on systemic discrimination and seeking groupwide relief, without the need for individualized showing of discrimination—an approach preferred by civil rights groups—as creating “a caste system in which an individual is unfairly disadvantaged for each person who is preferred,” which he regarded as being “as offensive to standards of human decency today as it was some 84 years ago when countenanced under Plessy v. Ferguson.”9 Ira Glasser, executive director of the American Civil Liberties Union, matched Reynolds’s acerbic tone when attacking his posture toward Title VII, reflecting the severity of the rift that opened up between the Reagan Justice Department and civil rights groups: Mr. Reynolds is the moral equivalent of those Southern segregationists of a generation ago standing in the schoolhouse door to defend segregation. Today, he stands at the workplace gate, defending discrimination in employment. His efforts disgrace the American dream of equal opportunity.10 It was not only interest groups on the far left that doubted Reynolds’s commitment to vigorous civil rights enforcement. Within a year of his appointment, more than half the attorneys at the CRD signed a petition objecting to the policies of their new leader.11 Under Reynolds’s leadership, the CRD sharply reversed course and shifted to a more conservative position, consistent with the new administration’s preferences, on a variety of civil rights issues then percolating through the federal courts. The CRD participated as an intervenor or an amicus in cases seeking judicial reconsideration of several Supreme Court decisions interpreting Title VII that had allowed race-conscious affirmative action plans as an appropriate remedy in Title VII class actions, without requiring every member of the minority class to prove that he or she was subjected to individualized discrimination in order to benefit under the remedial plan. Civil rights advocates insisted that such classwide affirmative action plans were necessary to address systemic bias rooted in a history of discrimination.12 The CRD also switched sides, and moved rightward and in opposition to civil rights groups, on high-profile civil rights issues relating to busing, higher education, and redistricting.13 The EEOC’s substantive legal posture underwent a similar transformation, again reflecting the promises that the new president had campaigned on. The agency began to oppose affirmative action arrangements, particularly the use of goals and timetables in settlement agreements.14 More generally, Clarence Thomas, appointed EEOC chair in 1982, took a noninterventionist approach. He eschewed the development of policy through interpretive guidelines, promulgating many fewer than his pre-
CIVIL RIGHTS ACT OF 1991
175
decessors.15 He took the stance that the EEOC would focus its attention only on individual intentional discrimination claims, to the exclusion of systemic discrimination claims seeking classwide relief based upon statistical disparities across groups, and he eliminated the agency’s special unit for handling such systemic litigation.16 Between 1983 and the end of Reagan’s second term, the EEOC effectively ceased bringing “disparate impact” claims.17 Such claims—in which facially neutral employment policies, such as aptitude tests, are challenged as disproportionately disadvantaging minority groups—can have far greater policy impact than individual intentional discrimination claims. The EEOC also experienced large reductions in available resources. Using the Office of Management and Budget, the Reagan administration cut budget requests for the agency. The agency’s budget, adjusted for inflation and caseload, dropped by 35 percent between 1979 and 1984, after having increased steadily from the EEOC’s inception until Reagan took office.18 These declining budgets under Reagan led to significant reductions in EEOC staff. As compared with 3,752 authorized EEOC staff positions in 1979, by 1987 the number had shrunk by 22 percent, to 2,941.19 Under Thomas’s leadership, according to numerous objectively measurable criteria, the EEOC’s administrative complaint process also became much more favorable to employers and unfavorable to people complaining of discrimination. During the Reagan administration, the average length of time the EEOC took to conduct investigations doubled, increasing from roughly five months in 1980 to roughly ten months in 1989. The proportion of cases in which the EEOC found against the claimant doubled between 1980 and 1987, rising from approximately 30 to 60 percent. Concomitantly, the proportion of cases in which the EEOC achieved a settlement resulting in some benefit to the claimant sank, from about 32 percent in 1980 until it bottomed out at between 12 and 15 percent in the years 1985 to 1989. The number of complaints for whom the agency obtained monetary relief fell by 81 percent between 1980 and 1985, from 15,328 to 2,964.20 Simply put, people complaining of discrimination did worse by virtually every objective measure under the Reagan-Thomas EEOC than they had during previous administrations. Enforcement efforts at the Office of Federal Contract Compliance Programs (OFCCP) were also reduced. OFCCP, housed in the Department of Labor, is responsible for ensuring that employers who contract with the federal government comply with employment discrimination laws. Like the EEOC, under Reagan the OFCCP saw sharp reductions in its budget and staffing.21 Following the administration’s direction, it also significantly relaxed the process for determining whether a government contractor underutilized minority and women workers relative to rel-
176
CHAPTER 6
evant labor pools, and lessened its use of goals, timetables, and affirmative action plans to facilitate progress for those who did.22 The effects on bottom-line enforcement statistics were remarkable. In 1980, in carrying out its compliance review function, OFCCP obtained back pay for 4,336 individuals, while in 1986 it obtained back pay for 499 individuals.23 The aggregate sum of back pay it recovered also plummeted.24 A staff report of the House Education and Labor Committee disclosed that OFCCP cases reported for enforcement to the solicitor of the Department of Labor dropped from 269 cases in 1980 to 22 cases in 1986.25 The Department of Labor has the authority to initiate proceedings seeking termination of government contracts, and debarment from future contracts, of a contractor with deficient workforce representation of women and racial minorities. Reagan also affected civil rights policy through his marked influence on the ideology of the federal judiciary, consistent with his campaign promises to appoint conservative judges who would reverse judicial trends toward social activism. Reagan appointed Sandra Day O’Connor to the Supreme Court in 1981, Anthony Scalia in 1986, when he also elevated William Rehnquist to chief justice, and Anthony Kennedy in 1987, creating a powerful conservative majority on civil rights issues. The magnitude of Reagan’s influence, in terms of the total share of the federal bench that he had appointed by the time he left office, was matched by only two other presidents in American history. Of the 736 sitting federal judges at the end of Reagan’s second term, he had appointed 346, or nearly half. The profile of his first-term appointments was 98 percent Republican, 93 percent white, and 92 percent male. The administration was unusually rigorous, relative to past practice, in its ideological vetting of prospective judicial appointments, and by the end of the Reagan presidency the law of employment discrimination clearly reflected the preferences of the conservative jurists that Reagan had appointed. A study of federal district judge voting patterns found that between 1981 and 1985, while Carter appointees had voted in favor of race discrimination plaintiffs at a rate of 59 percent, Reagan’s appointees did so at a rate of 13 percent.26 Reagan’s influence on civil rights implementation was considerable, and this much is well recognized. It is critical to stress, however, what he was not able to control, which has been largely ignored in discussions of civil rights enforcement in the Reagan years. He was not able to control private litigation. It was not for lack of awareness or concern. From the start of the Reagan administration, high-ranking lawyers appointed to positions of leadership in the bureaucracy complained of the growth in recent years of civil rights litigation instigated by the private bar, which they regarded as frequently being of questionable merit, as producing excessively large fee awards, and as imposing undue monetary burdens
CIVIL RIGHTS ACT OF 1991
177
on both business and governmental defendants.27 The administration sought to reduce civil rights and other attorney’s fee awards, according to Michael S. Greve, a conservative activist and legal analyst, as part of a broader program to “defund the left.” When the Reagan Administration took office in 1981, one of the priorities urged upon it was to cut federal funding for liberal and leftist advocacy groups. Well known conservative activists openly advocated a strategy of “defunding the Left.” This was necessary, they argued, for the success of the conservative social and deregulatory agenda. The incoming administration shared this assessment. President Reagan himself made no bones about his hostility to public interest litigators, to whom he once referred as “a bunch of ideological ambulance chasers doing their own thing at the expense of the poor who actually need help”; he also sensed that the liberal public interest movement was a primary obstacle to his campaign promises of “regulatory relief” and of advancing a conservative social agenda.28 The administration floated a number of proposals to limit civil rights fee awards, mainly through the mechanism of capping hourly rates far below what courts had been awarding under statutory fee shifting provisions. Based upon archival research, Jefferson Decker concludes that Reagan administration leadership saw civil rights fee-shifting legislation since the CRA of 1964 as a critical part of the incentive structure generating excessive civil rights litigation, and the goal of their fee-capping proposals, at bottom, was to “drive a stake through that incentive structure.” Ultimately, however, the administration’s attempts to limit fee awards met stiff opposition from both nonprofit and for-profit civil rights lawyers, and from important members of Congress, and the attempts were defeated.29 The private enforcement status quo was sticky and Reagan could not move it. There had been 5,492 private federal job discrimination suits filed in 1980—Carter’s last year in office—a number that had been roughly stable since 1975, averaging 5,410 private civil suits per year during those six years. In Reagan’s first three years in office, as the demobilization of the civil rights administrative enforcement machinery was under way, private job discrimination litigation in federal court nearly doubled, rising to 7,046 in 1981, to 8,311 in 1982, and to 10,002 in 1983. It then declined somewhat and plateaued, averaging 8,685 private job discrimination suits per year for the balance of Regan’s presidency, a 59 percent increase over the average figure of 5,479 during the Carter years. Between 1975 and 1988, private litigation represented approximately 96 percent of total enforcement actions, with EEOC prosecutions, averaging
178
CHAPTER 6
317 per year, accounting for only 4 percent.30 Total private civil rights litigation underwent a very similar pattern of growth: roughly stable during the Carter years, rising steadily during Reagan’s first term, and plateauing in Reagan’s second term at an average level 56 percent higher than during the Carter years.31 During this period that private employment discrimination litigation was growing, the plaintiffs’ employment discrimination bar was also becoming more organized. In 1985, the National Employment Lawyers Association, an association of lawyers predominantly working in the for-profit sector, for whom a primary practice area is representing plaintiffs in job discrimination claims, was formed.32 Employers during the Reagan years were actually far more likely to be sued under federal job discriminate laws than ever before. It is plausible, from a theoretical point of view, that this sharp increase in private enforcement litigation during the Reagan years actually resulted from the executive branch’s demobilization. As detailed above, in the Reagan years, by numerous objective measures, the national government secured far less relief for employment discrimination claimants than it had under previous administrations. Given that Title VII allows claimants dissatisfied with the results of the administrative process to proceed with private litigation, an administrative process that delivers fewer benefits to claimants is likely to produce more civil rights lawsuits. The decline, for example, in the number of claimants for whom the EEOC obtained monetary relief, from 15,328 in 1980 to 2,964 in 1985, would almost certainly be associated with an increase in the number of claimants leaving the agency dissatisfied, and thus more likely to seek redress through private litigation. Whatever the cause of the surge of private job discrimination litigation during the Reagan years, it powerfully demonstrates the capacity of private enforcement regimes to insulate at least a portion of the enforcement function from executive control. The Congressional Response During the 1980s, a predominantly Democratic Congress was acutely aware of the Reagan administration’s demobilization of the civil rights enforcement machinery, and the decade was marked by implacable conflict over civil rights policy between Congress and the executive branch. The conflicts ranged from confirmation battles over presidential appointments in which civil rights issues were central (the most notable of which was the failed nomination of Robert Bork to the Supreme Court), to the first vetoes of civil rights legislation since Reconstruction, to legislativeexecutive struggles over control of agencies with responsibility for implementing civil rights laws.33 The EEOC was a key focal point of this conflict. In the nine years between 1983 and enactment of the CRA of 1991,
CIVIL RIGHTS ACT OF 1991
179
Democratic-chaired congressional committees conducted no less than fifteen oversight hearings examining aspects of EEOC enforcement efforts, including (1) the substantive legal positions the agency advocated, (2) the limited nature of its enforcement litigation, and (3) its complaintprocessing practices.34 The tenor of the hearings was sharply critical and often combative. Committee Democrats excoriated the agency for what they regarded as a pro-employer bias in the agency’s position on legal issues of broad significance, including repudiation of consent decrees conferring classwide relief, affirmative action plans utilizing goals and timetables, and “disparate impact” theory, in which facially neutral employment policies, such as promotion tests, are challenged on the basis of statistical evidence establishing that they disproportionately exclude protected groups.35 The Congressional Black Caucus attacked chairman Thomas for grudgingly narrow interpretations of civil rights precedents, refusing to seek the full range of remedies provided by statute, and eschewing class actions and systemic pattern-or-practice cases.36 Congressional Democrats also criticized the agency for delay in its complaint processing, most dramatically illustrated by revelations in 1988 that the statute of limitations had expired on thousands of age discrimination claims while they languished before the agency, claims that Congress revived by special statute.37 Following an investigation of selected EEOC district offices, the House Committee on Education and Labor reported that, as compared to 1980 (Carter’s last year in office), by 1985 there had been a precipitous decline in the number of complainants for whom the agency obtained monetary benefits in the conciliation process, and a doubling in the percentage of charges in which the agency found “no cause” to believe that discrimination had occurred, while in some instances “no cause” findings were rendered without meaningful investigation.38 Finally, Democrats made abundantly clear that they regarded these deficiencies in EEOC performance as the result of the ideological preferences of a presidential administration that opposed robust civil rights enforcement. In 1983 oversight hearings, Democrat Augustus Hawkins, chair of the House Subcommittee on Employment Opportunities, decried “the extent to which the administration has gone to undermine the Nation’s civil rights enforcement efforts.”39 In 1986 hearings, Democrat Matthew Martinez, who took over as subcommittee chair in 1985, characterized the EEOC as “an agency which obstructs enforcement of the law [and] . . . has abdicated responsibilities which it has been charged with,” and stated that “the EEOC Commissioners seek to undermine the mission of the agency.”40 In 1987, Democrat Tom Lantos, chair of the House Subcommittee on Employment and Housing, complained during oversight hearings in reference to the EEOC: “[T]here is not much point in passing
180
CHAPTER 6
good laws if, in the implementation of the legislation, the purpose and the goal and the objective of sound legislation is undermined, eroded, or sabotaged . . . making a mockery of the solemn promise of the Civil Rights Act.”41 A report of the Congressional Black Caucus disclosed that “Thomas was so reluctant to bring class and systemic cases that Congress had to earmark EEOC funds specifically for that type of enforcement and threaten to cut the budget of the chair and members of the EEOC” in an effort to compel him to undertake such litigation.42 During the 1980s, Democrats in control of oversight committees were locked in an acrimonious struggle with the Reagan administration over control of the EEOC.
The Supreme Court Curtails Private Title VII Enforcement While Congress clashed throughout the 1980s with the president over civil rights policy, it was also sparring on the same subject with the Supreme Court, which increasingly reflected presidential preferences with the appointments of Justices O’Connor, Scalia, and Kennedy. This conflict heated up particularly in the latter half of the decade and into the early 1990s. In the five years from 1986 to 1990, Congress passed five pieces of civil rights legislation that were explicitly directed at overriding Supreme Court decisions interpreting civil rights statutes, and the tone of the legislative proceedings leading to that legislation was highly antagonistic toward the Court.43 It was this legislative-judicial discord that ultimately served as the catalyst for the CRA of 1991. The legislative momentum that culminated in passage of the CRA of 1991 was triggered more than two years earlier, in May and June 1989, by a series of five Supreme Court decisions interpreting Title VII handed down by a sharply divided Supreme Court. This is a remarkably large number of Supreme Court decisions to be issued interpreting a single title of a single statute in less than two months, and it signaled a clear move by the Court to turn back the rising tide of private Title VII lawsuits. All but one of the decisions were stridently divided, with an ascendant conservative wing carrying the day. Most of the cases were decided by the new five-justice majority, established the previous year with Reagan’s appointment of Anthony Kennedy, who joined Justices Rehnquist, Scalia, White, and O’Connor to form a new majority on many civil rights issues. The decisions did not overtly strike at the substantive right to a workplace free of discrimination for protected groups, but rather they constricted Title VII’s private enforcement regime. The most controversial of the cases was Wards Cove Packing Co. v. Atonio,44 which substantially revised the framework for adjudicating “disparate impact” cases established in the 1971 landmark Griggs v. Duke Power Co. (discussed in
CIVIL RIGHTS ACT OF 1991
181
the last chapter), an infrequent but important type of claim in which a plaintiff challenges a facially nondiscriminatory policy, such as aptitude tests, as having discriminatory effects on a protected group. Wards Cove tightened a plaintiff’s burden of proof, limited the nature of evidence she could rely upon, and expanded the range of defenses available to employers. In Price Waterhouse v. Hopkins, the Court held that even when discrimination was a “substantial factor” in an adverse employment action, the employer would be absolved of liability if it established that it would have made the same decision in the absence of the discriminatory motive.45 In Martin v. Wilks, it expanded standing for whites and males to challenge the legality of affirmative action consent decrees entered into by municipalities that benefited racial minorities and women.46 In Lorance v. AT&T Technologies, it restricted application of Title VII’s statute of limitations in cases challenging seniority systems so as to cause such claims to expire faster.47 In Independent Fed’n of Flight Attendants v. Zipes, it limited the potential entities from which successful Title VII plaintiffs could recover attorney’s fees under the statute’s fee-shifting provision.48 In addition to these five Title VII cases, also in the summer of 1989 a five-justice majority handed down Patterson v. McLean Credit Union, holding that the Civil Rights Act of 1866’s prohibition of race discrimination in private employment was limited to the formation and enforcement of contracts, and did not extend to postformation conduct, such as discriminatory harassment.49 While the legislative process aimed at overriding these decisions was under way, in March 1991 the Supreme Court handed down two additional decisions that influenced the content of the CRA of 1991’s amendments to Title VII. The decisions were again divided, with president Bush’s appointment of Justice Souter in 1990 adding an additional vote (if only initially) to the new conservative majority. In Boureslan v. Arabian American Oil Company, the Court held that American citizens employed by an American company in a foreign country were not covered by Title VII,50 and in West Virginia University Hospitals, Inc. v. Casey, it held that expert witnesses’ fees could not be recovered by prevailing plaintiffs as part of attorney’s fees under the Civil Rights Attorney’s Fees Awards Act of 1976, a decision that did not address Title VII but clearly indicated how the Court would resolve the same issue under Title VII, which has identical fee-shifting language.51 There could be no mistaking the fact that the new conservative Supreme Court majority sought to cut back Title VII’s private enforcement regime, and with it Title VII litigation. Addressing rules governing burdens of proof, standards of evidence, standing, statutes of limitations, attorneys’ fees, and expert witness costs, in every single one of the cases the Court adjusted these elements of Title VII’s private enforcement regime
182
CHAPTER 6
in the same direction—to the detriment of women and minority plaintiffs. National civil rights organizations, including the Leadership Conference on Civil Rights, the NAACP, and the ACLU, reacted with swift and decisive condemnation.52 In commenting on the ostensibly technical and procedural decisions, a member of the NAACP’s executive board lamented that President Reagan had reshaped the Supreme Court’s civil rights philosophy, while the organization’s executive director, Benjamin Hooks, characterized the Court as “more dangerous to this nation than any Bull Connor with a fire hose.”53 Conservatives no less than liberals recognized the deeply ideological significance of the decisions. Bruce Fein, a conservative constitutional lawyer who had worked in the Reagan Justice Department, characterized the decisions as reflecting “that we have a Reagan court on civil rights and social welfare. It means we won’t have the rampant social engineering emanating from federal judges.”54 Civil rights leaders were clear in their assessment of the decisions’ effects, and their analysis closely followed the law and economics model of the choice to litigate. Because the decisions decreased the probability of plaintiffs’ success and reduced potential economic recoveries, civil rights lawyers would litigate fewer cases, curtailing private enforcement of Title VII.55 Reporting on a core theme that emerged in a two-day strategy session following the decisions held by the LDF, attended by about eighty of the nation’s leading civil rights advocates, a staff reporter of the Wall Street Journal wrote: As a result of the decisions, civil-rights lawyers are turning away clients because many job-discrimination claims are viewed as too costly to litigate and almost impossible to win, the lawyers said. Some of the lawyers predicted that these decisions—combined with the increased difficulty of getting courts to require defendants to pay attorney’s fees and the tendency of some judges to dismiss novel claims as frivolous—will lead to a thinning of the ranks of civilrights lawyers.56 Civil rights activists sought a legislative solution.
Congress Overrides, and More Civil rights organizations and activists were the catalysts for override legislation.57 Longtime civil rights liberals Augustus Hawkins, chair of the House Education and Labor Committee, and Edward Kennedy, chair of the Senate Labor and Human Resources Committee, led the way among lawmakers. Reginald Govan, staff counsel on the House Education and Labor Committee, who was tasked with developing the override legisla-
CIVIL RIGHTS ACT OF 1991
183
tion on the House side, wrote a firsthand contemporaneous account of events between the initial mobilization to override the 1989 decisions and passage of the CRA of 1991.58 He reports that the Leadership Conference on Civil Rights constituted a drafting committee, which worked in close collaboration with congressional committee staff to draft the override bill that would be introduced in February 1990.59 The drafting committee included representatives of the NAACP, the LDF, the Lawyers’ Committee for Civil Rights Under Law, the Mexican-American Legal Defense Fund, the National Urban League, the American Civil Liberties Union, the Women’s Legal Defense Fund, the National Women’s Law Center, and People for the American Way.60 While representatives of lawyers’ associations representing for-profit lawyers, such as the American Bar Association and the National Employment Lawyers Association, would later testify in support of the bill, none were invited to participate in the process of crafting the law.61 In addition to formulating language to overturn the Supreme Court decisions issued in the summer of 1989, the drafting committee persuaded Hawkins and Kennedy to press beyond a narrow override and incorporate into the bill additional amendments adding the availability of uncapped compensatory and punitive damages, providing for trial by jury, overturning several decisions handed down in the mid-1980s restricting attorney’s fee awards, and extending Title VII’s statute of limitations from 180 days to two years.62 Using punitive and other monetary damages, more liberal fee award rules, and a larger limitations period, to increase private lawsuits in the face of the Reagan/Bush retrenchment on civil rights was not a new idea. Civil rights advocates had recently secured passage of the Fair Housing Act Amendments of 1988, which removed the $1000 cap on punitive damages, increased the limitations period, and authorized attorney’s fee awards regardless of the plaintiff’s ability to pay (overturning the limitation on FHA fee awards introduced by Senator Byrd in 1968, discussed in chapter 4). These FHA amendments had the purpose and effect of increasing private enforcement.63 The proposed new damages provisions would radically alter potential economic recoveries for Title VII plaintiffs, for whom back pay was the only form of monetary relief available. Under the proposed bill, compensatory damages would be available for all pecuniary losses resulting from discrimination (as opposed to back pay only), as well as for pain and suffering. Punitive damages could also be awarded, without limit, to punish the employer and deter future violations by it and other would-be violators. This change in available damages would transform claims that had previously lacked any monetary value under Title VII because they did not involve back pay, such as claims based upon religious or sexual harassment, into claims with potentially massive monetary value. Fur-
184
CHAPTER 6
ther, claims that had previously involved only modest monetary stakes for the plaintiff, so typical of Title VII’s back-pay limitation, would be worth much more. Also crucial was the drafting committee’s proposed right to trial by jury. The liability determination of whether a plaintiff had been discriminated against, and if so, the damages determination of how much money to award her, would be in the hands of a cross section of the population rather than a federal judge (at the time most likely to be a white male Republican). Whereas in 1963–64 civil rights advocates regarded bench trials as imperative because they feared bias at the hands of southern juries,64 in 1989 they were more fearful of bias at the hands of federal judges and strongly preferred jury trials.65 Reginald Govan, the committee counsel who worked directly with the drafting committee, explained that civil rights groups pressed for the jury trial amendment “because it stripped a federal judiciary increasingly comprised of Reagan/Bush appointees of its exclusive role as fact-finder in employment discrimination cases under Title VII.”66 These proposed damages and jury trial provisions would have the effect of giving to gender, religion, and national origin plaintiffs under Title VII benefits already enjoyed by race discrimination plaintiffs under the CRA of 1866. The latter law’s prohibition of interference with a person’s right to “make and enforce contracts” that “is enjoyed by white citizens” was interpreted in the early 1970s to apply to private employment, and it includes the right to compensatory and punitive damages as well as trial by jury. Federal courts have read the language “white citizens” as indicating that the law’s “clear purpose . . . [is] to provide for equality of persons of different races,” such that it does “not cover discrimination based on religion, sex, or national origin.”67 In pressing for adoption of the damages and jury trial provisions under Title VII, advocates stressed the equitable argument that it would bring other protected groups to parity with rights already enjoyed by racial minorities.68 The Leadership Conference’s position on cease-and-desist had traveled light years since H.R. 405 was proposed a quarter century earlier. Now that substantial Title VII amendments were under way, the possibility of giving the EEOC cease-and-desist powers came back onto the agenda, with proposals ranging from a pure administrative cease-anddesist model such as that proposed in H.R. 405, to an election-of-remedies model in which claimants could choose between an administrative and a judicial venue.69 Among the cease-and-desist advocates was Clifford Alexander, EEOC chair from 1967 to 1969 and unsuccessful proponent of cease-and-desist in the 1969 hearings leading to passage of the 1972 amendments to Title VII.70 In a Washington Post editorial titled “The EEOC Solution,” Alexander called for enactment of the cease-and-desist
CIVIL RIGHTS ACT OF 1991
185
powers that “civil rights leaders” had wanted in 1964, but had sacrificed “[t]o deter minority leader Sen. Everett Dirksen’s filibuster, which would have killed the landmark Civil Rights Act of 1964.” Alexander’s proposal of moving to the enforcement model sought by civil rights advocates in 1964, of course, would entail eliminating Title VII’s private right of action. “Only the lawyers, who stand to lose millions in fees,” remarked Alexander, “will complain.”71 Some on the House Education and Labor Committee thought the idea might have merit and asked the Leadership Conference drafting committee to consider it. The drafting committee, however, “categorically rejected this proposal.”72 In explaining lack of support for cease-and-desist, Richard Seymour, director of the Employment Discrimination Project of the Lawyers’ Committee for Civil Rights Under Law for the quarter century spanning 1977 to 2001, and a member of the drafting committee, rejected the possibility of the agency as an adequate vehicle for processing claims, partly due to underfunding, but mostly due to “lack of diligence and ineptitude.”73 Civil rights advocates regarded the agency with a particularly skeptical eye in 1989, after seven years under the chairmanship of Clarence Thomas, whom they regarded as ideologically hostile to their agenda.74 During his tenure at the Lawyers’ Committee, Seymour stated, “No one involved in private enforcement ever has been willing to give up the private right of action.” He continued: “Would anyone relinquish anything to give the EEOC cease-and-desist power? Nope. Is it ever foreseeable that they would? Nope. . . . At meetings of the Leadership Conference on Civil Rights, there has been no question about this for as long as I can remember.”75 Alexander, who had left government for a career in the corporate sector in 1969, was off the mark in sizing up potential opposition to ceaseand-desist. While he was right that civil rights leaders wanted a pure cease-and-desist framework in 1964, the context had changed a great deal. By the end of the 1960s, while still advocating cease-and-desist, they insisted on retaining a partial private right of action as an alternative, and to function as a safety valve, in a hybrid cease-and-desist / private litigation framework. In 1989, the Leadership Conference flatly repudiated cease-and-desist under any circumstances, hybrid or otherwise. While Alexander’s views had not changed, civil rights leaders’ had. More disappointed than ever in an agency headed by Clearance Thomas since 1982, they now looked back on the Dirksen compromise as a great gift. Indeed, just as they had built upon the Dirksen compromise by extending its feeshifting approach to other civil rights laws in the 1970s, in 1989 they sought to build upon it once more. By pushing beyond a simple override and adding the new monetary damages and jury trial provisions to Title VII, thereby incentivizing greater private enforcement, they sought to tilt
186
CHAPTER 6
the public/private balance in Title VII enforcement, already dominated by private litigation, toward further privatization. When introduced in February 1990, the override bill composed by the Leadership Conference drafting committee, in close coordination with Hawkins’s and Kennedy’s committee staff, encountered stiff opposition from the Bush administration, conservative Republicans in Congress, and the business community. Several entwined features of debate over the bill distinguished it from debates over private enforcement provisions between 1964 and 1976. In marked contrast with their relative passivity with respect to those earlier debates, major national business interests, such as those represented by the National Association of Manufacturers and the United States Chamber of Commerce, moved strongly to block the 1990 bill, particularly its provisions concerning disparate impact claims and new monetary damages.76 The steep growth of private job discrimination suits, and the attendant costs imposed on business, had by 1990 made job discrimination suits a serious business concern. The notable contentiousness of debates over the bill also stood in sharp contrast to the relative consensus achieved among nonsouthern legislators in the earlier legislative debates over private enforcement of civil rights laws.77 Party polarization was clearly in evidence. Debate was particularly belligerent in connection with two substantive grounds offered for opposition to the bill. First, a now familiar antilitigation and antilawyer theme was broadly advanced against the bill by Republicans and their business constituents. The proposed bolstering of Title VII’s private enforcement regime, they argued, would be an “engine of litigation,” would “provoke a torrent of costly litigation,” and would “prompt frivolous lawsuits and bankrupting damage awards.”78 It would yield a “bonanza . . . for plaintiffs and their lawyers,” “would do more to promote legal fees than civil rights,” and could best be characterized as “the plaintiffs’ lawyers’ full employment law of 1990.”79 This “litigation explosion” narrative—featuring Democrats providing benefits to plaintiffs with specious claims and their greedy lawyers, and imposing massive costs on business—was confined to only a few Southerners in the debates over the 1964 and 1976 acts.80 In debates leading to the CRA of 1991, it was the master narrative of the bill’s opponents. A second point of heated contention, reflecting the partisan antagonisms over civil rights that had grown in the 1980s, concerned assertions by many Republicans, echoing their galvanized business constituents, that the projected explosive growth in lawsuits if the bill passed, particularly “disparate impact” suits, would have the practical effect of causing hiring “quotas.”81 Employers seeking to avoid liability would give preferences to minorities and women to the detriment of more qualified white
CIVIL RIGHTS ACT OF 1991
187
males. This argument—that the bill would effectively induce affirmative action—proved a politically potent weapon against it, at least in the short run.82 These attacks, coupled with a veto threat by the president, succeed in causing liberal Democrats and the drafting committee to make some concessions sought by southern Democrats and moderate Republicans, such as capping punitive damages, in their quest for a veto-proof majority.83 In October 1990, both the House and Senate voted to pass a bill—in both chambers just shy of the two-thirds necessary to override a veto—that overturned nearly all of the offending decisions and provided for uncapped compensatory damages, punitive damages capped at $150,000, trial by jury, and a statute of limitations enlarged to two years.84 President Bush’s opposition remained firm, but he was also sufficiently concerned about the electoral repercussions of a veto that he wished to avoid appearing adamantly opposed to any civil rights legislation.85 He proposed a vastly weaker bill that no one thought Democrats would accept, and when they summarily rejected it, he vetoed what would have been the Civil Rights Act of 1990.86 It appeared that an override vote in the House would not succeed, so none was taken, and the Senate failed to override by a single vote.87 The 101st Congress ended without civil rights legislation. In the first session of the next Congress the issue was back on the agenda, and it was evident that civil rights liberals would have to make concessions relative to the bill that was passed in 1990 to attract sufficient support to override an anticipated veto. And so they did. Working with moderate Republicans led by Senator John Danforth, they conceded some ground to opponents of the 1990 bill on what had been hard-fought battles over the formulation of the burden of proof, available defenses, and allowable inferences from statistical evidence in disparate impact cases; they agreed to cap not just punitive damages, but also most forms of compensatory damages as well; and they ceased to insist upon increasing Title VII’s statute of limitations.88 While these moderating concessions were enlarging support for the bill, the political calculus shifted in favor of voting for it. In early October 1991, shortly after the first round of Senate hearings on the nomination of Clarence Thomas to the Supreme Court had concluded, the story on Anita Hill’s allegations of sexual harassment broke. The nation was then riveted by televised hearings on the subject from October 11 to October 13, and the seriousness and pervasiveness of sexual harassment on the job, for which Title VII provided no compensatory remedy but the proposed amendments would, became the focus of a national dialogue. This development dovetailed perfectly with the recent strategy of some of the bill’s proponents, who sought to neutralize the race politics of the quota issue
188
CHAPTER 6
by casting the bill primarily as a women’s rights bill, emphasizing that it provided monetary damages to victims of gender discrimination already enjoyed by race discrimination plaintiffs under the CRA of 1866.89 The White House and Republican senators were beset by charges that their reflexive disbelief of Hill’s allegations revealed insensitivity to women’s rights in general, and sexual harassment in particular, arousing anxiety among conservative Republicans about voting against another civil rights bill that increased remedies for gender discrimination at work.90 Public opinion polls conducted in January 1991 on the failed 1990 bill had shown remarkable public support for compensatory and punitive damages in gender discrimination claims, with 68 percent in favor and 25 percent against, and this was before the Hill-Thomas hearings.91 Republicans were also concerned about recent national media attention drawn by former Klansman turned self-styled “white nationalist” David Duke, who had made a strong second-place showing in the November Republican primaries for the Louisiana governorship. While the national Republican party and the president himself unambiguously repudiated Duke,92 they were clearly worried that the Duke spectacle would feed what moderate Republican James Jeffords of Vermont characterized as the “perception problem”: that the party was insensitive on race issues and at times exploited them.93 Democrats invoked Duke’s name repeatedly in the debates over the bill, charging that Republicans’ frequent repetition of the word “quota” represented a Duke-like strategy of using race-baiting code words.94 Polling in early 1991 had disclosed that a majority of respondents thought that President Bush was “just playing politics” by inaccurately denominating the proposed legislation a “quota bill.”95 With the 1992 elections looming, the Duke controversy fostered apprehension among many Republicans about opposing even compromise civil rights legislation, and, together with the charges of sexism against Republicans following the Hill-Thomas hearings, it eroded the number of Republicans prepared to stand with president against the compromise bill.96 Senators John Warner of Virginia and Ted Stevens of Alaska, who had voted to sustain the president’s veto in 1990,97 publicly announced that they had met with the president and notified him that they could not be counted upon to cast that vote again. “I felt that it was imperative that we did not have the divisiveness of not having a civil rights bill in the 1992 presidential and congressional campaigns,” Warner said. “We made it clear we felt it was necessary to get a bill.”98 The compromises reached on some of the more contentious issues, particularly those regarding disparate impact claims and damages caps, coupled with a political context more conducive to passage, produced veto-proof majorities in both chambers for a compromise bill. Missouri
CIVIL RIGHTS ACT OF 1991
189
Republican John Danforth introduced the compromise bill that would become the CRA of 1991 and played a leading role in brokering the deal.99 As with the Title VII’s private enforcement regime in 1964, the Fair Housing Act’s private enforcement regime in 1968, and the School Aid Act of 1972’s provision for attorney’s fee awards for winning plaintiffs in school desegregation litigation, it was once again a Republican who introduced the litigant-mobilizing compromise civil rights bill that would become law. The CRA of 1991, as finally passed, overrode a large majority, though not all, of the offending Supreme Court decisions of 1989 and 1991.100 The amendments to Title VII statutorily specified new evidentiary and proof standards that eased a plaintiff’s burden in disparate impact cases, enacted rules shifting certain costs of prosecution to defendants in cases won by plaintiffs, clarified statutes of limitation so as to make them more permissive in seniority cases, and restricted the scope of standing for individuals to challenge consent decrees creating affirmative action programs. These provisions of the CRA of 1991 sought, manifestly, to restore nearly all aspects of Title VII’s private enforcement regime to their condition prior to the summer of 1989. The CRA of 1991 also created new punitive and compensatory damages provisions, where the sum of (1) punitive damages, (2) compensatory damages for pain, suffering, and other nonpecuniary losses, and (3) future pecuniary losses, would be capped at between $50,000 and $300,000 in a gradated scheme increasing with the size of the employer.101 Compensatory damages for pecuniary losses— excluding future losses—would not be capped. Plaintiffs seeking compensatory or punitive damages would be entitled to trial by jury.102 When President Bush’s final efforts to rally conservative Republicans to sustain a veto failed, the president sought to negotiate some eleventhhour concessions from the bill’s proponents, but with his veto threat undermined he had lost any leverage to effect material changes.103 The Senate voted for the Civil Rights Act of 1991 93 to 5,104 and the House voted for it 381 to 38,105 a stunning bipartisan consensus for the act’s unambiguous increase in incentives for private litigation, enacted for the express purpose, as shown below, of increasing the number of private Title VII suits. Republicans had fought hard to minimize the extent of the law’s litigant-mobilizing effects. However, in the face of mounting public pressure and an impending election, and after securing significant concessions from liberal Democrats following the 1990 veto, they overwhelmingly voted for it. Addressing civil rights enforcement through incentivizing private litigation continued to command broad consensus even in an era in which the parties were deeply polarized over civil rights and there was much antilitigation rhetoric in the air. Recognizing clear defeat and not wanting to further alienate minority and women voters without any po-
190
CHAPTER 6
tential benefits for business, President Bush switched to a credit-claiming posture, asserted that the bill before him was not a “quota bill” like the one passed in 1990, and signed it into law on November 21, 1991.106
Legislative Historical Evidence of Lawmakers’ Motives Mobilizing Private Enforcers Why did Congress, following an agenda set by liberal members with strong ties to civil rights groups, pass a law straightforwardly intended to increase Title VII litigation? Both the House and Senate reports on the bills that led to the CRA of 1991 emphasized that private enforcement litigation was intended, in the compromise of 1964, to be a central part of the enforcement scheme for federal employment discrimination laws. “In enacting Title VII,” the House report stated, “Congress intended to vindicate the substantial public interest in a discrimination-free workplace by encouraging private citizens to enforce the statute’s guarantees,” and it identified as a key goal of the statute to “encourage . . . victims [of discrimination] to act as ‘private attorneys general’ by enforcing the statute for the benefit of all Americans.”107 The House and Senate reports further concurred that Title VII’s remedial scheme, both in its express content and as recently interpreted by the new conservative majority on the Supreme Court, was deficient in meeting this goal. There existed instead a condition of underenforcement of meritorious claims due to inadequate monetary incentives for plaintiffs to assert claims, and for attorneys to prosecute them. With respect to the modest monetary damages available under Title VII, the Senate report stated that limiting economic damages to back pay “often means that victims of intentional discrimination may not recover for the very real effects of the discrimination; and thus victims of intentional discrimination are discouraged from seeking to vindicate their civil rights.”108 The House report characterized the strict limitations on monetary recovery under Title VII as posing “significant disincentives to would-be enforcers,” and as likely having “a depressant effect on discrimination suits.”109 Allowance for compensatory and punitive damages would mobilize more private plaintiffs with meritorious claims to enforce their rights. Further, and critically, the bill’s advocates emphasized that even where monetary damages provide sufficient incentives for plaintiffs to initiate enforcement, retaining counsel to prosecute claims is another matter entirely, even in clearly meritorious cases. In December 1989, the Federal Courts Study Committee (FCSC), empanelled by Congress and charged with examining problems facing the federal courts, released a report stating that the monetary stakes in many Title VII cases were so small that,
CIVIL RIGHTS ACT OF 1991
191
even with the potential to recover attorney’s fees, claimants sometimes found it difficult to litigate because they were unable to retain counsel. The FCSC identified three reasons for plaintiffs’ difficulty in retaining counsel: delay, uncertainty of outcome, and difficulty in recovering fees by winning plaintiffs.110 While attorneys would be awarded fees if they prevailed, litigation nearly always involves long delay and uncertainty of outcome, and thus the prospect of fees is discounted accordingly. An attorney with purely economic motives who has the opportunity to be paid at market rate, in a timely fashion, and not contingent upon winning, would not accept Title VII cases.111 Further, in developing a framework for assessing whether a prevailing plaintiff’s fee petition was “reasonable,” many courts had adopted a practice of weighing whether the fee requested (based upon hours worked on the case) was “proportionate” to the economic recovery achieved for plaintiffs.112 If not, courts utilizing this approach would allow only a proportionate award even if it meant compensating attorneys at well below market rates, sometimes drastically so.113 In a 1989 article titled “Eliminating the Plaintiff’s Attorney in Equal Employment Litigation,” an EEOC regional attorney, with no financial stake in fee awards, characterized a series of federal court decisions in the middle to late 1980s curtailing fee awards in Title VII cases as “a disaster for plaintiffs . . . [and] a judicially constructed mine field through which many attorneys have decided not to travel.”114 These conditions, according to the FCSC, made it difficult for plaintiffs with meritorious but low-value cases to retain counsel. The Senate report observed that there was a “dearth of competent counsel willing to represent victims of discrimination despite many meritorious suits,”115 and that “private counsel representing plaintiffs in equal employment cases have become an endangered species, in many places already extinct.”116 The House report similarly found that, as a result of inadequate monetary incentives to attract counsel to prosecute claims, “The evidence suggests that plaintiffs are already encountering substantial difficulty in obtaining legal representation.”117 Lack of counsel for potential plaintiffs with meritorious but low-value claims, these legislators concluded, was a market problem in need of a market solution. Attorneys will only represent plaintiffs, the Senate report observed, citing the comments of a distinguished federal judge, if “the civil rights market . . . adequately compensate[s] them.”118 “[A]ttorneys, like other professionals and workers generally,” the House report echoed, “cannot afford to work for free. They have got to be paid.”119 The extent of monetary incentives for attorneys to enforce statutes can be regarded, it continued, “as a fuel that makes the machinery of adjudication work. If the fuel runs out, the machinery does not function and civil
192
CHAPTER 6
rights do not have the effect of protecting people whose interests are at stake.”120 Adding compensatory and punitive damages would allow contingency arrangements to counteract the discounting of fees that results from delay and uncertainty of outcome, and would reduce the problem of small awards (back pay only) serving as a justification for judges, applying the proportionality principle, to award fees far below market rates. The proposed new damages, then, were intended by civil rights advocates both to mobilize plaintiffs and to increase their chances of retaining counsel. With respect to the latter issue, the new damages were manifestly intended to transform the “civil rights market” so as to shore up the civil right bar. Separation of Powers Conflicts But what motivated Congress in 1989–91, after living with Title VII’s original private enforcement provisions for a quarter century, to push beyond a simple override of the offending decisions and tilt the enforcement framework toward greater privatization with the new damages and jury trial provisions? The legislative record is replete with evidence that separation of powers conflicts between Congress and the executive over implementation of Title VII drove Congress, in the CRA of 1991, to incentivize more private enforcement litigation. The CRA of 1991 was the apex of a decade-long struggle between Congress and the president over civil rights policy in general, and the EEOC in particular. Liberal Democratic legislators and civil rights groups argued, unambiguously, that increased private enforcement of Title VII was necessary because (1) they regarded the Reagan and Bush administrations as hostile to civil rights in general; (2) they were sorely dissatisfied with EEOC performance in particular; (3) they regarded deficiencies in EEOC performance as rooted in executive ideology; and (4) they believed that congressional oversight of the agency had failed. They sought, accordingly, to mobilize private litigants to do what the EEOC would not. Republican legislators’ and business groups’ opposition to the liberals’ proposals were also entwined with separation of powers struggles. They complained that the practical effect of the law would be to shift the regulatory implementation scheme from one centered upon bureaucracy to one dominated by private litigation, thereby diminishing executive power. Some actually urged that enforcement should be improved by strengthening the administrative power of the EEOC, a suggestion roundly repudiated by liberals. In the course of their arguments in favor of an enhanced private enforcement regime for Title VII, Democratic legislators frequently shifted into broad-gauged attacks on the Reagan-Bush record on civil rights.
CIVIL RIGHTS ACT OF 1991
193
They made clear that the proposed legislation was intended as a response to executive ideology, as well as, manifestly, the decisions of recent Republican judicial appointments who were, many Democrats believed, carrying out the Reagan agenda on civil rights. In the House, typical were statements by Major Owens of New York that the bill introduced by Hawkins could “prevent the poisonous civil rights polices of the Reagan administration from being permanently embedded in the laws”;121 by John Conyers of Michigan that “the President is leading the charge to reverse civil rights progress”;122 and by Donald Edwards of California that “we have generally a very hostile administration to civil rights legislation.”123 The same theme reverberated in the Senate, where Howard Metzenbaum of Ohio thundered: Over the last 9 years we have seen a marked increase in the tolerance for racism and sexism. The Reagan administration launched a campaign against civil rights. Ronald Reagan did more to set back the clock on civil rights than any other president in this century. . . . It is no wonder, then, that the US Supreme Court, with the three Reagan appointees tipping the balance, slammed the door in the face of the victims of discrimination.124 Such broadside attacks on the Reagan civil rights record—focused on both bureaucracy and courts—were also recurrent among witnesses during committee hearings, generally offered as prologue to pleas that Congress act swiftly against executive and judicial retrenchment on civil rights by bolstering Title VII’s private enforcement regime.125 Civil rights advocates argued that private litigants had to be mobilized to perform the enforcement task that the EEOC was failing to accomplish. Some characterized the agency as simply ineffectual. Regarding “enforcement by the EEOC and other agencies,” Chairman Hawkins pronounced that “they offer no help whatsoever,”126 and he characterized the EEOC’s administrative process as “useless.”127 House Democrat Jose Serrano of New York stated that the agency had failed even to investigate claims filed with it,128 alluding to a General Accounting Office report requested by Hawkins and released at the end of 1988 that found, across six EEOC district offices studied, that “41 to 82 percent of charges closed by the district offices were not fully investigated.”129 Ellen Vargyas of the National Women’s Law Center said of the EEOC’s administrative process: “It simply hasn’t worked.”130 Claimants and attorneys who had been through the agency process similarly testified that it did not meaningfully investigate their claims or otherwise provide help to them.131 Others regarded the agency as not just ineffectual and unresponsive, but also affirmatively harmful, both symbolically and practically. Benjamin Hooks, executive director of the NAACP, complained that the agency
194
CHAPTER 6
process “make[s] a mockery of one’s ability to bring claims of discrimination before administrative agencies.”132 A private civil rights practitioner characterized it as “a hurdle that one has to overcome,” and as “a thorn in the process,” producing delay that simply had to be endured before proceeding to court, where it would be possible to address discriminatory employment practices in a serious fashion.133 Representatives of the Hispanic community accused the agency of bias against them.134 Indeed, this view that the agency’s administrative procedures provided no benefits to claimants while imposing costs upon them led Vargyas to advocate eliminating the requirement that claimants exhaust the administrative process before filing an action in court.135 The EEOC, the bill’s advocates charged, was not only failing to deliver remedies to victims of discrimination, but concomitantly was failing to create meaningful compliance incentives for employers. “I don’t think being notified of a Title VII charge pending,” said Nancy Kreiter of the Women Employed Institute in the House hearings, “makes many employers shake in their boots these days. Certainly going through the EEOC process doesn’t.”136 In urging her colleagues in the House to increase available monetary damages for Title VII plaintiffs, Washington Democrat Jolene Unsoeld insisted that “[a] slap on the hand from the EEOC just isn’t good enough,” and that increased money damages would provide the kind of “meaningful deterrent” the EEOC was not delivering.137 This justification for the bill was repeated time and again in floor statements.138 Moreover, and this point is crucial to the present analysis, Democratic legislators made clear that they did not regard the EEOC’s deficiencies as an enforcer to be mere bureaucratic ineptitude, nor the result of simple resource constraints. Rather, they understood them to result from their ideological opponents’ control of the executive branch. Advocates of the bill identified as a key source of the EEOC’s weakness the fact that during the Reagan-Bush years the executive branch had refused to use the agency to implement civil rights laws vigorously. Such denunciations, of course, were not new. They closely paralleled charges by Democrats leveled regularly during the near-continuous EEOC oversight hearings throughout the 1980s—charges that EEOC leadership was obstructing and sabotaging enforcement. During committee hearings, in explaining the need for the bill’s punitive damages provision to mobilize private enforcers, Representative Donald Payne of New Jersey remarked: [T]his agency which is the appointed body of administration—have you followed the last eight, nine or ten years and observed the people who have been nominated for civil rights enforcement? I think that
CIVIL RIGHTS ACT OF 1991
195
the less you want to do to enforce civil rights the better your opportunities are to get nominated.139 “We had an EEOC during the Reagan years that did not function properly,” complained Ohio representative Mary Rose Oakar in her floor statement in support of the proposed enhanced private enforcement regime. “Ironically,” she continued, “during the last decade we had a Civil Rights Commission that had a director and a majority who worked against civil rights.”140 Chairman Hawkins offered the same observation, arguing that the employment discrimination laws had to be amended to better mobilize private enforcers because the “EEOC and the other law enforcement agencies of the Administration are not effective. They are led by individuals who do not even believe in their own law,”141 and who do not “believe in civil rights.”142 According to Hawkins, speaking as chair of the committee with oversight authority over the EEOC, oversight efforts had failed, and he regarded further attempts at oversight as futile. In his view, the agency was under the control of his ideological opponents, and the committee had been unable to monitor and control it effectively. Said Hawkins in the House hearings: The Subcommittee on Employment Opportunities of the Committee on Education and Labor has gone through this charade for almost a decade. We have given up even investigating the EEOC because it is useless to investigate the agency because they obviously are not going to protect against the type of [discrimination] cases that we have heard this morning. . . . I was surprised this morning that one of the corporate lawyers argued in favor of EEOC, defending EEOC. A rather strange case it seemed to me. It seemed that when they got control of EEOC, then the EEOC was okay because they had control of it. . . . In the last 10 years we have become so disgusted with what EEOC was doing that we discontinued monitoring it.143 Hawkins’s skepticism about congressional control of the EEOC was shared by interest groups testifying in favor of the act, who believed that an enhanced private enforcement regime was a preferable way to resolve perceived enforcement deficiencies. For example, Ellen Vargyas of the National Women’s Law Center urged, in her committee testimony, adoption of increased monetary damages in order to mobilize private enforcers in light of the EEOC’s weakness, and the great difficulty congressional Democrats would have in controlling it. [T]he experience with [EEOC] conciliation shows an approximately 13% success rate. This is not, in my view, something to be proud
196
CHAPTER 6
of. . . . Now, we certainly believe that . . . this committee has been in the forefront of efforts to make the EEOC seem more responsive, to give it more resources to help it along. But the solutions of an improved EEOC and a damages remedy are by no means mutually exclusive. Unfortunately, the experience in trying to have a more successful EEOC has taught us that if, in fact, it can be achieved, it will not be achieved easily.144 Civil rights advocates believed that, with the agency in the hands of their ideological adversaries, oversight by Congress was of limited value. Accordingly, they sought to mobilize private litigants and attorneys to carry out their policy goals. Hawkins’s observation that at least some advocates for business interests had become converts to bureaucratic regulation through the EEOC finds support in the record. Just as civil rights groups’ views on the relative importance of agency powers versus private litigation had changed a great deal since 1964, the views of conservative Republicans and business interests now also showed some movement. They argued that the proposed changes to Title VII’s private enforcement regime would radically alter the existing implementation scheme, tilting it further toward private litigation and away from bureaucracy, regrettably undermining the power of the EEOC. Republicans, management lawyers, business interest groups, and conservative academics argued that bureaucratic regulation would be a more effective policy tool than economic incentives for selfhelp by private actors through litigation, and thus, to the extent that there was an underenforcement problem, it should be addressed by giving the EEOC more money and power, thereby expanding its bureaucratic capacity. Opponents of incentivizing more private litigation argued that the remedial scheme of Title VII, as originally enacted, represented a deliberate choice of an administratively centered implementation model, founded on voluntary conciliation, rather than one based primarily upon litigation. The modest economic damages available to plaintiffs (back pay only) encouraged them to sincerely participate in the search for a voluntary administrative resolution. To substantially increase the economic value of claims with compensatory and punitive damages, to be awarded by juries, would markedly diminish plaintiffs’ and their attorneys’ willingness to submit voluntarily to resolutions supervised and crafted by the commission. Instead, plaintiffs and their attorneys would “hold out for a bigger payday in federal court,” as one management lawyer put it.145 According to Illinois Republican Harris Fawell, the net result, deplorably, would be “gutting the EEOC.”146 A representative of the National
CIVIL RIGHTS ACT OF 1991
197
Association of Manufacturers likewise testified in reference to the new damages provisions: Were this section enacted, any hope of conciliation and settlement through the EEOC would vanish. . . . [A]ttorneys would hold out to individuals the promise of six or seven figure judgments, with the accompanying six or seven figure legal fee. The purpose and function of the EEOC would be effectively terminated.147 This lament about the untoward weakening of administrative power to regulate business was repeated by other conservative Republicans legislators,148 representatives of the Bush administration,149 business interest groups,150 and management lawyers.151 Those most opposed to increasing incentives for private litigation suggested an alternative solution. While there were legitimate concerns about inadequate enforcement by the EEOC, some Republicans and their constituents argued, rather than diminishing executive power and effectively transferring more implementation authority to private litigants, lawyers, and courts, the better solution would be to give the agency more power and money. House Republican Steve Bartlett of Texas suggested that the EEOC could be made a more effective enforcer if it were given power to administer an alternative dispute resolution mechanism with teeth, such as binding arbitration, which is cease-and-desist by another name.152 House Republican Carlos Moorhead of California advocated an approach that, instead of increasing incentives for private litigation, “strengthens the existing settlement process under the Equal Employment Opportunity Commission.”153 Witnesses testifying on behalf of employer interests, acknowledging the agency’s weakness as an enforcer, also repeatedly took the same position in hearings. “[S]trengthen the EEOC’s existing administrative capabilities,” and “provid[e] sufficient enforcement capability to the EEOC to assure prompt investigation of claims,” counseled one management lawyer.154 “[B]eef up the administration of the statute, [and] see to it that claimants and victims of discrimination have an agency they can turn to for relief, they have confidence in, and an agency that has what it takes to enforce this statute as it was intended,” implored another.155 If necessary, restructure the agency as so to strengthen it, advised an economist testifying against the new damages provisions.156 Just don’t generate more lawsuits, they all agreed. In the end, it was a small minority of Republicans and their constituents that advocated addressing enforcement problems by shoring up the EEOC’s administrative power and providing it more resources. Most Republicans who opposed incentivizing more private lawsuits did not advocate the alternative of administrative state-building. It is significant and
198
CHAPTER 6
ironic, nevertheless, that in 1989–91 the only voices speaking in favor of strengthening the EEOC in congressional hearings and floor debates were conservative Republicans and witnesses representing employer interests. Surely Hawkins was at least partly correct that “when they got control of EEOC” they regarded it as less of a menace. Moreover, during the same period that administrative enforcement had abated during the Reagan-Thomas years, job discrimination lawsuits by private parties had grown considerably, plateauing in the mid-1980s at a level about 59 percent higher than during the Carter years. The prospect of triggering a new upward spiral in this litigation threat seemed worse, at least to some Republicans and employer interests, than strengthening the agency. They lost the fight, however. Roll Calls and Voting Alignments on Private Enforcement The partisan divisions in the legislative proceedings leading to the CRA of 1991 were sharp. The magnitude of the divisions that characterized the parties’ positions during the 1990 debates is well represented by the final roll calls on the bill that passed both chambers in 1990 but was vetoed by the president. The basic patterns in the two chambers were the same. Aggregating across both, Democrats voted in favor of the 1990 bill at a rate of 92 percent, and Republicans voted against at a rate of 80 percent.157 While Southerners in both parties were more likely to vote against the bill than other members of their parties, this tendency did not affect the basic party alignments.158 This partisan polarization on private civil rights enforcement is a clear reversal of voting patterns that prevailed from 1964 to 1976. Interestingly, however, the combination of moderating revisions in the 1991 Danforth bill that passed, alongside electoral pressures faced by Republicans, transformed partisan divisions into cross-party unity of even greater magnitude than in the earlier votes on private civil rights enforcement. In the end, Republican support for incentives for private civil rights lawsuits was stronger than ever before. On passage of the CRA of 1991, aggregating across the two chambers, Democrats voted in favor at a rate of 98 percent, Republicans did so at a rate of 82 percent,159 and Southerners were not a distinct voting bloc.160 Interest Groups on the Scene Lawyers completely dominated the House and Senate hearings, ranging from representatives of legal advocacy organizations, to private practitioners, to law professors. Testifying in support, representatives of civil rights legal advocacy organizations (e.g., the NAACP and the Women’s Legal Defense Fund) were by far the leading force, making twenty-three
CIVIL RIGHTS ACT OF 1991
199
appearances between both chambers. These were the groups that helped draft the bill originally introduced by Kennedy and Hawkins in 1990. The private for-profit bar played a far lesser role among forces supporting passage. Representatives of the American Bar Association and the National Employment Lawyers Association (an association of predominantly for-profit plaintiffs’ employment lawyers), and four for-profit plaintiffs’ civil rights lawyers were invited to testify in favor.161 Thus, among supporters of the bill, appearances by nonprofit civil rights group representatives outnumbered those of for-profit sector lawyers by about four to one. In opposition to the bill, several conservative nonprofit legal advocacy organizations (e.g., Washington Legal Foundation), and several national business organizations (e.g., the National Association of Manufacturers) testified. Interestingly, considerably more employment lawyers from the for-profit sector opposed the law than supported it: eleven managementside lawyers appeared in opposition to passage.162 In contrast with hearings on laws examined in the last two chapters, two changes are notable. First, the number of specifically legal advocacy organizations focusing on civil rights had grown significantly in number and had become by far the biggest presence in the hearings advocating in favor of incentives for private enforcement litigation. Second, whereas in the earlier hearings there was scarcely a single witness that expressed concern about excessive private civil rights litigation, in these hearings such witnesses were a major presence, and for-profit management-side lawyers were easily the largest contingent among them.
The Effects of the CRA of 1991 on Private Enforcement What were the consequences of the legislative struggles traced in this chapter? The CRA of 1991 provides a unique opportunity to evaluate the actual effects of the details of a private enforcement regime on rightsasserting behavior. When a right and its private enforcement regime are enacted at the same time, as is typically the case, it is impossible to know what level of private enforcement would have obtained if the same right had been embedded in a different (weaker or stronger) private enforcement regime. That is, it is not possible to assess the actual effects of particular economic incentives in a private enforcement regime on the mobilization of enforcers. Cases such as the CRA of 1991, in which there was an existing set of substantive rights embedded in some particular private enforcement regime, and then a subsequent change in important elements of the private enforcement regime, provide a natural experiment in which the effects of the changed elements on claiming behavior can be
200
CHAPTER 6
evaluated. A longitudinal study I conducted of the volume of private Title VII claims, from 1980 to 2002, finds that the rise in private enforcement that the bill’s supporters promised and its opponents feared was, in fact, brought about by the CRA of 1991.163 Before discussing the study’s findings, a note is in order about the data analyzed. Existing data on federal employment discrimination litigation is aggregated across all federal statutes barring job discrimination—with most claims arising under Title VII, the Age Discrimination in Employment Act (ADEA), and the Americans with Disabilities Act (ADA)—into a single “job discrimination” category. While Title VII claims are by far the most common within this category, claims under other statutes are significant in number as well. Federal litigation data is thus too broadgauged to systematically assess the effects of the CRA of 1991’s amendments to Title VII.164 One particularly confounding consideration is that Title I of the ADA of 1990, which prohibits employment discrimination based upon disability, became effective in July 1992, thus producing a surge of a new type of federal employment discrimination litigation shortly after the CRA of 1991 was passed. It is not possible with federal litigation data, lumped into the single “job discrimination” category, to assess whether any post-1991 increases in litigation resulted from the CRA of 1991’s changes to Title VII, the newly passed Title I of the ADA, other laws, or some combination.165 It does bear noting, though, that there was in fact an abrupt and steep increase in job discrimination lawsuits filed in federal courts immediately following enactment of the CRA of 1991, as reflected in figure 6.1.166 The number of private job discrimination suits increased by 211 percent in the six years following passage. While ADA claims certainly contributed to this increase, they cannot be the primary explanation. Among EEOC charge filings during this period, which are a precondition to proceeding with litigation under Title VII, the ADEA, and the ADA, disability claims were asserted in approximately 18 percent of the charges.167 Disability claims under the new ADA, representing this comparatively small fraction of total job discrimination claims before the EEOC, cannot explain the 211 percent increase in private lawsuits during the same period. The sudden and powerful growth of federal job discrimination litigation following passage of the CRA of 1991 is highly consistent with the efficacy of that law as an act of litigant mobilization, even if the federal litigation data is too broad-gauged to systematically establish the law’s effects. A high-quality alternative measure of private Title VII enforcement exists in the form of the number of charges of discrimination filed with the EEOC. As discussed more fully in chapter 4, as a legal precondition to filing an action in court, prospective plaintiffs under Title VII are required to file a charge with the EEOC and give the commission an opportu-
201
CIVIL RIGHTS ACT OF 1991 Total EEO litigation 25,000
20,000
15,000 1991 10,000
5,000 1975
1980
1985
1990
1995
2000
2005
Figure 6.1 Employment Discrimination Litigation in Federal Trial Courts, 1977–2005
nity to conduct an investigation and facilitate voluntary settlement, after which the complainant is entitled to proceed to court as a matter of right whatever the agency’s determination.168 EEOC charge data is maintained separately for each statute administered by the agency, and includes the nature of the discrimination alleged, and thus it permits one to isolate claims under Title VII, and among Title VII claims, to isolate claims on the basis of race, national origin, gender, and religion. A substantial proportion of federal employment discrimination claims are settled after an EEOC charge is filed because of the threat of litigation, but without formal litigation being instituted. Indeed, it is well recognized that potential liability “in the shadow of the law,” without formal legal action, profoundly shapes whether and how cases settle without litigation.169 Because EEOC filings reflect the fulfillment of a formal legal precondition to litigation, they capture both cases that ultimately enter litigation and those that settle after the threat of litigation is invoked. In the study I found that passage of the CRA of 1991 brought about a statistically and substantively significant increase in Title VII charges filed in all four protected categories (gender, race, national origin, and religion).170 In aggregate, it increased total Title VII charges by 58 percent. Though we should recall all of the qualifications already made about the limits of data on federal job discrimination lawsuits, figure 6.1 suggests the possibility of a considerably larger percentage increase in Title VII lawsuits brought about by the CRA of 1991. One plausible explanation
202
CHAPTER 6
for a greater percentage rise in litigation, as compared to the percentage rise in EEOC charges, is that increasing the expected value of Title VII claims in the CRA of 1991 may well have increased both the number of EEOC charges filed and the proportion of EEOC charges that resulted in a lawsuit. In addition to this basic finding regarding the volume of charges, the study yielded two other findings that are relevant to the larger themes of this book. First, it found that movement in the volume of Title VII charges over time was significantly correlated with movements in federal courts’ ideological preferences, utilizing the same measure of judicial ideology employed and described in chapter 3 of this book. Movements in judicial ideology in the conservative direction reduced the volume of Title VII charges: the more conservative the judiciary, the less likely claimants are to file charges of discrimination. One implication of this finding is that presidents can affect litigant mobilization under Title VII through judicial appointments. As we saw earlier in this chapter, in advocating for the CRA of 1991, civil rights groups and their allies in Congress complained that the cumulative effect of judicial appointments during the Reagan and Bush years was a judiciary inhospitable to their cause.171 The study found that the net increase in judicial conservatism during the eleven-year period from the time Reagan assumed the presidency in 1981 until passage the CRA of 1991, was associated with reducing quarterly Title VII charges by an estimated 29 percent. While certainly considerable, the magnitude of this effect on the volume of Title VII charges from over a decade of Republican judicial appointments is only half of that achieved, in the opposite direction, by the single legislative event of the CRA of 1991. The significant relationship between judicial ideology and claiming behavior supports one of the empirical suppositions underpinning the judicial friendliness and the judicial hostility hypotheses: that judicial ideology can significantly influence litigant mobilization or demobilization. It also provides corroborating evidence for the rational litigant model because it is consistent with the contention that plaintiffs and their attorneys are able to perceive movements of judicial ideology, and its effects on the behavior of judges and the doctrine they produce, and factor it into their assessment of the expected value of claims. A second interesting finding of the study is that while ideological movement in the judiciary influenced the volume of Title VII charges, this was not true for movement in the ideological preferences of the EEOC’s leadership, measured based upon the ideology of the appointing president. This is consistent with evidence presented in this chapter indicating that, in tandem with the Reagan administration’s demobilization of
CIVIL RIGHTS ACT OF 1991
203
the administrative enforcement apparatus, private enforcement litigation actually increased, appearing to be largely outside the control of the executive branch. Private enforcers kept coming, knowing that regardless of the agency’s judgment of their claim, they would be free to litigate. We know from the last two chapters that the existence of a private enforcement mechanism largely outside the control of the executive branch was consciously built into the governing statutory framework in the legislative compromise of 1964, and was consciously entrenched and consolidated in the legislative struggles over civil rights enforcement during the ensuing decade. The evidence presented in this chapter suggests that Title VII’s private enforcement regime succeeded, to a significant extent, in insulating enforcement from executive interference. In assessing the effects of the CRA of 1991, it is also noteworthy that, alongside the powerful post-1991 growth in private Title VII enforcement, membership in the National Employment Lawyers Association— the largest association of plaintiffs job discrimination lawyers—also grew by leaps and bounds. While membership had hovered around 900 or 1,000 lawyers in the few years leading up to the act, it jumped from about 1,000 members in 1991 to 1,700 in 1993, to 2,200 in 1995, to 3,200 in 1997, where it roughly plateaued (at about the same time that the post-1991 growth in federal job discrimination lawsuits also plateaued, per figure 6.1).172 In the six years following passage of the CRA of 1991, when private federal job discrimination lawsuits increased by 211 percent, NELA membership increased by about 220 percent. Job discrimination litigation as a practice specialty grew, and the job discrimination bar became more organized. Part of Congress’s explicit motivation for increasing the economic value of Title VII claims was to grow the infrastructure of private for-profit lawyers ready, willing, and able to prosecute Title VII claims on behalf of plaintiffs. The rising number of lawyers specializing in representing job discrimination plaintiffs, in tandem with that bar’s growing enforcement activity, was purposefully fostered by Congress with the goal of compensating for what it viewed as agency failures and judicial hostility. This leaves open an important question about the efficacy of private Title VII lawsuits as a regulatory tool to advance equality. What were the actual effects of that private enforcement activity on the ground, where people work? The introductory chapter noted that researchers have found considerable evidence that private lawsuits can be an effective tool in shaping the behavior of the regulated population across many policy domains. The findings have established both “specific deterrence” and “general deterrence” effects, where specific deterrence influences the future conduct of the entity sued, while general deterrence refers to ef-
204
CHAPTER 6
fects of lawsuits against members of the regulated population on other would-be violators.173 This body of research includes a number of studies specifically focusing upon job discrimination litigation. Alexandra Kalev and Frank Dobbin, using longitudinal data spanning 1971 to 2002 on employment opportunities for women and racial minorities in a large sample of private organizations, find that women and racial minorities made gains within organizations after the organizations were subjected to job discrimination lawsuits.174 In work examining a sample of both private and public organizations between 1971 and 2005, I have found that, independent of the effects of actually being sued, the rate of job discrimination litigation in the geographic areas in which organizations do business is a significant factor driving their adoption of policies calculated to increase employment opportunities for women and racial minorities—polices that have been demonstrated to actually be effective in achieving that goal.175 Charles Epp, studying a large sample of local governmental employers, finds that litigation “support structures” in the local community—measured as the size of the bar in the locality specializing in suits against government—were positively associated with governmental employers’ propensity to adopt serious measures directed at compliance with prohibitions against workplace sexual harassment.176 Together these studies, arriving at similar conclusions using a variety of different research designs and measurement strategies, suggest that private job discrimination litigation has been effective in advancing the policy goals of Title VII of the CRA of 1964.
Conclusion Implications of the Process-tracing Evidence I next assess how the process-tracing evidence in this chapter fits with the theories and evidence presented previously in the book. The role of the process-tracing evidence is to allow detailed interrogation of what caused Congress to enact a private litigation-centered enforcement regime; to facilitate assessment of how well the observed causes in the particular case support the causal theories of legislative enactment of private enforcement regimes presented in chapters 2 and 3; and to permit appraisal of how consistent the observed causes are with the empirical relationships demonstrated in the model presented in chapter 3. Legislative-Executive Conflict. The evidence in this chapter strongly supports the logic of the legislative-executive conflict hypothesis, which holds that conflict between Congress and the president over control of the bureaucracy can encourage legislative reliance upon private enforcement
CIVIL RIGHTS ACT OF 1991
205
regimes. This hypothesis was also supported by the legislative-executive conflict variables in the empirical models presented in chapter 3, and the process-tracing evidence in each of the last two chapters. The election of Ronald Reagan produced a polarizing realignment between Congress and the executive branch on civil rights. The 1980s were marked by implacable conflict between congressional Democrats and the Reagan administration over civil rights in general, and over control of the EEOC in particular. The EEOC was subject to near-continuous and highly combative oversight hearings throughout the decade, where Democrats charged that agency leadership was “undermining,” “obstructing,” and “sabotaging” the agency’s mission, to the point that Democratic leadership was “disgusted.” Congressional Democrats and civil rights groups were bitterly dissatisfied with the agency’s performance, which they regarded as ideologically biased in favor of employers, and which they saw as resulting from presidential subversion. To address perceived enforcement deficiencies, a Democratically controlled Congress, working closely with civil rights groups, statutorily bolstered Title VII’s private enforcement regime, increasing the economic value of claims for the stated purpose of escalating the mobilization of private litigants and lawyers to do what, in their view, the agency would not. Judicial Ideology. Judicial ideology was a significant factor driving the passage and content of the CRA of 1991. It was observed in chapter 2 that, from the standpoint of theory, congruence of legislative and judicial policy preferences should make private enforcement regimes more attractive to Congress, other things being equal (the judicial friendliness hypothesis). It was also observed that divergence of legislative and judicial preferences can also be counterbalanced, when Congress is endeavoring to mobilize litigants, by enacting ever more robust private enforcement regimes (the judicial hostility hypothesis). In strengthening Title VII’s private enforcement regime in the CRA of 1991, a Democratically controlled Congress was explicitly responding to Supreme Court decisions that it harshly characterized as ideologically motivated conservative retrenchment, and as hostile to civil rights. That is, congressional Democrats were motivated to bolster Title VII’s private enforcement regime by what they regarded as judicial hostility toward their policy preferences. It may at first seem ironic that congressional Democrats, working with civil rights groups, responded to the offending Supreme Court decisions not only by overruling their substance, but also by increasing incentives for plaintiffs to take their cases to the very judiciary that they regarded as hostile. On closer scrutiny it is readily explicable. As the judiciary moved further away from Congress in elaborating substantive Title VII law through its rule articulation function, Democrats were still able to
206
CHAPTER 6
increase execution of the rule enforcement function by statutorily increasing the expected monetary value of Title VII claims. What was their alternative? To address perceived enforcement deficiencies by enhancing administrative power, as some Republicans and business interests suggested? To civil rights advocates, the EEOC did not look like a more promising source of rule articulation than courts, and it certainly looked like a far less promising source of rule enforcement than adequately incentivized private lawsuits. From their point of view, therefore, enhancing Title VII’s private enforcement regime appeared to be a better move than bureaucratic state-building. This episode is useful in suggesting an answer to a puzzle posed by the theory and evidence presented earlier in this book regarding the effects of judicial ideology on the legislative choice of private enforcement regimes. The judicial ideology variables, measuring Congress’s distance from the judiciary, and measuring whether the judiciary was moving toward or away from Congress, proved insignificant in the empirical model in chapter 3, supporting neither the judicial friendliness nor the judicial hostility hypotheses. We were unable to judge, then, whether legislators and interest groups simply did not weigh judicial ideology when crafting enforcement regimes, or whether the effects of the countervailing hypotheses were canceling one another out in the data. In the historical evidence in the previous two chapters, policy actors (first on the right, then on the left) supported court-based implementation based partly upon the expectation that the judiciary would be an ideologically favorable venue, supporting the judicial friendliness hypothesis. The historical evidence in this chapter clearly showed the judicial hostility hypothesis in operation. On balance, then, the process-tracing evidence supports the significance of judicial ideology to the choice of private enforcement regimes, demonstrating that, from Congress’s point of view, both judicial friendliness and judicial hostility can operate, in quite different ways, to encourage the choice of private enforcement regimes. Observing support for both of the countervailing hypotheses also provides a plausible explanation for the insignificance of the judicial ideology variables in the empirical model in chapter 3, counseling against interpreting that finding to mean that judicial ideology is insignificant to actors constructing regulatory enforcement regimes. Veto Points and Compromise. The veto points and compromise hypothesis holds that the veto-ridden character of American lawmaking institutions can encourage private enforcement regimes as a compromise alternative to bureaucratic state-building. The 1989–91 debates focused upon whether to bolster an existing private enforcement regime, and the prospect of administrative state-building was never seriously before Congress.
CIVIL RIGHTS ACT OF 1991
207
The dynamics predicted by this hypothesis thus were not implicated. The underlying logic of the veto points and compromise hypothesis did enter the picture, however, when President Bush’s veto of the CRA of 1990 led to a winnowing down of the litigant mobilizing incentives that made it into the CRA of 1991: the compromise was less, rather than more, litigant mobilization. Policy Drift. The coalition drift hypothesis holds that current majority coalitions in Congress enact private enforcement regimes to insulate desired levels of enforcement of their regulatory laws from subversion by future majority coalitions with different and possibly distant preferences. The empirical model in chapter 3 supported this hypothesis, finding impending seat share losses by the majority party (electoral uncertainty) to be significantly associated with enactment of private enforcement regimes. However, as with the historical evidence leading to passage of Title VII, and surrounding the entrenchment and expansion of private civil rights enforcement in the 1965–76 period, the historical evidence leading to passage of the CRA of 1991 did not disclose that advocates of the law were motivated by concern about subversion by future legislative majorities. The bureaucratic drift hypothesis holds that current majority coalitions in Congress enact private enforcement regimes to insulate enforcement from subversion by bureaucrats pursuing their own personal preferences, conceived broadly to include not only bureaucrats’ ideological preferences, but also timidity, apathy, shirking, careerism, and capture. Attacks on the EEOC by congressional Democrats and civil rights groups went beyond claims that it was implementing a Reagan-Bush agenda in favor of employers and against civil rights enforcement—that is, they went beyond claims of executive subversion. The attacks crossed over into claims that the agency was simply hapless, ineffectual, “useless,” and characterized by “lack of diligence and ineptitude.” They also included charges that the agency had been captured by employer interests. Such claims, offered in support of legislation bolstering Title VII’s private enforcement regime, reflect the concerns animating the bureaucratic drift hypothesis. Issue Groups. The evidence in this chapter supports the issue group hypothesis, which maintains that issue groups, motivated by distrust of bureaucracy, are a significant source of demand for enactment of private enforcement regimes. This hypothesis was supported by the empirical model in chapter 3, in conjunction with the historical evidence for the 1965–76 period. Civil rights groups powerfully and effectively mobilized in response to the Supreme Court’s 1989 decisions, thereby placing the issue of Title VII amendments on the legislative agenda; they secured the
208
CHAPTER 6
support of important and sympathetic Democratic legislators with whom they had close ties; and under the organizational umbrella of the Leadership Conference, they worked hand-in-hand with key legislative committee staff in crafting a bill to strengthen Title VII’s private enforcement regime, many important elements of which were contained in the law ultimately passed. Like the historical evidence in the 1965–76 period, the evidence in this chapter showed that distrust of bureaucracy was among civil rights groups’ key motives in pressing for greater incentives for private enforcement. Some of the distrust clearly arose from the concerns identified by the issue group hypothesis. As just noted in the discussion of bureaucratic drift, civil rights groups attacked the EEOC as hapless, ineffectual, and inept. However, even more than in the Nixon years, issue groups’ distrust of bureaucratic power went beyond generalized loss of faith in bureaucracy, and was very importantly fueled by particularized distrust of the existing executive regime as ideologically hostile to their policy agenda. In this sense, as in the last chapter, the operation of issue group demand was entwined with legislative-executive conflict dynamics. Civil rights groups—with a long-standing preference for private enforcement regimes traced in the last chapter—were an important component of the governing Democratic party coalition in Congress. Civil rights groups and congressional Democrats together faced a conservative Reagan-Bush White House, and their shared distrust and antagonism toward it motivated their alliance in support of bolstering the private enforcement alternative. Rent-Seeking Lawyers. The rent-seeking lawyer hypothesis—which holds that lobbying by lawyers pursuing money is an important reason that Congress enacts private enforcement regimes—is not well supported by the historical evidence. It does appear, though, that the for-profit plaintiffs’ civil rights bar, as in the mid-1970s, may well have provided an auxiliary source of power to civil rights groups in the legislative process. The empirical model in chapter 3, and the historical evidence surrounding passage of Title VII, provided no support for the rent-seeking lawyer hypothesis. The for-profit plaintiffs’ civil rights bar that emerged in the first half of the 1970s clearly was part of the mobilization, orchestrated by civil rights groups, to secure fee shifting across the entire domain of civil rights statutes, though the causal significance of the for-profit civil rights bar in producing the legislative outcomes was ambiguous. The forprofit job discrimination bar continued to grow during the 1980s—apparently taking up, in part, enforcement opportunities opened up by the Reagan administration’s less interventionist use of the civil rights bureaucracy—and became more organized, with the formation of the National Employment Lawyers Association in 1985.
CIVIL RIGHTS ACT OF 1991
209
In the hearings leading to the CRA of 1991, in contrast with those leading to the Civil Rights Fees Act of 1976, the for-profit lawyers invited to testify were split between support for and opposition to the law, with most opposing. While defense-side lawyers no less than plaintiffsside lawyers stood to profit from increasing levels of job discrimination litigation, each group testified on the side of the law favoring its clients: the defendants’ lawyers opposed the law and the plaintiffs’ lawyers supported it. This evidence is consistent with the two bars having distinct ideological and policy orientations, and with them having commitments to different client populations. It is also possible that defense lawyers enhanced their professional standing (and thus economic interests) by testifying against the law, and that plaintiffs lawyers, consistent with the rent-seeking lawyer hypothesis, sought to foment litigation to enrich themselves. The testimony of the latter group in support of the bill is certainly consistent with the rent-seeking lawyer hypothesis, even if it is also consistent with other quite different explanations grounded in the pursuit of ideological, policy, and client goals. Although it is difficult to evaluate the effects of for-profit lawyers on the legislative outcome in 1991, as with the entrenchment and expansion of civil rights fee shifting in the first half of the 1970s, it is abundantly clear that civil rights groups were the catalysts and agenda setters. It was they who put the issue in front of Congress. When the quite large drafting committee was constituted, it was exclusively composed of nonprofit groups within the Leadership Conference on Civil Rights, and the forprofit plaintiffs bar was not given the opportunity to work with committee staff in crafting the bill. In the hearings, appearances in support of the bill by representatives of nonprofit civil rights groups outnumbered those of for-profit plaintiffs’ lawyers four to one. It may be that the for-profit civil rights bar added weight and power to civil rights groups’ lobbying efforts, and it may be that this support was pivotal to passage of the CRA of 1991. Believing the latter to be true would entail the historical counterfactual claim that civil rights groups could not have succeeded without the for-profit civil rights bar, a claim that seems to me highly improbable based upon the evidence presented, though not demonstrably false. Even if the for-profit civil rights bar was a necessary element of the winning coalition, issue groups pursuing their policy programs were undoubtedly the architects and leaders of the successful campaign to enhance Title VII’s private enforcement regime. If rent-seeking activity by for-profit civil rights lawyers played a causal role, it was to increase the power of civil rights groups in the legislative process. Party Alignments. The weak form of the party alignment hypothesis is supported by the evidence, and the strong form of it is rejected; this is
210
CHAPTER 6
consistent with the empirical model presented in chapter 3 and the historical evidence presented in the last chapter. The weak form of the hypothesis predicts that Democrats will be more likely to support private enforcement regimes than Republicans, and the strong form predicts consistent Republican opposition. Democrats in both chambers, working closely with civil rights group constituents, introduced the bills in 1990 that initiated the legislative process leading to the CRA of 1991, and Democrats were consistent supporters of stronger versions of litigantmobilizing elements of bills throughout the legislative process. As contrasted with the bipartisan support for private civil rights enforcement litigation between 1964 and 1976, during most of the legislative process, Republicans’ posture toward private Title VII enforcement appeared to have changed dramatically: many within the party attacked it vociferously. During the 1980s, the question of the desirability of private civil rights lawsuits—and litigation more generally—had emerged as a locus of intense party conflict. By the early 1980s, the Reagan administration sought reforms to reduce it. Republican opposition in 1989–91 appears readily explicable in terms of the interests openly pursued by important business constituents. At least in part as a result of growing levels of Title VII litigation and the imposition of associated costs, national business organizations mobilized forcefully against the proposed bill. On the interest group front, this was by far the most striking new development in 1989–91. Business groups collaborated with Republicans in opposition to the bill as energetically as civil rights groups collaborated with Democrats in support. However, paradoxically, in the vote on final passage, over 82 percent of Republicans voted in favor of the CRA of 1991’s litigant-mobilizing incentives, an even wider margin of Republican support than for any such provisions between 1964 and 1976. Moderate Republican Danforth introduced the compromise bill that became the CRA of 1991. Alongside the preferences of the business community, Republicans had to balance broader electoral pressures that caused them to want to avoid appearing hostile to civil rights, replicating an electoral dynamic observed in the CRAs of 1964 and 1968, and arguably in the School Aid Act of 1972. In response to their business constituents, they significantly mitigated some litigant-mobilizing elements of civil rights advocates’ initial proposal— such as by capping monetary damages and retaining Title VII’s very short statute of limitations—and in response to the broader electoral pressures they overwhelmingly voted for the compromise Danforth bill. Despite all the partisan rancor surrounding the legislative events of 1989–91, in an era of partisan polarization over public policy in general, over civil rights in particular, and over litigation as well, the CRA of 1991 was remarkable for the overwhelming bipartisan support it received.
CIVIL RIGHTS ACT OF 1991
211
Budget Constraints. The process-tracing evidence does not support the budget constraint hypothesis. While there were a few scattered references during the hearings and floor debates to the EEOC being overburdened and underresourced, this theme was not a significant part of the debate. Budget deficits aside, it is clear that the civil rights liberals who were the catalysts behind the CRA of 1991 simply did not regard the EEOC, even with ample resources, to be a plausible solution to the problem of underenforcement. Title VII’s Private Enforcement Regime, Unintended Consequences, and Republican Regret? The origins and development of federal job discrimination litigation raise a question about how the ultimate policy outcome fits with the intent of Title VII’s designers in 1963–64. There is no question that Title VII’s private enforcement regime proved, over time, to be far more robust than either conservative Republicans or liberal Democrats had expected. However, it would be a mistake to regard these effects as “unintended consequences” in the sense, often used by institutional scholars, of producing important effects in “realms outside those originally intended.”177 The purpose of the private enforcement regime was to provide for enforcement through private litigation, and its details were the subject of extensive scrutiny and controversy. Civil rights advocates insisted on fee shifting (among other provisions) with the specific aim and hope of mobilizing private enforcers. Civil rights opponents (Southerners) sought to reduce or eliminate litigant-mobilizing provisions that they warned would unleash a stampede of “ambulance chasing” lawyers “chosen and controlled by the NAACP,” but they were defeated. That the volume of Title VII litigation far exceeded the expectations of Title VII’s designers is a story of self-reinforcing policy feedback, not unintended consequences: the effects occurred in precisely the realm intended, but their magnitude, over time, was far greater than expected. The question naturally arises whether the conservative Republican defeat of liberal administrative state-building, in favor of an enforcement framework that ultimately generated a huge number of lawsuits against employers, was a mistake from their own point of view. If Republicans got more civil rights litigation than they bargained for, does this mean that they made a mistake, from the antiregulatory point of view of pivotal Republicans in 1963–64, in substituting private lawsuits for administrative power in Title VII? Did they end up worse off than if they had just accepted the liberals’ cease-and-desist administrative framework? If they had known then what we know now, would they do it again? Considering this question entails contemplating speculative historical counterfac-
212
CHAPTER 6
tuals. I briefly weigh it from the vantage point of the full transformation from H.R. 405 into the substitute bill offered by Dirksen that ultimately passed, as opposed to trying to parse more finely the effects of particular Republican changes along the way.178 Any answer, it seems, would turn on whether the EEOC created by H.R. 405 was adequately funded and politically supported. We know, of course, that the actual EEOC was not; however, the question arises whether, in the absence of the private lawsuit alternative, there would have been more broad-based pressure to fund and politically support an adjudicatory agency with cease-and-desist powers that was the exclusive source of enforcement. A strong agency would have been able to administratively adjudicate, at the state’s expense, a much larger number of complaints than the number of full-blown federal lawsuits that have actually been litigated. With the administrative model proposed in H.R. 405, an individual employee, without legal representation, could file a complaint and the Office of the Administrator would conduct an investigation. If the Office of the Administrator regarded the complaint as meritorious and could not achieve voluntary settlement, the administrator could then prosecute the complaint before the board. In the private litigation-centered alternative actually adopted in Title VII, the investigative and prosecutorial responsibilities were transferred to the employee. To have any meaningful chance of success, the employee has to retain counsel to carry out these functions in the much more procedurally complex, and costly, arena of federal litigation. Under this framework, even with the costs of investigation and prosecution being recoverable by winning plaintiffs, only a small fraction of total complaints initiated with the EEOC actually end up being prosecuted in court. A generously funded and aggressively mobilized agency with cease-and-desist powers could, in theory, have posed a greater enforcement threat than private lawsuits. It is, however, also possible—and I believe more probable—that even in the absence of private lawsuits to carry the enforcement burden, the EEOC would not (or at least not consistently) have been given the fiscal and political support necessary to undertake the gargantuan task of investigating and adjudicating the job discrimination claims arising out of the entire American workforce. In a historical counterfactual H.R. 405 scenario with a feeble agency, the agency would represent a much less forcefully interventionist implementation scheme than Title VII’s actual private litigation framework. Therefore, one’s conjecture about what congressional and executive support would have been for an NLRB for civil rights, without lawsuits, must necessarily inform one’s assessment of whether pivotal Republicans made a mistake, from an antiregulatory point of view. History most definitely does tell us what civil rights liberals believed about which historical counterfactual was more probable. Their
CIVIL RIGHTS ACT OF 1991
213
fierce defense of Title VII’s private enforcement regime in the late 1960s and early 1970s, even at the expense of an NLRB for civil rights, was clearly predicated significantly on a palpable loss of faith in the administrative state, and their growing antipathy toward those in control of it. It is tempting to infer from civil rights liberals’ strong embrace of Title VII’s private enforcement regime that this enforcement approach redounded to the detriment of antiregulation Republicans and business interests, suggesting that they made a mistake in 1963–64. It must be kept in mind, however, that in 1976 nonsouthern Republicans voted to ratify and extend the private enforcement bargain struck by their party brethren in 1963–64, and did so by a comparably wide margin. They did so despite the fact that, from civil rights liberals’ point of view, a fairly clear picture had emerged by 1976 of the assertiveness of Title VII’s private enforcement regime as an instrument of intervention. To be sure, by the early 1980s the Reagan administration sought to curtail private civil rights litigation. By 1989–91, Republicans’ tone toward private civil rights litigation had grown caustic, and their votes for the CRA of 1991, in the face of powerful electoral pressures, appeared to belie deep antagonism toward private job discrimination lawsuits. But even then, Republicans were not inclined to revisit the foundational compromise of 1963–64 repudiating administrative power in favor of private lawsuits. While the prospect of cease-and-desist was floated during the committee proceedings, and while a few Republicans did suggest that strengthening the agency would be a better solution to Title VII enforcement problems than incentivizing more private lawsuits, this was a minority view within the party, and no concrete proposals of this sort were actually offered. The Republican mind-set in 1989–91 was to limit lawsuits, not to bolster administrative power as an alternative to lawsuits. Thus, while the question “Would they do it again?” is quite speculative, post-1964 legislative developments surrounding private civil rights enforcement in general, and Title VII lawsuits in particular, do not suggest regretful longing by Republicans for an NLRB for civil rights.
Chapter 7 CONCLUSION AND IMPLICATIONS
State Capacity and Private Enforcement Regimes Private enforcement regimes are a critically important instrument that American state actors use to achieve their objectives in the process of policy implementation. Using economic incentives to mobilize private litigants and the private bar, members of Congress have used this instrument, which they recognize partly as an alternative to administrative power, with a high degree of self-consciousness since the founding of the American regulatory state in the late nineteenth century. With it they have constructed a massive private enforcement infrastructure and penetrated and controlled spheres of the economy and society spanning the waterfront of federal civil regulation. With it they have constructed the litigation state. State-centered social scientific and historical research has been largely confined within a narrow, bureaucracy-centered conception of what the state is, treating state capacity as a function of such variables as the number of bureaucratic personnel, the degree of their organizational centralization, and capabilities of resources extraction. This literature has emphasized the American state’s weakness, portraying it as laggardly, amateur, and anemic. The implicit comparative benchmark of strength for these judgments is generally a centralized bureaucratic European model of the state. While these scholars have often recognized that the absence of this form of strength in the American state is rooted in its institutional fragmentation, particularly its separation of powers structures, their gaze has remained too fixed on the ways in which institutional fragmentation has thwarted, truncated, and deformed efforts to build a state on the European model. They have been inattentive to the ways in which some of those very same institutional properties are also generative of a different form of state-building. As distinguished from the centralized bureaucratic European model of state strength, a great deal of American regulatory state control has taken the form of radically decentralized intervention by an army of litigants and lawyers licensed by the state and paid bounty by defendants at the state’s command. Because of the distinct structure of American political institutions, America’s regulatory state has taken a distinct form—one importantly dependent upon private litigation.
CONCLUSION
215
The sharp disjuncture between assessments of Title VII’s enforcement framework by state-centered scholars on the one hand, and those of civil rights activists on the other, highlight why an excessively bureaucracycentered conception of state capacity is so limiting. Scholarship examining the move from administrative power to private litigation in the compromise on Title VII’s enforcement framework often features an elegiac trope. When examining the movement from an administrative ceaseand-desist enforcement model to one founded on private lawsuits, this literature is characterized by narratives of evisceration, privation, and enfeeblement. It imagines the administrative enforcement power that Republicans took away in 1963–64 as substantial and muscular—a powerful state apparatus that might have intervened forcefully in the American workplace to move the nation toward greater equality. It treats the private enforcement powers that were substituted as meager, ineffectual, or scarcely worthy of mention, showing little or no regard for the powerful and effective army of private litigants and lawyers that the enforcement compromise ultimately mobilized.1 In contrast with most state-centered scholars’ view, within a few years of Title VII’s passage, civil rights activists lost confidence in executive implementation and grew committed to the efficacy of private litigation. In the late 1960s, they shifted to advocating a hybrid private litigation / cease-and-desist framework after observing a budget-strapped and overwhelmed agency, and an executive branch they regarded as lacking the political will, and at times the desire, to enforce the law aggressively. They thus refused to give up private litigation for a purely administrative framework. Alongside civil rights groups’ waning belief in administrative enforcement, they became increasingly confident in the efficacy of private litigation, with fee shifting, based upon both gratifying substantive results in courts and the high levels of private enforcement that ensued once the for-profit civil rights bar began to grow exponentially in the first half of the 1970s. When the issue of EEOC cease-and-desist power resurfaced in debates leading to passage of the CRA of 1991, civil rights groups, facing what they regarded as the most ideologically hostile EEOC leadership in the agency’s history, flatly rejected the prospect of vesting it with cease-and-desist authority, under any circumstances, hybrid or otherwise. Looking back to the compromise of 1964, civil rights groups—in stark contrast with many civil rights scholars—recognized in the Republicanorchestrated substitution of private lawsuits for administrative power a potent strengthening of the American state’s capacity to address job discrimination. The policy feedback effects of the 1964 act had a profound impact on policy activists. Ironically, during the presidency of George W. Bush, some union advocates began to complain that the National Labor
216
CHAPTER 7
Relations Act’s protections of worker rights were “notoriously underenforced” by the National Labor Relations Board, liberals’ onetime exemplar of effective administrative power. They began to argue for amending the act to provide workers a private right of action for some of the law’s core guarantees, citing Title VII as their model.2
American Political Institutions and Private Enforcement Regimes Legislative-Executive Conflict Interbranch conflict between Congress and the president over control of the administrative state drives Congress’s reliance upon private lawsuits for implementation of its statutory commands. Chapter 3 presented strong statistical evidence that Congress is significantly more likely to enact private enforcement regimes as the ideological preferences of Congress and the president diverge, a finding that proved robust across multiple measures of interbranch conflict. Chapters 4 to 6, using qualitative historical evidence, traced the causal processes at play in generating an enforcement framework dominated by private lawsuits for federal civil rights laws. This process-tracing evidence also reflected, repeatedly over time, the importance of legislative-executive competition for control of the administrative state in motivating Congress’s reliance upon private lawsuits. This was true in 1964, when pivotal Republicans feared an overzealous agency, and in both the Nixon-Ford and Reagan-Bush years, when Democrats complained of an underzealous one. In all three chapters, across multiple partisan configurations of legislative-executive conflict, distrust of the executive was critical in driving congressional enactment of private enforcement regimes. The statistical evidence of the correlation between legislative-executive ideological conflict and congressional enactment of private enforcement regimes, and the historical evidence showing that legislative-executive conflict was an important part of the causal processes materially driving the congressional choice of private enforcement regimes, are mutually reinforcing. Joined with past research on congressional delegation of power to the bureaucracy, these findings regarding legislative-executive conflict have important implications for the allocation of power among branches within the American separation of powers system, and between state and civil society. It has been well established in the bureaucratic delegation literature that legislative-executive conflict, or divided government, leads Congress to delegate less power to the bureaucracy.3 The evidence presented here shows that under precisely these conditions, in addition to seeking to constrain and limit bureaucratic power, Congress relies
CONCLUSION
217
more upon private litigants, lawyers, and courts to implement policy. The implications are that increasing degrees of ideological conflict between Congress and the executive will not only diminish executive power, but will correspondingly enhance the power of private litigants, lawyers, and courts in the enforcement of statutes and the elaboration of their substantive meaning. The case of civil rights presents a clear example of how private enforcement regimes can diminish executive power, and the Reagan years provide a particularly stark illustration. The legislative process that produced the enforcement framework for Title VII of the CRA of 1964 explicitly traded executive power for private lawsuits. Later, Reagan sought to reduce enforcement in the area of civil rights, including through the EEOC—and this is corroborated by numerous objective measures of administrative enforcement. Reagan also sought, unsuccessfully, reforms to limit attorney’s fee awards with the goal of reducing private civil rights litigation. Despite these efforts at executive retrenchment, and partly because of them, private civil rights actions surged during Reagan’s first term, ultimately increasing the aggregate enforcement levels faced by employers. EEOC enforcement efforts were already only a drop in the bucket of total enforcement when Reagan came to power, and they became an even smaller drop as the number of private actions grew. Later, during legislative debates leading to the CRA of 1991, Republicans complained bitterly that enhancing Title VII’s private enforcement regime would further weaken executive power to administer the law. When the CRA of 1991’s new litigation incentives mobilized even more private enforcement, Republicans well knew that the executive apparatus would recede yet further into the background of the implementation of job discrimination laws. The following two cases also demonstrate how private enforcement regimes, arising from legislative-executive conflict, can reduce executive power. Other Cases: Labor Unions and Clean Air
To reinforce the evidence in support of the legislative-executive conflict hypothesis, I briefly sketch two other cases, one in the area of labor policy (the Taft-Hartley Act of 1947), and the other in environmental policy (amendments to the Clean Air Act passed in 1970 and 1990). The purpose here is not to undertake a deep investigation of all of the causes of the legislative outcomes observed. Rather, it is to demonstrate the wide range of policy areas, the diversity of historical periods, and the variation in partisan interbranch configurations in the which the causal process has operated. I discuss the evidence in these cases linking legislative-executive conflict to the congressional choice of private enforcement in more detail elsewhere.4
218
CHAPTER 7
When Republicans took control of Congress in 1947, after fourteen years in the minority, labor law reform was at the top of their agenda. The Taft-Hartley Act of 1947, which amended the National Labor Relations Act of 1935, was manifestly aimed at scaling back the power of unions, which had grown assertive under the protective shelter of the NLRA. The law was passed by a “conservative coalition” of Republicans and southern Democrats, and among its most controversial provisions was the illegalization of certain forms of strikes, picketing, and boycotts, to be enforced via private lawsuits by businesses against unions for economic damages. This reliance upon private lawsuits for enforcement—among the most heatedly debated issues in the legislative proceedings—was an anomalous choice on its face. It departed radically from the already existing NLRA enforcement framework, which was robustly administrative, with the National Labor Relations Board empowered to adjudicate disputes under the act and issue binding cease-and-desist orders.5 Why did the conservative coalition choose private lawsuits to implement the new bans on union actions, instead of using the already existing administrative machinery created in the NLRA for the very purpose of implementing the act? Quite emphatically, the coalition did not trust the NLRB, which it regarded as ideologically biased in favor of labor unions and against business. This was nothing new. The NLRB under Roosevelt and Truman had pursued labor policies detested by the conservative coalition, and viewed by it as rooted in executive ideology rather than legislative intent, producing antagonism toward the agency among conservative legislators starting in the late 1930s.6 The House report on the bill leading to Taft-Hartley stated that investigation of the NLRB had “shown bias and prejudice to be rampant” in its enforcement practices,7 and the Senate report stated that “the Board has acquired a reputation for partisanship.”8 The two leading advocates for private lawsuits in the Senate were sponsor Robert Taft (R-OH) and Joseph Ball (R-MN). According to Taft, the NLRB had been “made up of . . . people who regarded themselves as crusaders to put a CIO union . . . in every plant in the United States,” and who perpetrated “outrages” and “injustices” against employers.9 According to Ball, the NLRB had been “rabid,” had “harassed employers,” and had “perverted the act and used it as an instrument of the CIO.”10 The conservative coalition believed that the use of private lawsuits prosecuted by injured businesses would be a far more effective enforcement tool than the prounion NLRB for ending the banned labor actions.11 Legislators hostile toward the NLRB had proposed bills starting in the late 1930s, and examined in high-profile and acrimonious congressional hearings, seeking to amend the NLRA so as to displace implementation authority from the NLRB to litigation and courts.12 Now in the majority, in Taft-
CONCLUSION
219
Hartley they mobilized private litigants to do what, in their view, the NLRB would not. Similar institutional dynamics drove legislative reliance upon private lawsuits to implement federal clean air legislation in 1970 and 1990, though this time with Democrats in control of Congress and Republicans in control of the White House. The Clean Air Act of 1970, which amended the Clean Air Act of 1963, contained the first “citizen suit” provision in federal environmental law. Though private rights of action were well known in federal regulatory statutes, they ordinarily were conferred upon people who suffered a tangible economic injury that involved violation of a statute, and allowed suit only against the entity causing the injury. In contrast, the 1970 act’s citizen suit provision conferred broad authority upon “any person” to file suit against violators of the law, and against the EPA for failure to carry out its mandatorily prescribed responsibilities under the act, with attorney’s fee awards for winning plaintiffs. This citizen suit provision thus sought to mobilize private litigation against the targets of regulation (the focus of this book), as well as using it as an instrument to control the bureaucracy. The Clean Air Act of 1990 bolstered the 1970 citizen suit provision, making new monetary penalties available, expanding the conduct covered by private suits, relaxing the evidentiary standards plaintiffs had to meet, and making violations easier to prove.13 What motivated the citizen suit innovation in 1970, and its expansion in 1990? Legislative-executive conflict was an important factor. When drafting the 1970 act, a Democratically controlled Congress doubted that the Nixon administration would enforce its statutory mandates vigorously, and it included the citizen suit provision with attorney’s fee awards to ensure adequate enforcement.14 The fact that the citizen suit provision specifically authorized suits to compel action by the EPA makes abundantly clear that Congress was concerned not only about supplementing the agency’s resources with private lawsuits, but also about its deviation from congressional mandates. Buttressing this view, the Democrat-led committee reports on the act stated that “[g]overnment initiative seeking enforcement of the Clean Air Act has been restrained,” and has “fail[ed] . . . to demonstrate sufficient aggressiveness”; that the citizen suit provision was intended to “motivate governmental agencies” to enforce the act; and that the provision would ensure “vigorous enforcement action” even if “agencies should fail in their responsibility” to enforce.15 In 1971, the Senate Committee Report on the Clean Water Act, passed over Nixon’s veto, emphasized that a similar provision could remedy “an almost total lack of enforcement” by the EPA.16 As in the case of civil rights policy, legislative-executive conflict over environmental policy intensified in the 1980s. According to a great many
220
CHAPTER 7
observers, Reagan undertook to withdraw the administrative enforcement machinery with the purpose of discouraging vigorous enforcement of environmental laws.17 Congressional Democrats took notice. Oversight hearings and investigations of the EPA by Democrat-led congressional committees were frequent in number and acrimonious in tone throughout the 1980s, and committee Democrats persistently leveled the same charge: the agency was willfully failing to implement Congress’s statutory mandates.18 When the Clean Air Act amendments of 1990 were debated, Democratic drafters and supporters of a stronger citizen suit provision expressly linked the need for more private enforcement to willful executive nonenforcement. Senate majority leader George Mitchell (D-ME) characterized EPA inaction as “nothing short of astonishing,” and concluded that “without such [citizen suit] litigation, key provisions of the act may remain unimplemented.”19 Representative Cardiss Collins (D-IL) declared that the EPA was “blatantly violating” congressional directives under the act,20 and explained, regarding the citizen suit amendments which she helped draft, that “[t]he objective is to ensure the rights of the clean air laws, even when the authorities have chosen not to pursue violations.”21 Representative Henry Waxman (D-CA), who also had a hand in crafting the citizen suit amendments, characterized the Reagan administration’s posture toward the Clean Air Act in the 1980s as a “fundamental assault,”22 and introduced a document into the record stating that in the Reagan years “the only effective enforcement of federal environmental laws was citizen involvement through lawsuits.”23 Unmistakably, legislative-executive conflict over control of the EPA was an important factor driving Congress to ratchet up private enforcement. In both 1970 and 1990, Congress sought to mobilize private litigants to do what, in its view, the EPA would not. Legislative-executive conflict as a source of private enforcement regimes cuts across many policy areas and is not unique to civil rights. Rival Hypotheses
The introduction to part II of this book observed that qualitative historical evidence could add significantly to the statistical findings on legislativeexecutive conflict if the two forms of evidence were mutually reinforcing. While this is true in general, it is especially significant to the causal status of legislative-executive conflict because some observers might claim that, notwithstanding the statistical findings, the broad increases in both legislative-executive conflict and legislative enactment of private enforcement regimes starting in the late 1960s are actually not causally linked. From such skeptics’ perspective, rival explanations for the simultaneous
CONCLUSION
221
increases might be that (1) it was simply a coincidence that policy activists learned about the utility of private enforcement regimes, leading to their increased use, at the same time that divided government became much more common; (2) increases in both legislative-executive conflict and enactment of private enforcement regimes arose from some other durable cause that emerged in the late 1960s (perhaps in political culture), while the two effects of that cause are not themselves causally linked; or (3) increased enactment of private enforcement regimes did occur in the late 1960s because of legislative-executive conflict, but then it became a routine, unreflective, boilerplate approach to legislating that reproduced itself in a manner unaffected by that conflict. The qualitative historical evidence provides a powerful answer to these alternative hypotheses to the extent that they are offered to deny the operation of legislative-executive conflict as a mechanism causing enactment of private enforcement regimes. The chapters on civil rights legislation, along with the labor and environmental examples just given, show that legislative-executive conflict motivated legislators’ self-conscious choice of private enforcement regimes, as an alternative to administrative power, in multiple domains of policy in repeated episodes covering about a half century. The civil rights chapters show that while civil rights groups’ preferences most certainly were transformed by learning (and other) feedback processes, these processes were entwined with legislativeexecutive conflict (in the Nixon-Ford years). As civil rights advocates learned, they were part of a Democratic legislative coalition facing a Republican White House which they regarded as committed to nonenforcement of civil rights laws. Subsequently, their desire for and ability to obtain private enforcement regimes were still critically shaped by institutional conflict between their mainly Democratic allies in Congress, and their Republican opponents in the White House (in the Reagan-Bush years). The historical evidence also shows that private enforcement regimes never became routine and uncontested, reproduced by Congress blindly without respect to strategic considerations of implementation. To the contrary, in the many legislative episodes between 1963 and 1991 examined, they were consistently the subject of conflict and contestation. Mobilization of private civil rights enforcers was the subject of close scrutiny when the foundational CRA of 1964 was fashioned, and had grown more (not less) controversial by the time the CRA of 1991 became law. Historical Patterns of Divided Government and Party Polarization
To appreciate the importance of the evidence linking legislative-executive conflict to private litigation in American policy, it must considered in the context of long-run historical patterns of divided government and
222
CHAPTER 7
party polarization. While private enforcement regimes have been used in federal regulation since at least the late nineteenth century, their significance has increased powerfully since the late 1960s, corresponding to the pervasiveness of divided government since the Nixon administration. In the first sixty-eight years of the twentieth century (1901 to 1968), one political party had control of both the executive and legislative branches 79 percent of the time. In the subsequent thirty-two years of the twentieth century, the parties divided control of those branches 81 percent of the time. In the mid-1960s, the estimated probability of divided government became more likely than not for the first time in the twentieth century, according to Charles Cameron, and it climbed steadily to a likelihood of roughly 80 percent by the close of the century. Moreover, during the years of divided government between Nixon taking office and the end of the twentieth century, 77 percent of the time the party more likely to undertake regulatory intervention (the Democrats) controlled one or both chambers of Congress, and the party more likely to oppose regulatory intervention (the Republicans) controlled the executive branch.24 The durability of the condition of divided government, primarily with Democratic legislators facing Republican presidents, that emerged in the late 1960s was exacerbated by another factor contributing to legislativeexecutive antagonism: party polarization. The notion that partisan division between legislative and executive branches will drive Congress to structure laws so as to circumvent executive interference with implementation is predicated upon the expectation of disagreements between the parties over policy. This is not static. In the Eisenhower years, for example, strong centrist elements in both parties allowed for a relatively cooperative lawmaking relationship between the executive and legislative branches despite divided government.25 However, ideological polarization between the parties starting between the late 1960s and the mid-1970s, depending upon one’s measure, eroded the bipartisan center in Congress and fueled party antagonisms. In the late 1960s, with Nixon’s first term, legislators not from the president’s party began to vote at markedly higher rates—as compared to members of the president’s party—against presidential initiatives, a trend that continued to slope upward through the end of the twentieth century.26 In the mid-1970s, ideological polarization between Democrats and Republicans, as measured by the distance between mean party NOMINATE scores (discussed in chapter 3) began to increase for the first time in the twentieth century, and the distance between the parties continued to grow through century’s end.27 The conjuncture of divided government and party polarization, primarily with Democratic legislators facing Republican presidents, over roughly the past four decades has been an important cause of the explosion of federal statutory litigation beginning at the end of the 1960s. In
CONCLUSION
223
an era of divided government and party polarization, Congress’s wariness of presidential commitment to construing and enforcing regulatory laws in accord with its wishes led it to turn more frequently to private enforcement regimes. In this era, private enforcement regimes have proliferated in the United States Code, and private statutory litigation has mounted apace. While private enforcement of federal statutes had been an important part of the American regulatory state since at least the Progressive Era, it underwent a seismic upward shift at the end of the 1960s, transforming the balance between public and private in federal policy implementation, and increasing the role of litigants, lawyers, and courts in fashioning and enforcing federal policy. Judicial Ideology The role of judicial preferences in the separation of powers dynamics producing private enforcement regimes is more ambiguous. We observed in chapter 2 that, from the standpoint of theory, it is not possible to generate a straightforward prediction about the relationship between judicial ideology and Congress’s enactment of private enforcement regimes. We observed, rather, that there were plausible hypotheses pointing in opposite directions. Judicial friendliness to congressional preferences would certainly make the judiciary a more attractive enforcement venue, other things being equal, thus encouraging Congress to enact private enforcement regimes. However, judicial hostility to legislative preferences can be counteracted, with respect to its effects on litigant mobilization, by increasing the magnitude of litigant-mobilizing incentives in a private enforcement regime. The statistical model in chapter 3 produced null findings for judicial ideology as an explanatory factor. While these null finding may be masking the hypothesized contradictory effects of judicial ideology, they are also consistent with judicial ideology simply being inconsequential as either an incentive or disincentive for enactment. The historical process-tracing evidence in chapters 4 to 6, however, suggests that judicial ideology is, in fact, relevant to legislators’ and interest groups’ assessment of the desired balance between administrative and judicial power in regulatory implementation. It also suggests that judicial ideology can indeed push in opposite directions at different moments in time for precisely the reasons hypothesized above. Republican support for private Title VII enforcement in 1963–64 was based partly on their belief that federal courts would be a more ideologically friendly implementation venue than a Kennedy or Johnson EEOC. However, contrary to conservatives’ and liberals’ expectations, federal courts proved an extraordinarily hospitable venue for Title VII plaintiffs in the late 1960s and early 1970s. This judicial friendliness, from liberals’ point of view,
224
CHAPTER 7
contributed materially to their rejection of the prospect of trading private lawsuits for enhanced administrative state power. It also encouraged their ultimately successful campaign to spread plaintiffs’ fee shifting across the field of civil rights. On the other hand, from the standpoint of civil rights liberals, by the end of the 1980s, when the new conservative majority on the Supreme Court handed down the decisions that triggered the movement for what became the CRA of 1991, the federal judiciary had shifted from ally to foe. Civil rights liberals did not turn away from courts, however. The executive branch hardly looked to them like a better implementation option. Instead, they ratcheted up Title VII’s private enforcement regime, unleashing more private litigants and lawyers than ever before to enforce the statute through courts. Thus, judicial ideology factored significantly into the forces that drove the creation of a Title VII enforcement framework dominated by private litigation, but it did so in contradictory ways over time. The fact that both judicial friendliness and judicial hostility encouraged enactment of private enforcement regimes maps clearly to the contradictory hypotheses about the influence of judicial ideology. Policy Drift The coalition drift hypothesis holds that current legislative majorities use private enforcement regimes, in the institutional environment of a sticky status quo, to insulate enforcement of their regulatory policies from subversion by future legislative coalitions. It was the only hypothesis to be robustly supported in the empirical models in chapter 3—which found a strong association between impending electoral losses and enactment of private enforcement regimes—while at the same time being wholly unsupported in the process-tracing evidence in part II of the book. I offer some conjecture about why this is so. Just as we have considered the possibility that the public legislative historical record may be an unlikely place to find rent-seeking lawyers advocating policies with the goal of self-enrichment, the public record may likewise be an unlikely place to find support for the coalition drift hypothesis. The core idea of coalition drift as a motivation for policy design, according to Terry Moe, is to “insulate from democratic control” and to “protect . . . from changes in group power and public authority.”28 Focusing on statutory constraints upon bureaucracy, Moe observes that “[t]oday’s enacting coalition . . . wants to ensure that tomorrow’s legislature cannot control the bureaucracy.”29 Applying the coalition drift hypothesis to private enforcement regimes, the logic is to insulate a regulatory enforcement mechanism from democratic control by tomorrow’s legislature. This strategy of policy implementation is susceptible to being regarded
CONCLUSION
225
as fundamentally antidemocratic since, by design, it endeavors to weaken the polity’s capacity for self-government in future periods. As such, it is plausible that legislators and interest groups, when motivated by fear of coalition drift, would not publicly articulate it as a justification for enactment of private enforcement regimes. The bureaucratic drift hypothesis holds that enacting coalitions use private enforcement regimes, in the institutional environment of a sticky status quo, to insulate from subversion by bureaucrats pursuing their own interests, including ideological agendas, timidity, apathy, shirking, careerism, and capture. While the bureaucratic drift hypothesis was not susceptible to straightforward testing in the statistical model, the model did find a positive correlation between the presence of issue groups in congressional hearings and enactment of private enforcement regimes, and scholars have argued that issue groups in the middle to late 1960s began to advocate private enforcement regimes as an alternative to dependence on administrative power because of concern about bureaucratic drift. This claim about issue group preferences and motivations was supported by the process-tracing evidence. Even before the transformation in civil rights group preferences took place, Republicans’ support for Title VII’s private enforcement regime in 1963–64 was partly motivated by their fear that the single-mission job discrimination agency sought by civil rights liberals would have a structural bias in favor of civil rights plaintiffs and against business. As a function of its structure, the proposed agency threatened to drift to the left, and this concern encouraged Republican support for the private enforcement alternative. Starting in the late 1960s and into the mid-1970s, civil rights groups’ support for private enforcement regimes was partly motivated by apprehensions about bureaucratic drift, particularly toward timidity, apathy, and capture. In the legislative struggles leading to the Civil Rights Act of 1991, the same sorts of concerns were again expressed by Democratic legislators and civil rights groups in support of increasing the incentives for private enforcement litigation. Thus, over the course of Title VII’s history, both conservatives and liberals have deployed private enforcement regimes as a guard against bureaucratic drift. Compromise The veto points and compromise hypothesis maintains that the many veto points in the American lawmaking process can encourage private enforcement regimes as a compromise alternative to bureaucratic statebuilding. Private litigation, with its air of private dispute resolution, is less visible and more ambiguous as a form of state intervention than bureaucratic state-building, and thus it may be the preferred alternative for
226
CHAPTER 7
legislators with antistatist preferences, a significant strand of the American political tradition, particularly as applied to the central state in the United States’ federalist system. This hypothesis was not susceptible to straightforward testing in the empirical model. It was, however, strongly supported by the process-tracing evidence. While the inferences that can be drawn from the study of one policy area are necessarily limited, it bears noting that compromise dynamics similar to those described below have been observed in the area of consumer protection legislation.30 In 1963–64, among conservative Republicans’ primary arguments against the strong EEOC sought by civil rights liberals was that it would entail excessive growth in the federal bureaucracy, and their control of the filibuster pivot in the Senate empowered them to refashion the law to better suit their preferences, which they did by substituting private litigation for bureaucratic power. In the campaign for EEOC cease-anddesist authority between 1965 and 1972, conservative Republicans again asserted arguments against growing federal bureaucracy, and fragmentation in the lawmaking process again contributed to the defeat of ceaseand-desist in favor of litigation: cease-and-desist was repeatedly quashed in committee; it twice passed one chamber but not the other; and in the bill that became the Equal Employment Opportunity Act of 1972, it was removed in a bargain to end a filibuster. The fact that private enforcement regimes privatize enforcement costs also contributed to their effectiveness as a compromise solution. Though the empirical model shows that congressional enactment of private enforcement regimes does not vary with budget deficits or surpluses over the course of the twentieth century, and budgetary considerations did not appear to materially affect the CRAs of 1964 or 1991, budgetary considerations did enter the picture during the period of entrenchment and expansion of private civil rights enforcement in the first half of the 1970s. Shep Melnick has observed that budget deficits emerged as a politically salient issue exactly during this period, enhancing the power of conservatives to oppose initiatives that would require spending, and increasing incentives for all legislators to shift regulatory implementation costs to the private sector, such as by relying upon lawsuits.31 Though courts ruled that they would not appoint counsel under Title VII’s appointmentof-counsel provision unless Congress appropriated funds to pay them, it did not do so. The budgetary issue was also clearly at play in the 1971 debates over how to provide for enforcement of federal school desegregation mandates. Republicans were not willing to appropriate money to pay fee awards for such lawsuits, but they were willing to support a feeshifting provision that would impose the fees on losing school districts. In debates over the Civil Rights Fees Act of 1976, budgetary efficiency emerged as a key justification on which Democrats and Republicans could agree, and it was explicitly braided with an antibureaucracy theme:
CONCLUSION
227
fee shifting does not add to the federal budget, and it does not grow the federal bureaucracy, the argument went. Chapter 5 provided considerable evidence that many Democrats and their civil rights group constituents probably did not believe that enforcement money would have been well spent on funding the Nixon-Ford bureaucracy, which they accused of willful nonenforcement of civil rights laws. For them, the twin tropes of fiscal conservatism and bureaucratic modesty were an effective way to forge an extremely broad consensus behind enlarging the private civil rights enforcement infrastructure, which they had already come to regard as more reliable and effective than bureaucracy.
American Exceptionalism and Private Enforcement Regimes The findings linking American political institutions to some kinds of litigation rates suggest an important qualification of accounts of American “litigiousness” grounded in the exceptional character of American culture. “[A]ll nations are exceptional,” remarked Sidney Verba, “but the United States is more exceptional than most.”32 Americans, according to this perspective, are a people apart, for better or worse. One particularly prominent strand of this notion of American exceptionalism understands the extensive role of legal rights, litigation, and courts in American policy as arising from an exceptionally individualist American political culture.33 Seymour Martin Lipset, in the fullest exploration of the cultural roots of American exceptionalism, observes that the United States is “a society profoundly rooted in law,” with a “potent orientation toward individual rights,” fostering “the American eagerness for legal settlements to disputes . . . [and] excessive litigiousness.”34 The American cultural soil, on this view, has proven fertile for the growth of a public mind-set of “rights talk” and “rights consciousness” that fosters a polity characterized by a strong proclivity toward rights-asserting behavior.35 That the United States is distinctive in the degree to which it utilizes the assertion of rights through litigation in policy implementation, and more broadly in the extent to which its citizens litigate to resolve disputes, is supported by cross-national empirical research.36 That American culture is causing the litigation, however, is not entirely clear. Cultural explanations for large political outcomes, such as the structure and magnitude of the welfare state, or the extent of litigation in policy implementation in a country, are quite difficult to demonstrate with evidence.37 While I do not discount the causal significance of culture on this ground, cultural accounts provide, at best, a radically incomplete explanation for the massive role of private litigation in federal policy implementation. Its distinct pattern of growth—exploding in the late 1960s, when other important forms of litigation in the very same courts remained flat—and its ex-
228
CHAPTER 7
tremely close mapping to legislative enactment of economic incentives for it, suggest that there is more to the story than a peculiar American propensity to sue. Further, the battle over Title VII in 1963–64 highlights a conceptual problem in unreflectively associating a culture of rights with a culture of lawsuits. Rights activists in 1963–64—historic avatars of American “rights consciousness”—believed that administrative machinery would be better than lawsuits to realize the rights they cared about. Their subsequent embrace of lawsuits was a pragmatic choice shaped by political realities unfolding on a concrete institutional terrain, not an expression of cultural preference. The institutional arguments advanced in this book, emphasizing links between American state structures and legislative creation of private enforcement regimes, have in common with cultural exceptionalist accounts an emphasis on distinctive features of the United States in cross-national context. This institutional cross-national point is importantly connected to debates about the unique character of the American administrative state. As discussed in chapter 2, James Q. Wilson has complained that American bureaucracy is distinctive, as compared to parliamentary regimes, in the extent to which its administrators are burdened by formal rules and constraints, obstructing their ability to execute their missions effectively, and weakening the administrative state.38 In response, Terry Moe argues that these distinctive attributes of American bureaucracy arise from the institutional structure of the separation of powers system: Congress cabins bureaucratic discretion in order to guard against subversion by presidents and future legislative majorities, a strategy that does not work or make sense in parliamentary regimes, where the executive arises from the legislature, the majority party controls both, and the status quo is easy for majorities to move.39 The arguments and evidence in this book link institutional distinctions between separation of powers and parliamentary systems to congressional reliance on private lawsuits for implementation. From a comparative cross-national point of view, “weak” American administrative state capacity on the one hand, and extensive private litigation in American policy implementation on the other, are linked outcomes of the same institutional causes and processes.
Political Parties, Interest Groups, and Private Enforcement Regimes Democrats and Republicans The evidence in this book suggests that the role of political parties in constructing the litigation state is more complex than that envisioned by the standard account of Democrats, acting as the party of “trial lawyers,”
CONCLUSION
229
being monopoly producers of legislation fostering private litigation. This partisan alignment perspective was discussed in chapter 3, which distinguished between a strong version of the party alignment hypothesis, according to which other causes of congressional enactment of private enforcement regimes would be conditional upon Democratic control of Congress, and a weak version, according to which Democratically controlled Congresses would simply be more likely to enact them than Republican ones. The statistical evidence provided no support for the strong version of the party alignment hypotheses, and the weak version was supported in two of the three models presented. Democratic control of Congress is associated with greater enactment of private enforcement regimes, with the magnitude of the effect being a meaningful but far from dominant factor, and clearly smaller than the effects of the significant institutional variables. The historical process-tracing evidence focusing on civil rights enforcement also problematizes the traditional party alignment account. We observed that it was Republicans who were the pivotal actors in making private lawsuits central to the CRA of 1964’s job discrimination provisions and the CRA of 1968’s fair housing provisions, in both instances as a means of defeating bureaucratic state-building. A Republican introduced the plaintiffs’ fee shift for school desegregation cases in the Emergency School Aid Act of 1972, after Republicans had succeeded in defeating a public fund to pay private lawyers to prosecute such suits. A Republican introduced the bill that became the CRA of 1991, a compromise measure aimed at increasing litigation incentives, but by a lesser degree than civil rights groups and liberal Democrats sought. To be sure, by the start of the 1970s it was liberal Democrats and their civil rights group constituents that were the prime movers behind getting the issue on the congressional agenda. But when they did push for major expansions of private civil rights enforcement leading to the Civil Rights Fees Act of 1976 and the CRA of 1991, both laws ultimately passed with overwhelming Republican support. Finally, we have also observed examples of Republican Congresses relying upon private enforcement regimes with the goal of assisting traditional Republican constituencies, such as in favor of business interests in the context of conflict with unions (the Taft-Hartley Act of 1947), for Cubans divested of property by the Castro regime (the Cuban Liberty and Democratic Solidarity Act of 1996), and against abortion providers (“Partial-Birth Abortion” Ban Act of 2003). These various episodes of Republican support for private enforcement regimes reveal dimensions of the partisan dynamics behind the construction of the litigation state that typically are absent from conventional party alignment accounts. On the whole, the evidence presented in this book suggests that differences
230
CHAPTER 7
between the parties’ level of support for private enforcement regimes, within the domain of federal statutory regulation, are differences in degree and not kind, with Democrats more likely to enact them. Issue Groups Interest groups are important sources of demand for private enforcement regimes, and they are an important part of party coalitions. The empirical model in chapter 3 revealed that the presence of issue advocacy groups in congressional hearings on regulatory legislation, where they are invited by legislators to testify in favor of their preferred outcome, is correlated with enactment of private enforcement regimes. The process-tracing evidence from the case of civil rights dovetailed with these statistical findings. Although civil rights groups initially opposed reliance on private lawsuits for Title VII enforcement, when this approach was imposed by Republicans, civil rights groups insisted on the availability of fee recovery for plaintiffs with the goal of addressing well-recognized economic obstacles to private civil rights enforcement. They subsequently were active in shaping the content of the Equal Employment Opportunity Act of 1972 so as to preserve Title VII’s private enforcement regime; they successfully advocated extending fee awards to school desegregation cases in 1972, and to voting rights cases in 1975; they were the catalysts behind getting the Civil Rights Fees Act of 1976 passed; and they put the CRA of 1991 on the agenda and wrote the first bill that was introduced on the way to its passage. We have seen that the causal role of issue groups cannot be neatly separated from the causal role of the separation of powers institutions in which they operate. Issue groups were critically motivated in their support for private enforcement by fears that the preferences of the enacting coalition, of which they were a part, would be subverted by hostile presidents and bureaucratic drift. These findings regarding issue advocacy groups also shed some light on the meaning of the evidence indicating that Democratically controlled Congresses are more likely than Republican ones to proactively mobilize private litigants for policy implementation. As discussed in chapter 3, the issue-oriented advocacy organizations that proliferated starting in the middle to late 1960s, including but not limited to legal advocacy organizations, focusing on such issues as civil rights, environmental, and consumer protection policy, both favored court-based implementation and emerged as a core constituency of the Democratic party coalition. To the extent that these groups have influenced legislation passed by Democrats, as we saw that they did repeatedly in the case of civil rights, it suggests that the Democrats’ greater propensity to enact pri-
CONCLUSION
231
vate enforcement regimes is being driven, at least in part, by this core constituency. The emergence of this legislative coalition adds to the configuration of elements that aligned starting around the late 1960s to produce the sea change toward greater congressional reliance on private enforcement regimes. Elements already identified are divided government, primarily with Democratic legislators fearing Republican presidents, and party polarization. Issue-oriented advocacy groups assumed a more important place within the Democratic party’s base, and their preference for courtbased implementation surged. Rent-Seeking Lawyers Claims that lawyers’ associations, acting in pursuit of self-enrichment, are important and effective sources of demand for enactment of private enforcement regimes are not well supported by the evidence presented in this book. The empirical model in chapter 3 revealed that the presence of representatives of lawyers’ associations in congressional hearings on regulatory legislation, where they are invited by legislators to testify in favor of their preferred outcome, is not associated with enactment of private enforcement regimes. The process-tracing evidence indicates that they simply were not a significant source of demand in the massive interest group lobbying campaign that led to the foundational CRA of 1964. Pivotal decisions to insert private lawsuits in Title VII came from conservative Republicans aiming to avoid bureaucratic regulation of business, particularly by a Democratically controlled agency, not to enrich lawyers in general, or plaintiffs lawyers in particular. The plaintiffs bar is not typically regarded as a constituency served by antiregulation Republicans with close ties to the business establishment, such as Everett Dirksen. The attorney’s fee provision in Title VII, which would come to serve the economic interests of the for-profit civil rights bar, was insisted upon by civil rights groups, which believed that it would be necessary in order to have any chance of mobilizing meaningful levels of private enforcement. The enforcement provisions of the CRA of 1964, and other enforcement regimes modeled on it, effectively created a private civil rights bar that had not previously existed. The for-profit civil rights bar thereby conjured into being assuredly did subsequently mobilize to entrench and expand opportunities for fee-generating civil rights litigation. However, it is abundantly clear from the historical record that it was civil rights groups that were the catalysts for and architects of the successful campaign for entrenchment and expansion of private lawsuits for civil rights
232
CHAPTER 7
implementation. It is also clear that their motivations were grounded in ardent civil rights advocacy. For-profit lawyers benefited, to be sure, but they were not the agenda setters for this legislation, nor was the legislation intended to benefit them. Rather, the for-profit civil rights bar was seen by civil rights groups as a resource to be leveraged, through the economic value of lawsuits, to create a powerful private enforcement infrastructure in the service of their policy agenda. To the extent that demand for legislation by rent-seeking lawyers contributed to reliance upon private lawsuits for civil rights implementation, it did so by bolstering the efficacy of civil rights groups in the legislative process. A very important qualification is in order about how the evidence in this book speaks to the relationship between parties and interest groups in influencing the extent of private litigation in policy implementation. The contention that Democrats are the allies and Republicans the enemies of the use of private litigation has been most frequently made in reference to “trial lawyers,” particularly as represented by the Association of Trial Lawyers of America (ATLA), a significant source of campaign contributions to the Democratic party.40 ATLA is an association of plaintiffs’ tort lawyers, or lawyers who represent plaintiffs seeking money damages for injuries caused by breaches of the common law of tort, ranging from individual slip-and-fall, automobile accident, and medical malpractice personal injury cases, to high-profile product liability class actions cases against pharmaceutical, tobacco, and asbestos manufacturers. Apparently owing to ATLA’s name, the phrase “trial lawyer”—though it makes reference neither to plaintiffs nor to torts, and though trials have lawyers for both defendants and plaintiffs—has become broadly associated with plaintiffs’ tort lawyers. The subject of this book has not been tort litigation, but rather has been the role of private lawsuits in the implementation of federal statutory regulation—such as antitrust, banking, environmental, civil rights, and labor regulation. Based upon a series of case studies, Thomas Burke argues that ATLA is a distinctively powerful association that exerts influence on legislators (primarily in obstructing legislation) in the context of tort reform that is unmatched by lawyers’ associations in other substantive areas of law.41 Accordingly, the interest group and partisan dynamics governing tort law on the one hand, and federal statutory regulation on the other, cannot be assumed to be the same. This book does not attempt to draw inferences based upon the evidence presented here about interest group efficacy or partisan alignments in the domain of tort. Further, the case of tort may counsel openness to the possibility that there will be distinctive areas of federal regulation in which lawyers’ financial interests have played a significant role in shaping legislative outcomes respecting private enforcement.
CONCLUSION
233
Democracy and Private Enforcement Regimes The very large role of courts and lawsuits in American policy implementation has drawn considerable criticism on a variety of grounds. Among the most frequent criticisms is that it is deeply undemocratic, unsuited to a political community committed to representative democracy and legislative supremacy. On this view, the American policy landscape, particularly since the 1960s, has been marked by judicial intervention into one policy domain after another that is properly the province of the legislative or executive branches. This “judicial imperialism” has flowed from the decisions of unelected (in the federal system), “activist,” and usurpatious judges, in conjunction with irresponsible plaintiffs’ lawyers pursuing narrow interests that would never succeed in the ordinary legislative process. These judges and plaintiffs’ lawyers lack the democratic legitimacy of the elected branches, which have been empowered by, and can be disciplined and controlled by, the polity through elections.42 The democratic challenge is typically leveled against litigation in general, and does not analytically parse distinctions of constitutional, statutory, or common-law claims. Its denunciations, however, consistently have embraced domains of federal statutory regulation such as civil rights, environmental, education, employment, and housing policy. Contrary to the antidemocratic critique, other scholars regard federal judges as deeply tied to democratic processes,43 see policymaking through elected representatives as but one dimension of democracy,44 and characterize litigation as a form of citizen participation in the enterprise of self-government, arguing that private lawsuits are a valuable and important facet of democratic life in America.45 The relationship between democracy and the legislative choice of private enforcement regimes versus administrative power is complex, involving difficult empirical issues and deep normative questions. This book does not enter that fray, other than to observe that the evidence presented here calls into question the descriptive accuracy of important aspects of the antidemocratic critique as applied specifically to the role of private enforcement regimes within the sphere of federal statutory regulation. By and large, the litigation state was constructed by Congress, the most representative branch in our federal government. If Congress wishes to enact regulatory laws that are to be implemented exclusively through the administrative state, as it well understands, it is free to do so within constitutional bounds. As discussed in chapter 1, the explosion of litigation in the federal system since the late 1960s is accounted for primarily by lawsuits to enforce federal statutes, and, as shown in chapter 3, this
234
CHAPTER 7
explosive pattern of growth tracked, in lockstep, Congress’s proactive inclusion in federal statutes of economic incentives for private enforcement. Moreover, in the case of civil rights legislation, we observed that in debates leading to passage of civil rights laws in 1964, 1968, 1972, 1975, 1976, and 1991, the question of whether to rely upon private lawsuits, and if so, to what degree, was the subject of intense scrutiny, debate, and negotiation by legislators and interest groups. The institutional organ of government that produced the litigation-dominated civil rights enforcement framework that prevails in the United States was not only legislative, but was, over and over, a robustly democratic example of the legislative process. This, of course, does not make the enforcement framework wise or sensible. Serious analysts of regulation disagree about that question and will continue to do so. It was, however, most certainly democracy at work.
NOTES
Chapter 1 An Introduction to Private Enforcement Regimes 1. Annual Report of the Administrative Office of the United States Courts, 1997–2007, table C2. 2. See, e.g., Walter K. Olson, The Litigation Explosion: What Happened When America Unleashed the Lawsuit (New York: Thomas Talley Books–Dutton, 1991); Walter K. Olson, The Excuse Factory: How Employment Law Is Paralyzing the American Workplace (New York: Martin Kessler Books–Free Press, 1997); Patrick M. Garry, A Nation of Adversaries: How the Litigation Explosion Is Reshaping America (New York: Plenum Press, 1997); Michael D. Johnston, “The Litigation Explosion, Proposed Reforms, and Their Consequences,” Brigham Young University Journal of Public Law 21 (2007): 179–207. 3. Anthony Bell, “‘Killer Lawyers’ Leave Trail of Blood,” Los Angeles Business Journal, May 21, 2001: 67. 4. E.g., Robert A. Kagan, Adversarial Legalism: The American Way of Law (Cambridge: Harvard University Press, 2001); R. Shep Melnick, Between the Lines: Interpreting Welfare Rights (Washington, DC: Brookings Institution, 1994); R. Shep Melnick, Regulation and the Courts: The Case of the Clean Air Act (Washington, DC: Brookings Institution, 1983); Thomas F. Burke, Lawyers, Lawsuits, and Legal Rights: The Battle over Litigation in American Society (Berkeley and Los Angeles: University of California Press, 2002). 5. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI of Civil Rights Act of 1964 (Washington, DC: U.S. Government Printing Office, 1968), 5–6; Bureau of National Affairs, State Fair Employment Laws and Their Administration: Texts, Federal State Cooperation, Prohibited Acts (Washington, DC: Bureau of National Affairs, 1964). 6. Bureau of National Affairs, State Fair Employment Laws, 237–38. 7. Statement of Senator James Abourezk (D-SD), in support of the Civil Rights Attorney’s Fees Awards Act of 1976, Congressional Record, 94th Cong., 2nd sess., September 29, 1976, 17052. 8. See, e.g., Theda Skocpol, “Bringing the State Back In: Strategies of Analysis in Current Research,” in Bringing the State Back In, ed. Peter B. Evans, Dietrich Rueschemeyer, and Theda Skocpol (Cambridge: Cambridge University Press, 1985), 3–37, 15–16; Jess Gilbert and Carolyn Howe, “Beyond ‘State v. Society’: Theories of the State and New Deal Agricultural Policies,” American Sociological Review 56 (1991): 204–20, 206. 9. Christopher H. Foreman, Jr., “Regulatory Agencies,” in International Encyclopedia of the Social and Behavioral Sciences, ed. Neil J. Smelser and Paul B. Baltes (Amsterdam: Elsevier Science, 2001), 12982–85, 12982. 10. James Q. Wilson, Bureaucracy: What Government Agencies Do and Why They Do It (New York: Basic Books, 1989), 311.
236
NOTES
11. For a discussion of Weber’s view of the state, see Reinhard Bendix, Max Weber: An Intellectual Portrait (Garden City, NY: Anchor Books, 1962), 417– 23. For examples of this orientation in studies of American political development (APD), see Richard Bensel, Yankee Leviathan: The Origins of Central State Authority in America (New York: Cambridge University Press, 1990); Theda Skocpol, Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States (Cambridge: Harvard University Press, 1992); Stephen Skowronek, Building a New American State: The Expansion of National Administrative Capacities, 1877–1920 (Cambridge: Cambridge University Press, 1982); Edwin Amenta, Bold Relief: Institutional Politics and the Origins of Modern American Social Policy (Princeton: Princeton University Press, 1998). For a discussion of this orientation in political-institutional analysis of state capacity more broadly, see Nicholas Pedriana and Robin Stryker, “The Strength of a Weak Agency: Enforcement of Title VII of the 1964 Civil Rights Act and the Expansion of State Capacity, 1965–1971,” American Journal of Sociology 110 (2004): 709–60. 12. Skocpol, “Bringing State Back In,” 15–16. 13. Ira Katznelson, “Flexible Capacity: The Military and Early American State Building,” in Shaped by War and Trade: International Influences on American Political Development, ed. Ira Katznelson and Martin Shefter (Princeton: Princeton University Press, 2002), 86. 14. William J. Novak, “The Myth of the ‘Weak’ American State,” American Historical Review 113 (2008): 752–72. 15. See, e.g., Skowronek, Building New American State; Bensel, Yankee Leviathan; Skocpol, Protecting Soldiers and Mothers; Victoria C. Hattam, Labor Visions and State Power: The Origins of Business Unionism in the United States (Princeton: Princeton University Press, 1993); William E. Forbath, Law and the Shaping of the American Labor Movement (Cambridge: Harvard University Press, 1991); Howard Gillman, The Constitution Besieged: The Rise and Demise of Lochner Era Police Powers Jurisprudence (Durham: Duke University Press, 1993). 16. Paul Frymer makes this point in “Law and American Political Development,” Law and Social Inquiry 33 (2008): 779–803. His recent book, Black and Blue: African Americans, the Labor Movement, and the Decline of the Democratic Party (Princeton: Princeton University Press, 2007), is a notable exception to this limitation in APD scholarship. 17. The characterization of APD’s treatment of courts and litigation in the last two paragraphs is indebted to review articles on the topic by Paul Frymer, “Law and American Political Development,” and John D. Skrentny, “Law and the American State,” Annual Review of Sociology 32 (2006): 213–44. 18. Peter J. Katzenstein, Between Power and Plenty: Foreign Economic Policies of Advanced Industrial States (Madison: University of Wisconsin Press, 1978), 297. 19. Oscar G. Chase, “American ‘Exceptionalism’ and Comparative Procedure,” American Journal of Comparative Law 50 (2002): 277–301. 20. Kagan, Adversarial Legalism, 11–12, 105–6; Robert Cooter and Thomas Ulen, Law and Economics, 4th ed. (Boston: Pearson Addison Wesley, 2004), 411.
NOTES
237
21. Marc Galanter and David Luban, “Poetic Justice: Punitive Damage and Legal Pluralism,” American University Law Review 42 (1993): 1393–1463; see also Frances K. Zemans, “Legal Mobilization: The Neglected Role of the Law in the Political System,” American Political Science Review 77 (1983): 690–703. 22. Charles R. Epp, Making Rights Real: Activists, Bureaucrats, and the Creation of the Legalistic State (Chicago: University of Chicago Press, 2009); Alexandra Kalev and Frank Dobbin, “Enforcement of Civil Rights Law in Private Workplaces: The Effects of Compliance Reviews and Lawsuits over Time,” Law and Social Inquiry 31 (2006): 855–79; Sean Farhang, “Private Lawsuits, General Deterrence, and State Capacity: Evidence from Job Discrimination Litigation,” Goldman School of Public Policy Working Paper (2010); Frymer, Black and Blue; Michael Kent Block, Frederick Carl Nold, and Joseph Gregory Sidak, “The Deterrent Effect of Antitrust Enforcement,” Journal of Political Economy 89 (1981): 429–45. 23. Albert J. Reiss, “Selecting Strategies of Social Control of Organizational Life,” in Enforcing Regulation, ed. Keith Hawkins and John M. Thomas (Boston: Kluwer-Nijhoff, 1984), 23–35. 24. Michael Mann, The Sources of Social Power, vol. 2: The Rise of Classes and Nation-States, 1760–1914 (New York: Cambridge University Press, 1993), 59. 25. Katznelson, “Flexible Capacity,” 86. 26. Christopher Howard, The Hidden Welfare State: Tax Expenditures and Social Policy in the United States (Princeton: Princeton University Press, 1997); Jacob Hacker, The Divided Welfare State: The Battle over Public and Private Social Benefits in the United States (New York: Cambridge University Press, 2002). 27. Daniel P. Carpenter, The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innovation in Executive Agencies, 1862–1928 (Princeton: Princeton University Press, 2001). 28. Robert C. Lieberman, Shaping Race Policy: The United States in Comparative Perspective (Princeton: Princeton University Press, 2005). 29. Novak, “Myth of Weak American State,” 24. 30. Annual Report of the Administrative Office of the United States Courts, 1999–2009, table C-2. 31. See, e.g., Jeb Barnes and Mark C. Miller, “Governance as Dialogue,” in Making Policy, Making Law: An Interbranch Perspective, ed. Mark C. Miller and Jeb Barnes (Washington, DC: Georgetown University Press, 2004); Kagan, Adversarial Legalism; Melnick, Between the Lines; Michael McCann, “Litigation and Legal Mobilization,” in The Oxford Handbook of Law and Politics, ed. Keith E. Whittington, R. Daniel Kelemen, and Gregory A. Caldeira (New York: Oxford University Press, 2008), 522–40, 532–35 (summarizing the literature on interest group / law reform litigation). 32. Earlier years had many fewer cases, and thus I used a stratified sampling procedure to ensure that I obtained a sufficient number of cases in earlier periods to get a reliable representation of party and lawyer characteristics over time. A random sample of 25 cases per year was coded for the years 1900 to 2004. 33. David S. Law, “Strategic Judicial Lawmaking: Ideology, Publication, and Asylum Law in the Ninth Circuit,” University of Cincinnati Law Review 73
238
NOTES
(2005) 817–66, 826; Erica Weisgerber, “Unpublished Opinions: A Convenient Means to an Unconstitutional End,” Georgetown Law Journal 97 (2009): 621– 55, 627. Federal appellate courts began making extensive use of unpublished opinions in the early 1970s. Stephen L. Wasby, “Unpublished Court of Appeals Decisions: A Hard Look at the Process,” Southern California Interdisciplinary Law Journal 14 (2004): 67–124, 70. 34. Alexis de Tocqueville, Democracy in America, ed. J. P. Mayer (New York: Harper and Row, 1969), 270. 35. The data for this table was extracted from the Annual Report of the Administrative Office of the United States Courts, 1942–2005, table C-2. The private statutory litigation figures reflect the cases classified by the Administrative Office as private / federal question / statutory cases, excluding prisoner petitions and deportation cases. Unfortunately, pre-1942 reports do not distinguish between United States plaintiff and private plaintiff claims. 36. Marc Galanter, “Reading the Landscape of Disputes: What We Know and Don’t Know (and Think We Know) about Our Allegedly Contentious and Litigious Society,” University of California at Los Angeles Law Review 31 (1983): 4–71; Marc Galanter, “The Day after the Litigation Explosion,” Maryland Law Review 46 (1986): 3–39; Arthur R. Miller, “The Pretrial Rush to Judgment: Are the ‘Litigation Explosion,’ ‘Liability Crisis,’ and Efficiency Cliches Eroding Our Day in Court and Jury Trial Commitments?” New York University Law Review 78 (2003): 982–1134, 985–96; Lawrence M. Friedman, Total Justice (Boston: Beacon Press, 1987). 37. Kagan, Adversarial Legalism, chap. 1 (reviewing the comparative literature). 38. Seymour Martin Lipset, American Exceptionalism: A Double-Edged Sword (New York: Norton, 1996); Anthony King, “Ideas, Institutions and the Policies of Government,” British Journal of Political Science 3 (1973): 291–314. 39. Lipset, American Exceptionalism, 270. 40. Bayless Manning, “Hyperlexis: Our National Disease,” Northwestern University Law Review 71 (1977): 767. 41. Jerold S. Auerbach, “A Plague of Lawyers,” Harper’s, October 1976, 37– 44, 42. 42. “Chief Justice Urges Greater Use of Arbitration,” New York Times, August 22, 1985, A21. 43. Galanter, “Landscape of Disputes,” 10. 44. Mary Ann Glendon, Rights Talk: The Impoverishment of Political Discourse (New York: Free Press, 1991); see also Garry, A Nation of Adversaries; Olson, Litigation Explosion; Philip K. Howard, The Death of Common Sense: How Law Is Suffocating America (New York: Random House, 1994); Peter Huber, Liability: The Legal Revolution and Its Consequences (New York: Basic Books, 1988). 45. Michael Schudson, The Good Citizen: A History of American Civil Life (Cambridge: Harvard University Press, 1999), chap. 6; Samuel Walker, The Rights Revolution: Rights and Community in Modern America (New York: Oxford University Press, 1998); Eric Foner, The Story of American Freedom (New York: Norton, 1998), 299–305.
NOTES
239
46. Christian Wollschlager, “Exploring Global Landscapes of Litigation Rates,” in Soziologie des Rechts: Festschrift fur Erhard Blankenburg zum 60. Geburtstag, ed. Jurgen Brand and Dieter Strempel (Baden-Baden: Nomos Verlagsgesellschaft, 1998), 577–88. 47. Irving R. Kaufman, “Reform for a System in Crisis: Alternative Dispute Resolution in the Federal Courts,” Fordham Law Review 59 (1990): 1–39. 48. Friedman, Total Justice. 49. David S. Clark, “Civil Litigation Trends in Europe and Latin America since 1945: The Advantage of Intracountry Comparisons,” Law and Society Review 24 (1990): 549–69; Friedman, Total Justice; Wollschlager, “Exploring Global Landscapes.” 50. The data for this table was extracted from the Annual Report of the Administrative Office of the United States Courts, 1942–2005, table C-2. The private statutory litigation figures reflect cases classified by the Administrative Office as private / federal question / statutory cases, excluding prisoner petitions and deportation cases. The tort figures reflect cases classified, under tort, as private diversity cases. 51. I note that while increases in the amount in controversy (the amount of money that must be demanded by a plaintiff to trigger diversity jurisdiction under 42 U.S.C. § 1332) can depress tort filings while not affecting filings under federal statutes, no such change occurred in the periods I compare in the text in which private statutory litigation grew faster than tort litigation. 52. Friedman, Total Justice. 53. David Collier, Henry E. Brady, and Jason Seawright, “Sources of Leverage in Causal Inference: Toward and Alternative View of Methodology,” in Rethinking Social Inquiry: Diverse Tools, Shared Standards, ed. Henry E. Brady and David Collier (Lanham: Rowman and Littlefield, 2004), 229–66, 248–58; John Gerring, Case Study Research: Principles and Practices (New York: Cambridge University Press, 2007), 43–48.
Chapter 2 Institutional Foundations of Private Enforcement Regimes 1. Mathew McCubbins, Roger Noll, and Barry Weingast, “Administrative Procedures as Instruments of Control,” Journal of Law, Economics, and Organization 3 (1987): 243–77; see also Mathew D. McCubbins and Thomas Schwartz, “Congressional Oversight Overlooked: Police Patrols versus Fire Alarms,” American Journal of Political Science 28 (1984): 165–79; and Mathew McCubbins, Roger Noll, and Barry Weingast, “Structure and Process, Politics and Policy: Administrative Arrangements and Political Control of Agencies,” Virginia Law Review 75 (1989): 431–82. 2. Charles R. Shipan, Designing Judicial Review: Interest Groups, Congress and Communications Policy (Ann Arbor: University of Michigan Press, 1997). 3. McCubbins and Schwartz, “Congressional Oversight Overlooked,” 175 (emphasis added).
240
NOTES
4. Cooter and Ulen, Law and Economics, chap. 10; Mitchell Polinsky and Steven Shavell, “Punitive Damages: An Economic Analysis,” Harvard Law Review 111 (1998): 869–962; George L. Priest and Benjamin Klein, “The Selection of Disputes for Litigation,” Journal of Legal Studies 13 (1984): 1–55; Richard A. Posner, “An Economic Approach to Legal Procedure and Judicial Administration,” Journal of Legal Studies 2 (1973): 399–458. 5. Robert J. MacCoun, “Voice, Control, and Belonging: The Double-Edged Sword of Procedural Fairness,” Annual Review of Law and Social Science 1 (2005): 171–201. 6. Michael W. Giles and Thomas D. Lancaster, “Political Transition, Social Development, and Legal Mobilization in Spain,” American Political Science Review 3 (1989): 817–33. 7. Kristin Bumiller, The Civil Rights Society: The Social Construction of Victims (Baltimore: Johns Hopkins University Press, 1988). 8. Robert J. MacCoun, “Media Reporting of Jury Verdicts: Is the Tail (of the Distribution) Wagging the Dog?” DePaul University Law Review 55 (2006) 539– 62, 539. 9. Galanter and Luban, “Poetic Justice”; Earl Johnson, Jr., “Lawyers’ Choice: A Theoretical Appraisal of Litigation Investment Decisions,” Law and Society Review 15 (1980): 567–610; Herbert M. Kritzer, “Contingency Fee Lawyers as Gatekeepers in the Civil Justice System,” Judicature 81 (1997): 22–29; Herbert M. Kritzer, “Lawyer Fees and Lawyer Behavior in Litigation: What Does the Empirical Literature Really Say?” Texas Law Review 80 (2002): 1943–83. 10. Cooter and Ulen, Law and Economics, 390. 11. 42 U.S.C. § 2000e-5(f)(1). 12. Congress was explicit that it was not creating a private right of action under OSHA: the law provides that it should not “be construed to supersede or in any manner affect any workmen’s compensation law or to enlarge or diminish or affect in any manner the common law or statutory rights, duties, or liabilities of employers and employees under any law with respect to injuries, diseases, or death of employees arising out of, or in the course of, employment.” 29 U.S.C. § 653(b)(4). For courts construing this language accordingly, see, e.g., Russell v. Bartley, 494 F.2d 334 (6th Cir. 1974); Skidmore v. Traveler’s Ins., 356 F.Supp. 670 (E.D. La.), aff’d, 483 F.2d 67 (5th Cir. 1973). 13. These amendments to the original act (34 Stat. 262) are codified as 15 U.S.C. § 298, and the reasons for the amendments are set forth in House Report No. 928, 91st Cong., 2nd sess., 1970. 14. 42 U.S.C. § 2000e-5(e)(1). While the limitations period is 180 days if the claim is filed only with the EEOC, it can be extended to 300 days if it is first filed with a state civil rights agency. Plaintiffs must file with the EEOC within these time limits in order to have the subsequent right to pursue a court action. 15. 31 U.S.C. § 3731(b). 16. John F. Vargo, “The American Rule on Attorney Fee Allocation: The Injured Person’s Access to Justice,” American University Law Review 42 (1993): 1567–1636; Kritzer, “Lawyer Fees”; John Leubsdorf, “Toward a History of the
NOTES
241
American Rule on Attorney Fee Recovery,” Law and Contemporary Problems 47 (1984): 9–36; Werner Pfenningstorf, “The European Experience with Attorney Fee Shifting,” Law and Contemporary Problems 47 (1984): 37–124. 17. I am aware of no instance in which Congress has adopted a straight loserpays arrangement on the model of the English rule. In at least one instance, such a rule has been adopted at the state level. With respect to medical malpractice claims only, Florida adopted a loser-pays rule as part of tort reform aimed at reducing filings, but subsequently repealed the rule because of a number of unanticipated consequences which caused the law’s original advocates to push for its elimination. Kritzer, “Lawyer Fees,” 1949–50. 18. Kritzer, “Lawyer Fees” (reviewing literature). 19. Under the standard American rule, the plaintiff knows that EC will be her own expected costs of litigation whether she wins or loses; under the British rule, the plaintiff knows that EC will be 0 if she wins, since the defendant will have to pay her fees, and will be her own expected costs of litigation, plus the defendant’s expected costs of litigation, if she loses; under a one-way plaintiff’s shift, the plaintiff knows that EC will be her own expected costs of litigation if she loses, and will be 0 if she wins. Thus, if the plaintiff loses, she is in the same position as if she loses under the American rule (she pays her own fees and costs), and better off than under the English rule. If the plaintiff wins, she is in the same position as if she won under the English rule (the defendant pays her fees and costs), and better off than under the American rule. 20. Steven Shavell, “Suit, Settlement and Trial: A Theoretical Analysis under Alternative Methods for the Allocation of Legal Costs,” Journal of Legal Studies 11 (1982): 55–81; Frances Kahn Zemans, “Fee Shifting and the Implementation of Public Policy,” Law and Contemporary Problems 47 (1984): 187–210. 21. 42 U.S.C. § 2000e-5(k). Evidence of Congress’s intent is discussed in chapter 4. 22. Galanter and Luban, “Poetic Justice”; Polinsky and Shavell, “Punitive Damages.” 23. Civil Rights Act of 1964, § 706(g) (as passed), reprinted in Bureau of National Affairs, The Civil Rights Act of 1964: Text, Analysis, Legislative History; What it Means to Employers, Businessmen, Unions, Employees, Minority Groups (Washington, DC, 1964). 24. 29 U.S.C. § 216(b) (the Equal Pay Act incorporates the damages provisions of the Fair Labor Standards Act of 1938). 25. Sherman Antitrust Act of 1890, § 7 (as passed), reprinted in Irving J. Sloan, ed., American Landmark Legislation: Primary Materials (Dobbs Ferry, NY: Oceana, 1976–79), vol. 3. 26. Emergency Price Control Act of 1942, § 205(e), 56 Stat. 23, Public Law No. 77-421. 27. Cooter and Ulen, Law and Economics, 431–32. 28. Robert D. Cooter and Daniel L. Rubinfeld, “An Economic Model of Legal Discovery,” Journal of Legal Studies 23 (1994): 435–63. 29. Interstate Commerce Act of 1887, § 16, reprinted in Sloan, American Landmark Legislation, vol. 2.
242
NOTES
30. Interstate Commerce Act of 1887, Sloan, American Landmark Legislation, 2:280 (comments of Senator Spooner (R-WI)); 2:368 (comments of Senator Edmunds (R-VT)); 2:466 (comments of Representative Crisp (D-GA)). 31. 78 Stat. 241, § 706(a). 32. Indeed, a majority of courts have held that it is within the discretion of the trial court to reject introduction of an EEOC determination into evidence, based upon the judge’s assessment of its probative value relative to its potential prejudicial effects. See, e.g., Coleman v. Home Depot, Inc., 306 F.3d 1333 (3rd Cir. 2002); Paolitto v. Crawford, 151 F.3d 60, 65 (2nd Cir. 1998); Johnson v. Yellow Freight System, Inc., 734 F.2d 1304, 1309 (8th Cir. 1984). While a small minority of courts have expressed a strong presumption that an EEOC determination is admissible as evidence, e.g., Plummer v. Western Int’l Hotels Co., 656 F.2d 502, 505 (9th Cir. 1981), such courts have not required that the determination be accorded any particular weight by the fact finder, and thus a determination in favor of a plaintiff most certainly does not shift the burden to the defendant. 33. Marc Galanter, “Why the ‘Haves’ Come Out Ahead: Speculations on the Limits of Legal Change,” Law and Society Review 9 (1974): 95–160. 34. Charles R. Epp, The Rights Revolution: Lawyers, Activists, and Supreme Courts in Comparative Perspective (Chicago: University of Chicago Press, 1998). 35. Evidence for this proportion in the context of civil rights is provided in chapters 5 and 6 of this book. See also Richard H. Sander and E. Douglass Williams, “Why Are There So Many Lawyers? Perspectives on a Turbulent Market,” Law and Social Inquiry 14 (1989): 431–79, 470. 36. Morris P. Fiorina, “Legislative Choice of Regulatory Forms: Legal Process or Administrative Process?” Public Choice 39 (1982): 33–66; C. L. Schultze, The Public Use of Private Interests (Washington, DC: Brookings Institution, 1977). 37. Fiorina, “Legislative Choice”; for other scholars recognizing this basic regulatory choice, see also Eugene Bardach and Robert A. Kagan, Going by the Book: The Problem of Regulatory Unreasonableness, rev. ed. (Philadelphia: Transaction Publishers, 2002); Burke, Lawyers, Lawsuits; William N. Jr. Eskridge, Phillip P. Frickey, and Elizabeth Garrett, Cases and Materials on Legislation: Statutes and the Creation of Public Policy, 3rd ed. (Saint Paul: West Group, 2001), 1099; Thomas W. Gilligan, William J. Marshall, and Barry R. Weingast, “Regulation and the Theory of Legislative Choice: The Interstate Commerce Act of 1887,” Journal of Law and Economics 32 (1989): 35–61; Edward L. Glaeser and Andrei Shleifer, “The Rise of the Regulatory State,” Journal of Economic Literature 41 (2003): 401–25; Peter L. Strauss, Todd D. Rakoff, and Cynthia R. Farina, Gellhorn and Byse’s Administrative Law: Cases and Comments (New York: Foundation Press, 2003), 24–25. 38. Robert Dahl, “Decision-Making in a Democracy: The Supreme Court as a National Policy-Maker,” Journal of Public Law 6 (1957): 279–95; Martin Shapiro, “Political Jurisprudence,” Kentucky Law Journal 52 (1964): 294–345. 39. For excellent recent reviews of this literature, see Howard Gillman, “Courts and the Politics of Partisan Coalitions,” in Whittington, Kelemen, and Caldeira, Oxford Handbook of Law and Politics; Jeb Barnes, “Bringing the Courts Back In: Interbranch Perspectives on the Role of Courts in American Politics and Policy
NOTES
243
Making,” Annual Review of Political Science 10 (2007): 25–44; Mark A. Graber, “Constructing Judicial Review,” Annual Review of Political Science 8 (2005): 425–51. 40. Jack M. Balkin and Stanford Levinson, “Understanding the Constitutional Revolution,” Virginia Law Review 87 (2001): 1045–1109; see also Howard Gillman, “How Political Parties Can Use the Courts to Advance Their Agendas: Federal Courts in the United States, 1875–1891,” American Political Science Review 96 (2002): 511–24. Similar arguments have been made about why a ruling coalition would choose to create the institution of constitutional judicial review in the first place, divesting itself of power in the short run, but increasing the durability of its constitutional program in the long run. Tom Ginsburg, Judicial Review in New Democracies: Constitutional Courts in Asian Cases (New York: Cambridge University Press, 2003); Ran Hirschl, Toward Juristocracy: The Origins and Consequences of the New Constitutionalism (Cambridge: Harvard University Press, 2004). Ginsburg calls this an “insurance theory” of judicial review, and Hirschl calls it “hegemonic preservation.” 41. Keith Whittington, “‘Interpose Your Friendly Hand’: Political Supports for the Exercise of Judicial Review by the United States Supreme Court,” American Political Science Review 99 (2005): 583–96. 42. William M. Landes and Richard A. Posner, “The Independent Judiciary in an Interest-Group Perspective,” Journal of Law and Economics 18 (1975): 875–901; Gillman, “Courts and Partisan Coalitions,” 648. 43. Kagan, Adversarial Legalism, 15. 44. See also Friedman, Total Justice. 45. Burke, Lawyers, Lawsuits, 173–74; see also Joseph L. Smith, “Congress Opens the Courthouse Doors: Statutory Changes to Judicial Review under the Clean Air Act,” Political Research Quarterly 58 (2005): 139–49. 46. Terry M. Moe, “The Politics of Bureaucratic Structure,” in Can the Government Govern? ed. John E. Chubb and Paul E. Peterson (Washington, DC: Brookings Institution, 1989); Terry M. Moe, “Political Institutions: The Neglected Side of the Story,” Journal of Law, Economics, and Organization 6 (1990): 213–53; Terry Moe and Michael Caldwell, “The Institutional Foundations of Democratic Government: A Comparison of Presidential and Parliamentary Systems,” Journal of Institutional and Theoretical Economics 150 (1994): 171–95. 47. Moe, “Politics of Bureaucratic Structure”; “Political Institutions.” 48. Randall Calvert, Mark Moran, and Barry Weingast, “Congressional Influence over Policymaking: The Case of the FTC,” in Congress: Structure and Policy, ed. Mathew McCubbins and Terry Sullivan (New York: Cambridge University Press, 1987); Barry R. Weingast and Mark J. Moran, “Bureaucratic Discretion or Congressional Control? Regulatory Policymaking by the Federal Trade Commission,” Journal of Political Economy 91 (1983): 765–800; Barry R. Weingast, “The Congressional-Bureaucratic System: A Principal-Agent Perspective (with Applications to the SEC),” Public Choice 44 (1984): 147–77. 49. Moe, “Politics of Bureaucratic Structure”; “Political Institutions”; Moe and Caldwell, “Institutional Foundations.” 50. Moe, “Politics of Bureaucratic Structure,” 282.
244
NOTES
51. David Epstein and Sharyn O’Halloran, Delegating Powers: A Transaction Cost Politics Approach to Policy Making under Separate Powers (New York: Cambridge University Press, 1999). 52. John D. Huber and Charles R. Shipan, Deliberate Discretion? The Institutional Foundations of Bureaucratic Autonomy (New York: Cambridge University Press, 2002). 53. Keith E. Hamm and Roby D. Robertson, “Factors Influencing the Adoption of New Methods of Legislative Oversight in the U.S. States,” Legislative Studies Quarterly 6 (1981): 133–50. 54. Kagan, Adversarial Legalism, 48–49; Burke, Lawyers, Lawsuits, 14–15, 173; R. Shep Melnick, “From Tax-and-Spend to Mandate-and-Sue: Liberalism after the Great Society,” in The Great Society and the High Tide of Liberalism, ed. Sidney Milkis and Jerome M. Mileur (Amherst: University of Massachusetts Press, 2005); Tom Ginsburg and Robert A. Kagan, “Introduction: Institutionalist Approaches to Courts as Political Actors,” in Institutions and Public Law: Comparative Approaches, ed. Tom Ginsburg and Robert A. Kagan (New York: Peter Lang, 2005), 6–7. 55. Epstein and O’Halloran find that during divided government, while Congress is less likely to delegate to executive agencies, it is more likely to make “nonexecutive” delegations of authority, which are measured by an amalgamation of delegations of authority to state agencies, local authorities, and the courts. Delegating Powers, 156–57. However, no separate analysis of delegation to the courts is provided, nor do they discuss in detail what they count as delegation to courts, and thus while their findings are suggestive, it is not possible to conclude from their work whether or not Congress delegated more power to courts during periods of divided government. 56. E.g., David Mayhew, Congress: The Electoral Connection (New Haven: Yale University Press, 1974). 57. Moe, “Political Institutions”; Moe and Caldwell, “Institutional Foundations”; Terry Moe, “The Positive Theory of Public Bureaucracy,” in Perspectives on Public Choice: A Handbook, ed. Dennis Mueller (Cambridge: Cambridge University Press, 1997), 455–80, 468–69. 58. Kenneth A. Shepsle, “Bureaucratic Drift, Coalition Drift, and Time Consistency: A Comment on Macey,” Journal of Law, Economics, and Organization 8 (1992): 111–18, 114–15. 59. Moe, “Politics of Bureaucratic Structure,” 274–75 (emphasis added). 60. Ellen M. Immergut, “The Rules of the Game: The Logic of Health PolicyMaking,” in Structuring Politics: Historical Institutionalism in Comparative Analysis, ed. Sven Steinmo, Kathleen Thelen, and Frank Longstreth (New York: Cambridge University Press, 1992), 57–89, 63–64. 61. Matthew S. Shugart and John M. Carey, Presidents and Assemblies: Constitutional Design and Electoral Dynamics (Cambridge: Cambridge University Press, 1992); Kent R. Weaver and Burt A. Rockman, “Assessing the Effects of Institutions,” in Do Institutions Matter? ed. Kent R Weaver and Burt A. Rockman (Washington, DC: Brookings Institution, 1993); James L. Sundquist, “Needed: A Political Theory for the New Era of Coalition Government in the United States,” Political Science Quarterly 103 (1988–89): 613–35.
NOTES
245
62. Landes and Posner, “Independent Judiciary”; David W. Brady and Craig Volden, Revolving Gridlock Politics and Policy from Jimmy Carter to George W. Bush, 2nd ed. (Boulder: Westview Press, 2005); Sven H. Steinmo, “American Exceptionalism Reconsidered: Culture or Institutions?” in The Dynamics of American Politics: Approaches and Interpretations, ed. Lawrence C. Dodd and Calvin Jillson (Boulder: Westview Press, 1994), 106–31; Weaver and Rockman, “Assessing Effects of Institutions.” 63. Moe, “Political Institutions,” 240. 64. Kagan, Adversarial Legalism, 49; Burke, Lawyers, Lawsuits, 173–74; Landes and Posner, “Independent Judiciary”; Ginsburg and Kagan, “Introduction: Institutionalist Approaches,” 6. 65. Joel D. Aberbach, Keeping a Watchful Eye: The Politics of Congressional Oversight (Washington, DC: Brookings Institution, 1990); Eskridge, Frickey, and Garrett, Cases and Materials on Legislation, 1129–73; Frank B. Cross, “Shattering the Fragile Case for Judicial Review of Rulemaking,” Virginia Law Review 85 (1999): 1243–1334, 1295–96. 66. Cross, “Shattering the Fragile Case,” 1303–5; Charles R. Shipan, “Interest Groups, Judicial Review, and the Origins of Broadcast Regulation,” Administrative Law Review 49 (1997): 549–84, 555–56; Shipan, Designing Judicial Review, 10; McCubbins, Noll, and Weingast, “Structure and Process,” 444–45. 67. Landes and Posner, “Independent Judiciary”; the assumption of judicial fidelity to the intent of the enacting coalition is doubted, for example, by Richard A. Epstein, “The Independence of Judges: The Uses and Limitations of Public Choice Theory,” Brigham Young University Law Review 827 (1990): 827–55, 851–52; see also Jeffrey A. Segal and Harold J. Spaeth, The Supreme Court and the Attitudinal Model Revisited (Cambridge: Cambridge University Press, 2002). 68. McCubbins, Noll, and Weingast, “Administrative Procedures,” 247. 69. McCubbins, Noll, and Weingast, “Administrative Procedures,” 248, 254–55. 70. On the obstacles to statutory override that create this bureaucratic latitude, see also Cross, “Shattering the Fragile Case,” 1303–4 (summarizing the literature on the topic). 71. McCubbins, Noll, and Weingast, “Structure and Process”; “Administrative Procedures.” 72. Murray Horn and Kenneth Shepsle, “Commentary on ‘Administrative Arrangements and the Political Control of Agencies’: Administrative Process and Organizational Form as Legislative Responses to Agency Costs,” Virginia Law Review 75 (1989): 499–508, 504; see also Moe, “Positive Theory,” 468–69. 73. The logic of McNollgast’s argument would appear to call for measures, in multidimensional policy space, of relevant committee chairs, floor medians (possibly the filibuster in the Senate), and the president, to calculate the size of the space in which bureaucratic drift can occur without triggering a legislative reversal. “Structure and Process,” 435–40. I am aware of no research that has endeavored to construct such a complex measure. If one were to construct, in the alternative, a crude measure of the size of the space in which bureaucratic drift can occur, in one-dimensional space, as the distance from Congress (measured as a single point) to the president, one would have a simple measure of legislative-executive conflict, which is already incorporated in the model to test
246
NOTES
the legislative-executive conflict hypothesis. That the ideological distance between Congress and the president encourages enactment of private enforcement regimes by legislators seeking to avoid presidential subversion, rather than encouraging enactment of private enforcement regimes by Congress and the president collaborating together to avoid subversion by bureaucrats, is amply borne out by the qualitative evidence in the second part of the book. 74. See, e.g., Hattam, Labor Visions; Robert C. Lieberman, Shifting the Color Line: Race and the American Welfare State (Cambridge: Harvard University Press, 1998); Sean Farhang and Ira Katznelson, “The Southern Imposition: Congress and Labor in the New Deal and Fair Deal,” Studies in American Political Development 19 (1995): 1–30; Theda Skocpol and Gretchen Ritter, “Gender and the Origins of Modern Social Policies in Brittan and the United States,” Studies in American Political Development 5 (1991): 36–93; Sven Steinmo, Taxation and Democracy: Swedish, British and American Approaches to Financing the Modern State (New Haven: Yale University Press, 1996). 75. Steinmo, “American Exceptionalism Reconsidered,” 126. 76. Howard, The Hidden Welfare State, 10–12; Hacker, The Divided Welfare State, 24–25, 45–46. 77. Kagan, Adversarial Legalism, 15–16, 50–51, 193–94; Burke, Lawyers, Lawsuits, 13–14, 172–73; Ginsburg and Kagan, “Introduction: Institutionalist Approaches,” 8; Lipset, American Exceptionalism, 21. A detailed example of the dynamic described in this paragraph is provided in chapter 4, which examines the legislative struggle that produced the job discrimination title of the Civil Rights Act of 1964. 78. M. Elizabeth Magill, “Agency Choice of Policymaking Form,” University of Chicago Law Review 71 (2004): 1383–1447; Strauss, Rakoff, and Farina, Gellhorn and Byse’s Administrative Law, 11. 79. See Lynn Mather, “Theorizing about Trial Courts: Lawyers, Policymaking, and Tobacco Litigation,” Law and Social Inquiry 23 (1998): 897–940 (reviewing literature on policy effects of litigation). 80. Jeb Barnes, Overruled? Legislative Overrides, Pluralism, and Contemporary Court-Congress Relations (Stanford: Stanford University Press, 2004). 81. Christopher B. Busch, David L. Kirp, and Daniel F. Schoenholz, “Taming Adversarial Legalism: The Port of Oakland’s Dredging Saga Revisited,” Journal of Legislation and Public Policy 2 (1999): 179–216. 82. Roy B. Flemming, John Bohte, and B. Dan Wood, “One Voice among Many: The Supreme Court’s Influence on Attentiveness to Issues in the United States, 1947–92,” American Journal of Political Science 41 (1997): 1224–50. 83. Marc Galanter, “Case Congregations and Their Careers,” Law and Society Review 24 (1990): 371–95; Michael McCann, “Reform Litigation on Trial,” Law and Social Inquiry 21 (1992): 457–82. 84. Michael McCann, Rights at Work: Pay Equity Reform and the Politics of Legal Mobilization (Chicago: University of Chicago Press, 1994); “Reform Litigation on Trial.” 85. Robert A. Katzmann, “Making Sense of Congressional Intent: Statutory Interpretation and Welfare Policy,” Yale Law Journal 104 (1995): 2345–67, 2346 (emphasis added).
NOTES
247
86. Melnick, Between the Lines; Regulation and the Courts; William N. Eskridge, Jr., Dynamic Statutory Interpretation (Cambridge: Harvard University Press, 1994); Paul Frymer, “Acting When Elected Officials Won’t: Federal Courts and Civil Rights Enforcement in US Labor Unions, 1935–85,” American Political Science Review 97 (2003): 483–99; Richard A. Posner, “Legal Formalism, Legal Realism, and the Interpretation of Statutes and the Constitution,” Case Western Law Review 37 (1987): 179–217; Max Radin, “Statutory Interpretation,” Harvard Law Review 43 (1930): 863, 870–71; Martin Shapiro, Law and Politics in the Supreme Court (London: Free Press of Glencoe, 1964). 87. Jerry L. Mashaw, Greed, Chaos, and Governance: Using Public Choice to Improve Public Law (New Haven: Yale University Press, 1997); Peter L. Strauss, “On Resegregating the Worlds of Statute and Common Law,” Supreme Court Review 9 (1994): 429–540. 88. Katzmann, “Making Sense of Congressional Intent”; Theodore J. Lowi, The End of Liberalism: The Second Republic of the United States (New York: Norton, 1979); Mashaw, Greed, Chaos, and Governance; Melnick, Between the Lines; Abner J. Mikva, “Reading and Writing Statutes,” University of Pittsburgh Law Review 48 (1987): 627–37; Posner, “Legal Formalism.” 89. Melnick, Between the Lines; Douglas R. Arnold, The Logic of Congressional Action (New Haven: Yale University Press, 1990); Patrick S. Atiyah and Robert S. Summers, Form and Substance in Anglo-American Law: A Comparative Study of Legal Reasoning, Legal Theory, and Legal Institutions (New York: Oxford University Press, 1987); George Lovell, Legislative Deferrals: Statutory Ambiguity, Judicial Power, and American Democracy (Cambridge: Cambridge University Press, 2003); Epstein and O’Halloran, Delegating Powers; Morris P. Fiorina, Congress, Keystone of the Washington Establishment (New Haven: Yale University Press, 1977); Fiorina, “Legislative Choice”; Huber and Shipan, Deliberate Discretion?; Moe, “Politics of Bureaucratic Structure”; Mikva, “Reading and Writing Statutes.” 90. Eskridge, Dynamic Statutory Interpretation; Mikva, “Reading and Writing Statutes”; Richard A. Posner, “Statutory Interpretation—in the Classroom and in the Courtroom,” University of Chicago Law Review 50 (1983): 800–822. 91. Donald R. Songer, Sue Davis, and Susan Haire, “A Reappraisal of Diversification in the Federal Courts: Gender Effects in the Courts of Appeals,” Journal of Politics 56 (1994): 425–39; Sean Farhang and Greg Wawro, “Institutional Dynamics on the U.S. Court of Appeals: Minority Representation under Panel Decision Making,” Journal of Law, Economics, and Organization 20 (2004): 299–330. 92. James J. Brudney, Sara Schiavoni, and Deborah J. Merritt, “Judicial Hostility toward Labor Unions? Applying the Social Background Model to a Celebrated Concern,” Ohio State Law Journal 60 (1999): 1675–1771; Deborah Jones Merritt and James J. Brudney, “Stalking Secret Law: What Predicts Publication in the United States Courts of Appeals,” Vanderbilt Law Review 54 (2001): 71–121. 93. Richard Revesz, “Environmental Regulation, Ideology, and the D.C. Circuit,” Virginia Law Review 83 (1997): 1717–72. 94. Daniel M. Schneider, “Empirical Research on Judicial Reasoning: Statutory Interpretation in Federal Tax Cases,” New Mexico Law Review 31 (2001):
248
NOTES
325–52; Daniel M. Schneider, “Assessing and Predicting Who Wins Federal Tax Trial Decisions,” Wake Forest Law Review 37 (2002): 473–529. 95. Law, “Strategic Judicial Lawmaking.” 96. Charles M. Cameron, Veto Bargaining: Presidents and the Politics of Negative Power (New York: Cambridge University Press, 2000). 97. Epstein and O’Halloran, Delegating Powers; Huber and Shipan, Deliberate Discretion?; Hamm and Robertson, “Factors Influencing Adoption.” 98. Landes and Posner, “Independent Judiciary.” 99. Segal and Spaeth, The Supreme Court and the Attitudinal Model. 100. Margaret H. Lemos, “The Consequences of Congress’s Choice of Delegate: Judicial and Agency Interpretations of Title VII,” Vanderbilt Law Review (2010, forthcoming); Pierce, Shapiro, and Verkuil, Administrative Law, 21, § 1.9. 101. Stuart Minor Benjamin and Arti K. Rai, “Who’s Afraid of the APA? What the Patent System Can Learn from Administrative Law,” Georgetown Law Journal 95 (2007) 269-336, 311-12; Burke, Lawyers, Lawsuits, 14-15. 102. Cross, “Shattering the Fragile Case,” 1295-96, 1303–5; McCubbins, Noll, and Weingast, “Structure and Process,” 444–45. 103. See, e.g., Tamar Frankel, “Implied Rights of Action,” Virginia Law Review 67 (1981): 553–85; Andrew M. Siegel, “The Court against the Courts: Hostility to Litigation as an Organizing Theme in the Rehnquist Court’s Jurisprudence,” Texas Law Review 84 (2006): 1097–1202; Ray Terry, “Eliminating the Plaintiff’s Attorney in Equal Employment Litigation: A Shakespearian Tragedy,” Labor Lawyer 5 (1989): 63–81; chapter 6 of this book. 104. Kagan, Adversarial Legalism, 45-46; Burke, Lawyers, Lawsuits, 15, 18081; Ginsburg and Kagan, “Introduction: Institutionalist Approaches,” 8. 105. Ginsburg and Kagan, “Introduction: Institutionalist Approaches,” 8. 106. Burke, Lawyers, Lawsuits, 15, 180-81 107. Matthew C. Stephenson, “Public Regulation of Private Enforcement: The Case for Expanding the Role of Administrative Agencies,” Virginia Law Review 91 (2005): 93–173; Farhang, “Private Lawsuits”; Robert Kagan, “Adversarial Legalism: Tamed or Still Wild?” New York University Journal of Legislation and Public Policy 2 (1999): 217–45. 108. Stephenson, “Public Regulation”; Farhang, “Private Lawsuits.” 109. See especially Moe, “Political Institutions”; Moe and Caldwell, “Institutional Foundations.” 110. Wilson, Bureaucracy, chap. 16. 111. Moe, “Political Institutions,” 241. 112. Moe, “Political Institutions,” 240. 113. Kagan, Adversarial Legalism, 6–9 (reviewing literature). 114. Kagan, Adversarial Legalism, 16. 115. For discussion of these assumptions in rational choice institutionalism, see Peter Ordershook, “The Emerging Discipline of Political Economy,” in Perspectives on Positive Political Economy, ed. James Alt and Kenneth Shepsle (Cambridge: Cambridge University Press, 1990), 9–30, 13; James Alt and Kenneth Shepsle, “Editors’ Introduction,” in Perspectives on Positive Political Economy, 1–5, 2; Robert O. Keohane, After Hegemony: Cooperation and
NOTES
249
Discord in the World Political Economy (Princeton: Princeton University Press, 1984), 80. 116. Paul Pierson, Politics in Time: History, Institutions, and Social Analysis (Princeton: Princeton University Press, 2004), chap. 4; Robert Jervis, System Effects: Complexity in Political and Social Life (Princeton: Princeton University Press, 1997), chap. 4; Hugh Heclo, Modern Social Politics in Britain and Sweden (New Haven: Yale University Press, 1974), 305–16; Theda Skocpol, “The Origins of Social Policy in the United States,” in The Dynamics of American Politics, ed. Lawrence C. Dodd and Calvin Jillson (Boulder: Westview Press, 1994); Eric Patashnik, “After the Public Interest Prevails: The Political Sustainability of Policy Reform,” Governance 16 (2003): 203–34.
Chapter 3 An Empirical Model of Enactment of Private Enforcement Regimes 1. E.g., Burke, Lawyers, Lawsuits; Frymer, “Acting When Elected Officials Won’t”; McCubbins, Noll, and Weingast, “Administrative Procedures”; McCubbins, Noll, and Weingast, “Structure and Process”; Melnick, Regulation and the Courts; Between the Lines; McCubbins and Schwartz, “Congressional Oversight Overlooked”; Charles R. Shipan, Designing Judicial Review. 2. Leubsdorf, “History of American Rule”; Pfenningstorf, “European Experience.” 3. Buckhannon Bd. & Care Home v. W. Va. Dep’t of Health & Human Res., 532 U.S. 598, 602 (2001); Key Tronic Corp. v. United States, 511 U.S. 809, 819 (1994). 4. Shavell, “Suit, Settlement and Trial”; Zemans, “Fee Shifting”; Kritzer, “Lawyer Fees.” 5. Alyeska Pipeline Serv. Co. v. Wilderness Soc’y, 421 U.S. 240, 263 (1975). 6. Galanter and Luban, “Poetic Justice”; Charles T. McCormick, On the Law of Damages (Chicago: Foundation Press, 1935), 85–87. 7. Smith v. Wade, 461 U.S. 30, 58 (1983); see also United States v. Snepp, 595 F.2d 926, 941 (4th Cir. 1978); TVT Records v. Island Def Jam Music Group, 2003 U.S. Dist. LEXIS 15271 (S.D.N.Y. 2003). 8. Jackson v. Johns-Manville Sales Corp., 781 F.2d 394, 403 (5th Cir. 1986). 9. Sedima, S.P.R.L. v. Imrex Co., 473 U.S. 479, 493 (1985). See also David Luban, “Response: A Flawed Case against Punitive Damages,” Georgetown Law Journal 87 (1998): 359–80. 10. Andrea Catania, “State Employment Discrimination Remedies and Pendent Jurisdiction under Title VII: Access to Federal Courts,” American University Law Review 32 (1983): 777–838, 807 n. 141; Note, “Practice and Potential of the Advisory Jury,” Harvard Law Review 100 (1987): 1363–81, 1374–75; Mark R. Pettit, “The Right to Jury Trial under Title VII of the Civil Rights Act of 1964,” University of Chicago Law Review 37 (1969): 167–80, 167; see also chapter 4 of this book. 11. See chapter 5 of this book.
250
NOTES
12. Cass R. Sunstein, “What’s Standing after Lujan? Of Citizen Suits, ‘Injuries,’ and Article III,” Michigan Law Review 91 (1992): 163–236. 13. Robert G. Blakey and Scott D. Cessar, “Equitable Relief under Civil RICO: Reflections on Religious Technology Center v. Wollersheim: Will Civil RIOC be Effective Only against White-Collar Crime?” Notre Dame Law Review 62 (1987): 526–623, 531 n. 17. 14. 16 Stat. 140–41, §§ 2–4. 15. Floor debates on the ICA of 1887, reprinted in Sloan, American Landmark Legislation, 3:280–81. 16. Floor debates on the ICA of 1887, Sloan, American Landmark Legislation, 3:471–72. 17. Floor debates on the ICA of 1887, Sloan, American Landmark Legislation, 3:466. 18. Congressional Record, 51st Cong., 1st sess. (1889), 21:2456. 19. Congressional Record, 51st Cong., 1st sess. (1889), 21:2456. 20. Congressional Record, 51st Cong., 1st sess. (1889), 21:3147. 21. Only provisions that applied to de novo court actions by private plaintiffs, and explicitly provided for recovery of attorney’s fees, or multiple or punitive damages, were included in the dataset. The procedure for collecting the dependent variable is described in an appendix to this chapter. 22. In a handful of instances Congress provided in a statute that prevailing plaintiff-enforcers of the statute would not be entitled to punitive damages (no analogous provisions occurred concerning attorney’s fees). I did not subtract these instances from the annual count because such provisions only occurred in laws in which Congress did create a private right of action and some economic damages, giving rise to Congress’s desire to make explicit that it did not intend such damages to include punitive damages. It would be erroneous to count such litigation fostering laws as antilitigation laws. In order to evaluate the robustness of the findings presented below to this choice, I examined an alternative specification in which I subtracted these provisions from the annual count, and the findings were statistically and substantively unaffected. 23. The private statutory litigation figures reflect cases classified by the Administrative Office of the United States Courts as private / federal question / statutory cases, excluding prisoner petitions and deportation cases. 24. Public Law No. 75-718. 25. Public Law No. 88-352. 26. Public Law No. 103-297. 27. Public Law No. 80-101. These provisions in Taft-Hartley are not contained in the dependent variable because, while allowing recovery of economic damages for injury to a business and the costs of suit, 29 U.S.C.A. § 187, it did not contain a punitive or multiple damages provision, or allow recovery of attorney’s fees. 28. Public Law No. 104-114. 29. Public Law No. 108-105. 30. Frymer, Black and Blue; Phillip B. Heymann and Lance Liebman, “No Fault, No Fee: The Legal Profession and Federal No-Fault Automobile Insurance Legislation,” in The Social Responsibilities of Lawyers, ed. Phillip B. Heymann and Lance Liebman (Westbury: Foundation Press, 1988), 309; Alan S. Rau, Ed-
NOTES
251
ward F. Sherman, and Scott R. Peppet, The Process of Dispute Resolution: The Role of Lawyers, 4th ed. (New York: Foundation Press, 2006), 56; Michelle J. White, “Legal Complexity and Lawyers’ Benefit from Litigation,” International Review of Law and Economics 12 (1992): 381–95, 394–95. 31. White, “Legal Complexity,” 394. 32. Frank R. Baumgartner and Beth L. Leech, “Interest Niches and Policy Bandwagons: Patterns of Interest Group Involvement in National Politics,” Journal of Politics 63 (2001): 1101–1213, 1191. 33. R. Shep Melnick, “Courts and Agencies,” in Miller and Barnes, Making Policy, Making Law, 93; R. Shep Melnick, “Separation of Powers and the Strategy of Rights: The Expansion of Special Education,” in The New Politics of Public Policy, ed. Marc Landy and Martin Levin (Baltimore: Johns Hopkins Press, 1995), 23–46; Michael McCann, Taking Reform Seriously: Perspectives on Public Interest Liberalism (Ithaca: Cornell University Press, 1986); Burke, Lawyers, Lawsuits; Kagan, Adversarial Legalism, 38–39, 47; Karen O’Connor and Lee Epstein, “Bridging the Gap between Congress and the Supreme Court: Interest Groups and the Erosion of the American Rule Governing Award of Attorney’s Fees,” Western Political Quarterly 38 (1985): 238–49; David Vogel, “The ‘New’ Social Regulation in Historical and Comparative Perspective,” in Regulation in Perspective, ed. Thomas K. McCraw (Cambridge: Harvard University Press, 1981), 155–85; William L. Andreen, “The Evolving Law of Environmental Protection in the United States: 1970–1991,” Environmental Planning and Law Journal 9 (1991): 96–110, 98; Martin Shapiro, Who Guards the Guardians? Judicial Control of Administration (Athens: University of Georgia Press, 1988), 55–77; Martin Shefter, Political Parties and the State: The American Historical Experience (Princeton: Princeton University Press, 1994), 86–94. 34. There is a debate among scholars over why this alignment occurs. Some maintain that legislators adopt positions in exchange for interest groups’ support in one form or another. Rebecca Morton and Charles Cameron, “Elections and the Theory of Campaign Contributions,” Economics and Politics 4 (1992): 79–108; Andreas Dür and Dirk de Bièvre, “The Question of Interest Group Influence,” Journal of Public Policy 27 (2007): 1–12, 5–6; Robert A. Dahl, Who Governs? Democracy and Power in an American City (New Haven: Yale University Press, 1961), 226. Others maintain that interest groups enjoy a comparative advantage relative to legislators, their staff, and their party in procuring private and costly information about a constituency’s preferences, and they can share the information strategically to alter legislators’ views about their electoral self-interest, and thereby change legislative behavior in the direction of the group’s preferences. John Mark Hansen, Gaining Access: Congress and the Farm Lobby, 1919–1981 (Chicago: University of Chicago Press, 1991), 3; John R. Wright, Interest Groups and Congress (Boston: Allyn and Bacon, 1994), 81. Yet others maintain that interest groups do not so much change legislators’ positions as “assist natural allies in achieving their own, coincident objectives” by supplying legislators with costly policy information and expertise, strategic political intelligence and advice, and auxiliary labor for the legislator’s staff. Richard L. Hall and Alan V. Deardorff, “Lobbying and Legislative Subsidy,” American Political Science Review 100 (2006): 69–84. The first two theories would straightforwardly explain how inter-
252
NOTES
est groups induce legislators to adopt private enforcement regimes. Even under the third theory, interest groups would be in a privileged position to argue the importance and utility of private enforcement regimes, as a policy tool, to advance groups’ and legislators’ coincident policy goals. Thus, while the three theories differ in how groups participate in a legislative coalition to influence legislative outputs, all three agree that groups can in fact do so. 35. E.g., Kagan, Adversarial Legalism, 148–52; Burke, Lawyers, Lawsuits, 47–48; Sara Parikh, “How the Spider Catches the Fly: Referral Networks in the Plaintiffs’ Personal Injury Bar,” New York Law School Law Review 51 (2006–7): 243–83, 250–51. 36. E.g., Randolph I. Gordon and Brook Assefa, “A Tale of Two Initiatives: Where Propaganda Meets Fact in the Debate over America’s Health Care,” Seattle Journal for Social Justice 4 (2006): 693–744; Roberta Romano, “The SarbanesOxley Act and the Making of Quack Corporate Governance,” Yale Law Journal 114 (2005): 1521–1611; Stephen C. Yeazell, “Brown, the Civil Rights Movement, and the Silent Litigation Revolution,” Vanderbilt Law Review 57 (2004): 1975–2003. 37. Hodge F. O’Neal, “Class Actions and Limited Vision: Opportunities for Improvement through a More Functional Approach to Class Treatment of Disputes,” Washington University Law Quarterly 83 (2005): 1127–1269, 1142 n. 61; Glenn S. Koppel, “Populism, Politics, and Procedure: The Saga of Summary Judgment and the Rulemaking Process in California,” Pepperdine Law Review 24 (1997): 455–561, 475. 38. Stephen J. Ware, “Money, Politics and Judicial Decisions: A Case Study of Arbitration Law in Alabama,” Journal of Law and Politics 15 (1999): 645–86, 657–58. 39. Romano, “Sarbanes-Oxley Act,” 1561; see also Koppel, “Populism, Politics, and Procedure,” 475. 40. On issue groups as part of the Democratic party coalition, see Shefter, Political Parties and the State, 86-94; Michael S. Greve, “Why Defunding the Left Failed,” Public Interest 89 (1987): 91–106, 101; Burke, Lawyers, Lawsuits, 44; Koppel, “Populism, Politics, and Procedure,” 475. Regarding Democrats being the more interventionist party in the twentieth century, see Keith Poole and Howard Rosenthal, Congress: A Political-Economic History of Roll Call Voting (New York: Oxford University Press, 1997); Scott James, Presidents, Parties, and the State: A Party System Perspective on Democratic Regulatory Choice, 1884–1936 (New York: Cambridge University Press, 2000). 41. Burke, Lawyers, Lawsuits, 15–16; Melnick, “From Tax-and-Spend”; Kagan, Adversarial Legalism, 15–16; O’Connor and Epstein, “Bridging the Gap.” 42. Steinmo, Taxation and Democracy; Harold Wilenski, The Welfare State and Equality: Structural and Ideological Roots of Public Expenditures (Berkeley and Los Angeles: University of California Press, 1975). 43. Epstein and O’Halloran, Delegating Powers, 129. 44. Poole and Rosenthal, Congress; Keith T. Poole, “Recovering a Basic Space from a Set of Issue Scales,” American Journal of Political Science 42 (1998): 954–93.
NOTES
253
45. Nolan McCarty, “Presidential Vetoes in the Early Republic: Changing Constitutional Norms or Electoral Reform?” Journal of Politics 71 (2009): 369–84. 46. McCarty’s presidential scores are calculated using the DW-NOMINATE procedure on presidential positions taken on roll calls in the House. In order to make these scores comparable to the House and Senate common space scores, I regressed House common space scores on House DW scores and obtained the transformation equation, which I then applied to the presidential DW scores to move them into the common space. Presidential common space scores calculated by Poole and Rosenthal only go back to the Eisenhower administration. 47. Michael W. Giles, Virginia A. Hettinger, and Todd Peppers, “Measuring the Preferences of Federal Judges: A Common Space Alternative,” typescript, 2001. 48. Rather than seeing the floor medians as critical, some theories emphasize the importance of the majority party. See, e.g., Gary W. Cox and Mathew D. McCubbins, Setting the Agenda: Responsible Party Government in the U.S. House of Representatives (Cambridge: Cambridge University Press, 2004). Thus, in an alternative specification, presidential distance was measured as the distance from the president to the medians of the majority party in both chambers, averaged across them. Other theories emphasize the importance of veto players, and thus in two additional alternative specifications presidential distance was measured as the distance from the president to the floor median, and as the distance from the president to the majority party median, in the chamber furthest away from the president, operationalizing the notion that the most distant chamber’s preferences are decisive because if they are not satisfied, that chamber could exercise its veto powers. In these three alternative specifications the statistically and substantively significant variables remained statistically and substantively significant, and likewise the insignificant variables remained insignificant. More complex veto pivot models, such as Keith Krehbiel’s, do not yield predictions about whether Congress would include private enforcement regimes in regulatory legislation, and do not point to operationalizable strategies for identifying legislators pivotal to that decision for purposes of measuring their distance to the president. See Keith Krehbiel, Pivotal Politics: A Theory of U.S. Lawmaking (Chicago: University of Chicago Press, 1998). The literature focusing on the size of the gridlock interval is concerned with legislative productivity (ascertaining the range of possible movement away from the policy status quo, or the range of opportunity for new legislation). The goal of this model is quite different. Given movement in the status quo through the enactment of new regulatory legislation, this model seeks to explain a particular aspect of the content of the law—whether a specific implementation strategy is adopted. Further, while Krehbiel’s gridlock interval measures when it is possible to move the status quo and adopt a new policy, within the set of policies outside the gridlock interval that can be adopted, in order to know which pivotal players in Krehbiel’s game (filibuster, floor median, veto override, or president) will be most important in deciding which of the feasible policies is adopted, one has to know where the status quo is relative to those players. Krehbiel, Pivotal Politics, 35. In the vast majority of empirical settings, including the present one, it is not possible to identify the position of the status quo (no law) within the ideological space of Krehbiel’s game, and thus it is not possible to identify the legislator most powerful in shaping the law’s content for purposes of measuring distance to the president.
254
NOTES
49. Sarah A. Binder, Minority Rights, Majority Rule: Partisanship and the Development of Congress (New York: Cambridge University Press, 1997), 73. 50. E.g., Peter J. May, “Social Regulation,” in The Tools of Government: A Guide to the New Governance, ed. Lester M. Salamon (New York: Oxford University Press, 2002), 156–85; Thomas O. McGarity, “Regulatory Reform in the Reagan Era,” Maryland Law Review 45 (1986): 253–73; Peter H. Schuck, The Limits of Law: Essays on Democratic Governance (Boulder: Westview Press, 2000). 51. McGarity, “Regulatory Reform,” 254. 52. McGarity, “Regulatory Reform,” 255; Schuck, The Limits of Law, 123; Vogel, “New Social Regulation.” 53. This is not entirely surprising. Issue groups in general are a far larger force than organizations focused upon litigation. There were 1,017 issue organizations in the sample, not including litigation advocacy groups, and 66 litigation advocacy groups. Further, many issue organizations that are not litigation advocacy groups clearly are actively aware of and concerned with the role of courts in implementation, as shown, for example, by the frequency with which issue groups file amicus briefs. Thus, the line between issue groups and legal advocacy organizations, as it pertains to their likely views on private enforcement regimes, is not a bright one. 54. Kagan, Adversarial Legalism, 152; Vogel, “New Social Regulation.” In addition, this claim is central to the contention of the party alignment hypothesis that Republican opposition to civil legal liability is importantly determined by the preferences of the party’s business constituents. 55. David Mayhew uses this measure in Divided We Govern: Party Control, Lawmaking, and Investigations, 1946–1990 (New Haven: Yale University Press, 1991). 56. There may be circumstances in which one would want to distinguish between “on budget” and “off budget” spending, where legislators arguably perceive fewer repercussions for off-budget spending. Melnick, Between the Lines, 272. However, off-budget spending primarily concerns the allocation of money for spending programs and trust funds, not the day-to-day regulatory implementation activities of agencies and departments. Allen Schick and Felix LoStracco, The Federal Budget: Politics, Policy, Process (Washington, DC: Brookings Institution Press, 2000); Eric M. Patashnik, Putting Trust in the U.S. Budget: Federal Trust Funds and the Politics of Commitment (New York: Cambridge University Press, 2000). The distinction is thus not relevant to the hypothesis being tested here. 57. Alba Conte, Attorney Fee Awards, 3rd ed. (Saint Paul: Thomson/West, 2004), 1:667–701; Kerry D. Florio, “Attorneys’ Fees in Environmental Citizen Suits: Should Prevailing Defendants Recover?” Boston College Environmental Affairs Law Review 27 (2000): 707–40, 722–32; David Berger, “Court Awards of Attorney’s Fees: Litigating Antitrust, Civil Rights, Public Interest, and Securities Cases—Prevailing Party Concepts in Court Awards of Attorneys’ Fees,” Practicing Law Institute Litigation and Administrative Practice Course Handbook Series 324 (1987): 41–87, 77–82; Herbert B. Newberg, Attorney Fee Awards (Colorado Springs: Shepard’s/McGraw-Hill, 1986), 178–85; Richard E. Larson, Federal
NOTES
255
Court Awards of Attorney’s Fees (New York: Harcourt Brace Jovanovich, 1981), 85–97. 58. Colin A. Cameron and Pravin K. Trivedi, Regression Analysis of Count Data (New York: Cambridge University Press, 1998). 59. Cameron and Trivedi, Regression Analysis of Count Data; A. Tobias and M. J. Campbell, “Time Series Regression for Counts Allowing for Autocorrelation,” Stata Technical Bulletin Reprints, vol. 8 (College Station: Stata Press, 1998). 60. J. Schwartz, C. Spix, G. Touloumi, L. Bachárová, T. Barumamdzadeh, A. le Tertre, T. Piekarski, A. Ponce de Leon, A. Ponka, G. Rossi, M. Saez, and J. P. Schouten, “Methodological Issues in Studies of Air Pollution and Daily Counts of Deaths or Hospital Admissions,” Journal of Epidemiology and Community Health 50 (1996 Supplement 1): S3–S11. 61. K. Katsouyanni, J. Schwartz, C. Spix, G. Touloumi, D. Zmirou, A. Zanobetti, B. Wojtyniak, J. M. Vonk, A. Tobias, A. Ponka, S. Medina, L. Bacharova, and H. R. Anderson, “Short Term Effects of Air Pollution on Health: A European Approach Using Epidemiologic Time Series Data: The APHEA Protocol,” Journal of Epidemiology and Community Health 50 (1996 Supplement 1): S12–S18. 62. Tobias and Campbell, “Time Series Regression.” 63. Peter Kennedy, A Guide to Econometrics (Cambridge: MIT Press, 1994), 253–54. 64. Robert F. Engle and Clive W. Granger, “Co-integration and Error Correction: Representation, Estimation and Testing,” Econometrica 55 (1987): 251–76. 65. Damoder N. Gujarati, Basic Econometrics, 4th ed. (Boston: McGraw Hill, 2003), 822.
Introduction to Part II 1. Annual Report of the Administrative Office of the United States Courts, 1997–2007, table C-2. 2. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 5–6; Bureau of National Affairs, State Fair Employment Laws. 3. Bureau of National Affairs, State Fair Employment Laws, 237–38. 4. David Collier, James Mahoney, and Jason Seawright, “Claiming Too Much: Warnings about Selection Bias,” in Brady and Collier, Rethinking Social Inquiry, 85–102, 93. 5. Alexander George and Timothy McKeown, “Case Studies and Theories of Organizational Decision Making,” Advances in Information Processing in Organizations 2 (1985): 21–58, 35. 6. Robert H. Bates, Avner Greif, Margaret Levi, Jean-Laurent Rosenthal, and Barry R. Weingast, Analytic Narratives (Princeton: Princeton University Press, 1998), chap. 1. 7. Harry Eckstein, “Case Study and Theory in Political Science,” in Handbook of Political Science, ed. Fred I. Greenstein and Nelson W. Polsby (Reading: AddisonWesley, 1975), 7:79–137, 119; see also Stephen Van Evera, Guide to Methods for Students of Political Science (Ithaca: Cornell University Press, 1997), 79–80.
256
NOTES
8. Annual Report of the Administrative Office of the United States Courts, 1997–2007, table C-2. 9. See, e.g., Olson, Litigation Explosion; The Excuse Factory; Garry, A Nation of Adversaries; Johnston, “Litigation Explosion.” 10. Van Evera, Guide to Methods, 79; see also Robert K. Yin, Case Study Research: Design and Methods (Thousand Oaks: Sage, 1994), 40; Gerring, Case Study Research, 178–84. 11. Yin, Case Study Research, 41–44. 12. Eckstein, “Case Study and Theory,” 124; Van Evera, Guide to Methods, 82. 13. Van Evera, Guide to Methods, 86–87. 14. Annual Report of the Equal Employment Opportunity Commission, 1997–2007; Annual Report of the Administrative Office of the United States Courts, 1997–2007, table C-2. 15. Bell, “Killer Lawyers.” 16. Van Evera, Guide to Methods, 83. 17. Gerring, Case Study Research, 104. 18. Mary Frances Derfner, “One Giant Step: The Civil Rights Attorney’s Fees Awards Act of 1976,” Saint Louis University Law Journal 21 (1977): 441–51. 19. O’Connor and Epstein, “Bridging the Gap,” 239. 20. Burke, Lawyers, Lawsuits, 15–16; Kagan, Adversarial Legalism, 15; Melnick, “From Tax-and-Spend.” 21. Frymer, Black and Blue, 14–15, 76–94, 130–37. 22. Frymer, Black and Blue, 78. 23. Frymer, Black and Blue, 7, 14. 24. See, e.g., Heymann and Liebman, “No Fault, No Fee,” 309; Rau, Sherman, and Peppet, Process of Dispute Resolution, 56; White, “Legal Complexity,” 394–95; Gordon and Assefa, “Tale of Two Initiatives,” 693; Romano, “SarbanesOxley Act.”
Chapter 4 Foundations: The Civil Rights Act of 1964 1. See Eric Schickler, Disjointed Pluralism: Institutional Innovation and the Development of the U.S. Congress (Princeton: Princeton University Press, 2001), 13. 2. Robert A. Caro, Master of the Senate (New York: Alfred A. Knopf, 2002), 78–108; William N. Eskridge, “Reneging on History? Playing the Court/ Congress/President Civil Rights Game,” California Law Review 79 (1991): 613– 84, 619 and n. 27; Daniel Rodriguez and Barry R. Weingast, “The Positive Political Theory of Legislative History,” University of Pennsylvania Law Review 151 (2003): 1417–1542; Anthony S. Chen, The Fifth Freedom: Jobs, Politics, and Civil Rights in the United States, 1941–1972 (Princeton: Princeton University Press, 2009), chap. 5. 3. Charles Whalen and Barbara Whalen, The Longest Debate: A Legislative History of the 1964 Civil Rights Act (Washington, DC: Seven Locks Press, 1985), 4.
NOTES
257
4. Thomas Byrne Edsall and Mary D. Edsall, Chain Reaction: The Impact of Race, Rights, and Taxes on American Politics (New York: Norton, 1992), 36; James L. Sundquist, Politics and Policy: The Eisenhower, Kennedy, and Johnson Years (Washington, DC: Brookings Institution, 1968), 221–71; Denton L. Watson, Lion in the Lobby: Clarence Mitchell, Jr.’s Struggle for the Passage of Civil Rights Laws (New York: Morrow, 1990), 485–625; Whalen and Whalen, The Longest Debate, 13; Chen, The Fifth Freedom, chap. 5; Dean J. Kotlowski, Nixon’s Civil Rights: Politics, Principle, and Policy (Cambridge: Harvard University Press, 2001), 157–66. 5. Edward G. Carmines and James A. Stimson, Issue Evolution: Race and the Transformation of American Politics (Princeton: Princeton University Press), 46. 6. Edsall and Edsall, Chain Reaction, 36; Sundquist, Politics and Policy, 230. 7. Rodriguez and Weingast, “Positive Political Theory,” 1456. 8. I rely upon the ICPSR roll call dataset codes for party and state in the Eighty-eighth Congress to arrive at these numbers. 9. Chen, The Fifth Freedom, chap. 5; David B. Filvaroff and Raymond E. Wolfinger, “The Origin and Enactment of the Civil Rights Act of 1964,” in Legacies of the 1964 Civil Rights Act, ed. Bernard Grofman (Charlottesville: University Press of Virginia, 2000), 9–32, 14–18; Lieberman, Shaping Race Policy, 160–61; Rodriguez and Weingast, “Positive Political Theory,” 1481–87. 10. 71 Stat. 634, Public Law No. 85-315. 11. 74 Stat. 86, Public Law No. 86-449. 12. Daniel M. Berman, A Bill Becomes a Law: The Civil Rights Act of 1960 (New York: Macmillan, 1962); Caro, Master of the Senate, 910–1012; Sundquist, Politics and Policy, 222–50; Chen, The Fifth Freedom, chap. 5. 13. Morroe Berger, Equality by Statute: The Revolution in Civil Rights (Garden City, NY: Doubleday, 1967), 160; Jack Greenberg, Race Relations and American Law (New York: Columbia University Press, 1959), 109 n. 17, 309; Michael I. Sovern, Legal Restraints on Racial Discrimination in Employment (New York: Twentieth Century Fund, 1966), 79. 14. Greenberg, Race Relations, 271; Judith Stein, Running Steel, Running America: Race, Economic Policy, and the Decline of Liberalism (Chapel Hill: University of North Carolina Press, 1998), 84; Watson, Lion in the Lobby, 362–63; Dennis L. Wright, “State Legislative Response to the Federal Civil Rights Act: A Proposal,” Utah Law Review 9 (1964–65): 434–58, 449. 15. See, e.g., Vargo, “American Rule”; Leubsdorf, “Toward a History of the American Rule. 16. Michael A. Bamberger and Nathan Lewin, “The Right to Equal Treatment: Administrative Enforcement of Antidiscrimination Legislation,” Harvard Law Review 72 (1960–61): 526–89, 526; Greenberg, Race Relations, 16, 283; Arthur Earl Bonfield, “State Civil Rights Statutes: Some Proposals,” Iowa Law Review 49 (1963–64): 1067–1129, 1113–14. 17. Greenberg, Race Relations, 16; see also Stein, Running Steel, Running America, 84; Bonfield, “State Civil Rights Statutes,” 1113. 18. Jack Greenberg, correspondence with the author, August 22, 2007. 19. Greenberg, Race Relations, 22.
258
NOTES
20. Watson, Lion in the Lobby, 362–63; Greenberg, Race Relations, 138. 21. Bamberger and Lewin, “Right to Equal Treatment,” 526; Greenberg, Race Relations, 271; Stein, Running Steel, Running America, 84; R. Wayne Walker, “Title VII: Complaint and Enforcement Procedures and Relief and Remedies,” Boston College Industrial and Commercial Law Review 7 (1966): 495–524, 501. 22. Bonfield, “State Civil Rights Statutes,” 1113; Greenberg, Race Relations, 101. 23. Francis J. Vaas, “Title VII: Legislative History,” Boston College Industrial and Commercial Law Review 7 (1966): 431–58, 433–35. 24. House Report No. 570, 88th Cong., 1st sess., 1963. 25. H.R. 405, 88th Cong., 1st sess., 1963, reprinted in Hearings on Miscellaneous Proposals Regarding the Civil Rights of Persons Within the Jurisdiction of the United States Before Subcommittee No. 5 of the House Committee on the Judiciary, 88th Cong., 1st sess., 1963, 2328–36, §§ 8, 10; House Report No. 570, 2–3, 9–11. 26. H.R. 405, §§ 9–10, 13–14; House Report No. 570, 4–6, 12–13. 27. H.R. 405, § 11; House Report No. 570, 11. 28. H.R. 405, § 11; House Report No. 570, 17. 29. H.R. 405, § 10; House Report No. 570, 10–11. 30. The only role for trial courts (federal district courts) was that in the event the administrator found reasonable cause to credit the claim of discrimination, did not secure voluntary compliance, and failed to proceed with a hearing within a “reasonable time,” the complainant was empowered to petition a district court for an order directing the administrator to proceed with the hearing. H.R. 405, § 10(c). The court was not empowered to adjudicate the merits of the case, but rather only to direct the administrator to hold a hearing pursuant to the terms of H.R. 405. 31. Robert D. Loevy, To End All Segregation: The Politics of the Passage of the Civil Rights Act of 1964 (Lanham, MD: University Press of America, 1990), 56–57; Joseph Rauh, “The Role of the Leadership Conference on Civil Rights in the Civil Rights Struggle of 1963–64,” in The Civil Rights Act of 1964: The Passage of the Law That Ended Racial Segregation, ed. Robert D. Loevy (Albany: State University of New York Press, 1997), 49–75, 54–56; Whalen and Whalen, The Longest Debate, 16, 36–38, 46, 66. 32. House Report No. 187, 79th Cong., 1st sess., 1945; Hugh Davis Graham, The Civil Rights Era: Origins and Development of National Policy, 1960–1972 (New York: Oxford University Press, 1990), 132. 33. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 5–6; Bureau of National Affairs, State Fair Employment Laws. 34. Greenberg, Race Relations, 15–17, 114, 271; Bonfield, “State Civil Rights Statutes,” 1117; see also House Report No. 570, 5; House Report No. 238, 92nd Cong., 1st sess., 1971, 10–13; George P. Sape and Thomas J. Hart, “Title VII Reconsidered: The Equal Employment Opportunity Act of 1972,” George Washington Law Review 40 (1972): 824–89, 841. 35. Greenberg, Race Relations, 114, 154–207; Greenberg, correspondence with the author, August 22, 2007; Loevy, To End All Segregation, 331; Raymond
NOTES
259
Wolfinger, “American Business and the Civil Rights Bill,” typescript, University of California, Berkeley (1968), 5–6; Wright, “State Legislative Response,” 448–49. 36. House Report No. 570, 17. 37. House Report No. 570, 20. 38. House Report No. 570, 19. 39. House Report No. 570, 19; see also 15, 17. 40. House Report No. 570, 15, 17. 41. Melvyn Dubofsky, The State and Labor in Modern America (Chapel Hill: University of North Carolina Press, 1994), chap. 6; James A. Gross, The Reshaping of the National Labor Relations Board: National Labor Policy in Transition, 1937–1947 (Albany: State University of New York Press, 1981); Seymour Scher, “The Politics of Agency Organization,” Political Research Quarterly 15 (1962): 328–44; discussion of the Taft-Hartley Act in the conclusion to this book. 42. House Report No. 570, 15 (emphasis added). 43. House Report No. 1370, 87th Cong., 2nd sess., 1962. 44. Graham, The Civil Rights Era, 129–31. 45. House Report No. 570, 15. 46. Vaas, “Title VII: Legislative History,” 434. 47. Whalen and Whalen, The Longest Debate, 13. 48. Filvaroff and Wolfinger, “Origin and Enactment,” 15–16. 49. Vaas, “Title VII: Legislative History,” 434. 50. Loevy, To End All Segregation, 31, 53, 68–72; Filvaroff and Wolfinger, “Origin and Enactment,” 14; Whalen and Whalen, The Longest Debate, 26–27, 95. 51. Graham, The Civil Rights Era, 125–27; Vaas, “Title VII: Legislative History,” 35. 52. Whalen and Whalen, The Longest Debate, 42; Filvaroff and Wolfinger, “Origin and Enactment,” 19–20. 53. Loevy, To End All Segregation, 66–73; Whalen and Whalen, The Longest Debate, 40–58. 54. Filvaroff and Wolfinger, “Origin and Enactment,” 18; Loevy, To End All Segregation, 55, 68. 55. Filvaroff and Wolfinger, “Origin and Enactment,” 21; Graham, The Civil Rights Era, 129–32; Whalen and Whalen, The Longest Debate, 56–58, 63. 56. Whalen and Whalen, The Longest Debate, 58. 57. §§ 707(a) and (b) of H.R. 7152, as reported by the House Judiciary Committee, reprinted in House Report No. 914, 88th Cong., 1st sess., 1963, 12. 58. Lieberman, Shaping Race Policy, 160. 59. Clifford L. Alexander, “The EEOC Solution,” Washington Post, July 25, 1989, A11; Herbert Hill, “Lichtenstein’s Fictions: Meany, Reuther and the 1964 Civil Rights Act,” New Politics 7 (1999): 82–107, 92; Chen, The Fifth Freedom, chap. 5. 60. House Report No. 914, Part II, Additional Views of William M. McCulloch, John V. Lindsay, William T. Cahill, Garner E. Shriver, Clark MacGregor, Charles McC. Mathias, and James E. Bromwell, 150. 61. House Report No. 914, Minority Report, 68.
260
NOTES
62. House Report No. 914, Minority Report, 69. 63. House Report No. 914, 12, § 707(c). 64. House Report No. 914, 3, § 204(b). 65. House Report No. 914, 12, § 707(e). 66. Filvaroff and Wolfinger, “Origin and Enactment,” 22; Rodriguez and Weingast, “Positive Political Theory,” 1463. 67. Lieberman, Shaping Race Policy, 161; Rodriguez and Weingast, “Positive Political Theory,” 1474–96; John G. Stewart, “The Civil Rights Act of 1964: Tactics I,” in Loevy, Civil Rights Act of 1964, 211–73; Chen, The Fifth Freedom, chap. 5. 68. For a detailed accounting, see Vaas, “Title VII: Legislative History.” 69. Murray Kempton, “Dirksen Delivers the Souls,” New Republic, May 2, 1964, 9. 70. Wolfinger, “American Business,” 29; see also 19–20. 71. Wolfinger, “American Business,” 11–23; Filvaroff and Wolfinger, “Origin and Enactment,” 31 n. 11. The one significant change in coverage that they unsuccessfully sought was to eliminate gender as a protected classification. 72. Wolfinger, “American Business,” 24–25; Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 2012 (in the House-passed bill, section 707(a) allowed suit “on behalf of” an aggrieved person; section 707, providing for the filing of charges of discrimination, contained no statute of limitations; and section 707(c) required permission of one member of the commission to file a civil suit). 73. Congressional Record, 88th Cong., 2nd sess., March 26, 1964, 6449. 74. Congressional Record, 88th Cong., 2nd sess., March 26, 1964, 6449. 75. Graham, The Civil Rights Era, 146–47. 76. Stewart, “Civil Rights Act of 1964,” 256. 77. Comment, “Enforcement of Fair Employment under the Civil Rights Act of 1964,” University of Chicago Law Review 32 (1965): 430–70, 432–35; Vaas, “Title VII: Legislative History,” 450–52; Rodriguez and Weingast, “Positive Political Theory,” 1491–97. 78. John G. Stewart, legislative assistant to Hubert Humphrey in 1963–64, interview by the author, December 29, 2008. 79. Stewart, “Civil Rights Act of 1964,” 245–59; interview by the author, December 29, 2008. See also “Dirksen Influence the Key to Successful Cloture Vote,” Congressional Quarterly Weekly, May 15, 1964, 947–50. 80. Stewart, “Civil Rights Act of 1964,” 249. 81. Comment, “Enforcement of Fair Employment under the Civil Rights Act of 1964,” 450–51; Loevy, To End All Segregation, 258; Rodriguez and Weingast, “Positive Political Theory,” 1492; Stewart, “Civil Rights Act of 1964,” 211, 257–59. 82. 78 Stat. 261, § 707(a). 83. 78 Stat. 260, § 706(e); Stewart, “Civil Rights Act of 1964,” 252. 84. Richard J. Pierce Jr., Sidney A. Shapiro, and Paul R. Verkuil, Administrative Law and Process, 3rd ed. (New York: Foundation Press, 1999), 21, § 1.9; A. Michael Froomkin, “Reinventing the Government Corporation,” University of Illi-
NOTES
261
nois Law Review 1995 (1995): 543–634, 558; Elena Kagan, “Presidential Administration,” Harvard Law Review 114 (2001): 2245–2385, 2279; Lloyd N. Cutler and David R. Johnson, “Regulation and the Political Process,” Yale Law Journal 84 (1975): 1395–1418, 1402–9; see also Jerry L. Mashaw, “Agency-Centered or Court-Centered Administrative Law? A Dialogue with Richard Pierce on Agency Statutory Interpretation,” Administrative Law Review 59 (2007): 889–904, 891. 85. Graham, The Civil Rights Era, 146. See also Stein, Running Steel, Running America, 85. 86. Alfred M. Blumrosen, Modern Law: The Law Transmission System and Equal Opportunity (Madison: University of Wisconsin Press, 1993), 47. 87. John G. Stewart, legislative assistant to Hubert Humphrey in 1963–64, interview by the author, December 29, 2008. 88. Dirksen introduced omnibus amendments to the CRA of 1964 twice. He did so first on May 26, 1964, Congressional Record, 88th Cong., 2nd sess., May 26, 1964, 11926–35 (the Title VII provisions are at 11930–34), and a second time, with further revisions, on June 10, 1964, Congressional Record, 88th Cong., 2nd sess., June 10, 1964, 13310–19 (the Title VII provisions are at 13314–18). The proposed Title VII amendments were the same in both versions. The private right of action appears at Congressional Record, 88th Cong., 2nd sess., May 26, 1964, 11932–33, §706(e), and Congressional Record, 88th Cong., 2nd sess., June 10, 1964, 13316, §706(e). 89. Congressional Record, 88th Cong., 2nd sess., May 26, 1964, 11932–33, § 706(e) and (k), and Congressional Record, 88th Cong., 2nd sess., June 10, 1964, 13316, § 706(e) and (k). The precise language of the fee shift was that fees could be awarded to “prevailing parties.” In Newman v. Piggy Park Enters, Inc., 390 U.S. 400 (1968), the Supreme Court held that prevailing plaintiffs could recover fees under this language as a matter of course but that prevailing defendants could do so only in extraordinary circumstances. As discussed below, this reading of the CRA of 1964’s fee shift is consistent with legislative intent. 90. 78 Stat. 260–61, § 706(e) and (k). 91. See, e.g., Lassiter v. Dept. of Social Services, 452 U.S. 18 (1981); U. S. ex rel. Gardner v. Madden, 352 F.2d 792 (9th Cir. 1965). 92. O’Connor and Epstein, “Bridging the Gap.” 93. The Civil Rights Act of 1870, also known as the Force Act 1870, created voting rights protections aimed at combating terrorism orchestrated by the Ku Klux Klan against African American voters in the Reconstruction South, where states had proved unwilling or unable to control widespread violence against African Americans. The statute’s voting rights protections were enforceable through civil actions for monetary damages, along with attorney’s fees for prevailing plaintiffs. 16 Stat. 140–41, §§ 2–4. None of the other Reconstruction civil rights laws contained a plaintiffs’ fee-shifting provision. Most Reconstruction legislation governing voting rights, including the voting provisions of the CRA of 1870, were repealed in 1894 after Democrats retook control of Congress. Greenberg, Race Relations, 135; Bernard Schwartz, Statutory History of the United States, Civil Rights (New York: Chelsea House, 1970), part 1, p. 445. Fee shifting was also employed, for example, in such major legislation as the Interstate Commerce
262
NOTES
Act of 1887, the Sherman Antitrust Act of 1890, the Securities Exchange Act of 1934, and the Fair Labor Standards Act of 1938. It was hardly a new idea in 1964. 94. Title VII’s limitation to back pay was likely desired by civil rights advocates because additional monetary damages risked triggering a right to trial by jury for defendants under the Seventh Amendment, which civil rights advocates feared would seriously weaken the effects of Title VII in the South, where all-white juries, selected with racist jury procedures, would be unlikely to find liability. Blumrosen, Modern Law, 48, 367 n. 29. 95. O’Connor and Epstein, “Bridging the Gap,” 239–40; H.R. 7152, § 204(b), 88th Cong., 1st sess., 1963, reprinted in Hearings on Miscellaneous Proposals Regarding the Civil Rights of Persons Within the Jurisdiction of the United States Before Subcommittee No. 5 of the House Committee on the Judiciary, 88th Cong., 1st sess., 1963, Serial No. 4, Part I, 653. 96. Jack Greenberg, correspondence with the author, August 22, 2007. 97. Blumrosen, Modern Law, 48. 98. Charles F. Abernathy, Cases & Materials on Civil Rights & Constitutional Litigation, 4th ed. (Saint Paul: West Law School, 2006). 99. Sovern, Legal Restraints, 79–80; Greenberg, Race Relations, 16–17, 22, 100–114, 138, 270–83; Walker, “Title VII,” 501–2; Watson, Lion in the Lobby, 362–63. 100. David Filvaroff, special assistant in the Department of Justice in 1963–64, correspondence with the author, November 25, 2008. 101. Raymond Wolfinger, an aide to Senator Hubert Humphrey in 1963–64, interview by the author, November 12, 2008; John G. Stewart, legislative assistant to Hubert Humphrey in 1963–64, interview by the author, December 29, 2008. Wolfinger reports that Hubert Humphrey was an open conduit of communication between civil rights groups and Justice Department lawyers. According to Stewart, they were largely the same Justice Department lawyers that had drafted the original version of the Kennedy administration’s bill, which contained a public accommodations title with a private right of action and fee recovery for successful plaintiffs. See H.R. 7152, § 204(b), 88th Cong., 1st sess., 1963, reprinted in Hearings on Miscellaneous Proposals Regarding the Civil Rights of Persons Within the Jurisdiction of the United States Before Subcommittee No. 5 of the House Committee on the Judiciary, 88th Cong., 1st sess., 1963, Serial No. 4, Part I, 653. 102. John G. Stewart, legislative assistant to Hubert Humphrey in 1963–64, interview by the author, December 29, 2008. 103. Whalen and Whalen, The Longest Debate, 167–68; Loevy, To End All Segregation, 227–30; Watson, Lion in the Lobby, 598. 104. Watson, Lion in the Lobby. 105. Watson, Lion in the Lobby, 606, and chap. 18 generally; Raymond Wolfinger, an aide to Senator Hubert Humphrey in 1963–64, interview by the author, December 11, 2008. 106. Congressional Record, 88th Cong., 2nd sess., June 4, 1964, 12722. 107. Congressional Record, 88th Cong., 2nd sess., June 4, 1964, 12724. 108. Congressional Record, 88th Cong., 2nd sess., June 4, 1964, 13839.
NOTES
263
109. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 1011; Congressional Record, 88th Cong., 2nd sess., May 26, 1964, 11930–34, § 706(a) (Title VII as introduced by Dirksen the first time); Congressional Record, 88th Cong., 2nd sess., June 10, 1964, 13314–18, § 706(a) (Title VII as introduced by Dirksen the second time); 78 Stat. at 260, § 706(a) (Title VII as passed). 110. Graham, The Civil Rights Era, 146. 111. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 2012. 112. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 1012–13; Congressional Record, 88th Cong., 2nd sess., May 26, 1964, 11930–34, § 706(b) (Title VII as introduced by Dirksen the first time); Congressional Record, 88th Cong., 2nd sess., June 10, 1964, 13314–18, § 706(b) (Title VII as introduced by Dirksen the second time); 78 Stat. at 259–60, § 706(b) (Title VII as passed); Rodriguez and Weingast, “Positive Political Theory,” 1492. 113. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 2012–13. 114. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 1012–13; Congressional Record, 88th Cong., 2nd sess., May 26, 1964, 11930–34, § 706(d) (Title VII as introduced by Dirksen the first time); Congressional Record, 88th Cong., 2nd sess., June 10, 1964, 13314–18, § 706(d) (Title VII as introduced by Dirksen the second time); 78 Stat. at 259–60, § 706(d) (Title VII as passed). For a discussion of the unusual shortness of the limitations period of Title VII, as passed, see Deborah L. Brake, “The Failure of Title VII as a Rights-Claiming System,” North Carolina Law Review 86 (2008): 859–935, 866–70. 115. Equal Employment Opportunity Commission, Legislative History of Titles VII and XI, 2012–13. 116. Walker, “Title VII,” 495–524, 506. See also Comment, “Enforcement of Fair Employment under the Civil Rights Act of 1964,” 454, 467. 117. Vaas, “Title VII: Legislative History,” 454. See also 437. 118. Congressional Record, 88th Cong., 2nd sess., June 10, 1964, 12722. 119. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14188. 120. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14188. 121. Congressional Record, 88th Cong., 2nd sess., June 16, 1964, 13878. 122. Congressional Record, 88th Cong., 2nd sess., June 16, 1964, 13879 (ICPSR Roll Call No. 366). 123. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14201, 14213–14. 124. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14214. 125. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14213. 126. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14214 (ICPSR Roll Call No. 406). 127. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14182 (ICPSR Roll Call No. 396). Southerners had consistently advocated criminal rather than civil proceedings in the civil rights context on the expectation that the higher burden of proof, coupled with trial by all-white southern juries, would
264
NOTES
make convictions all but impossible. Greenberg, Race Relations, 15; Watson, Lion in the Lobby, 365. 128. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14196 (ICPSR Roll Call No. 401). 129. Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14201 (ICPSR Roll Call No. 405). 130. Congressional Record, 88th Cong., 2nd sess., June 12, 1964, 13647–48 (ICPSR Roll Call No. 330). As with the same amendment offered with respect to Title VII, Thurmond proposed, as an alternative to civil liability, the use of ordinary criminal process with associated jury trials and proof beyond a reasonable doubt. 131. Congressional Record, 88th Cong., 2nd sess., June 16, 1964, 13934 (ICPSR Roll Call No. 383). 132. Congressional Record, 88th Cong., 2nd sess., June 16, 1964, 13904–5 (ICPSR Roll Call No. 371). 133. Congressional Record, 88th Cong., 2nd sess., June 12, 1964, 13669 (ICPSR Roll Call No. 337). 134. Congressional Record, 88th Cong., 2nd sess., June 15, 1964, 13839–40 (ICPSR Roll Call No. 361). 135. “Intensive Lobbying Marked House Civil Rights Debate.” Congressional Quarterly Weekly, February 21, 1964, 364–66. 136. Hearings on Miscellaneous Civil Rights Legislation Before Subcommittee No. 5 of the House Judiciary Committee, 88th Cong., 1st sess., 1963. 137. Hearings on Equal Employment Opportunity Before the Subcommittee on Labor of the House Committee on Education and Labor, 88th Cong., 1st sess., 1963. 138. Richard Berg, “Equal Employment Opportunity under the Civil Rights Act of 1964,” Brooklyn Law Review 31 (1964): 62–97, 96–97; Comment, “Enforcement of Fair Employment under the Civil Rights Act of 1964,” 430; Graham, The Civil Rights Era, 189–90; John D. Skrentny, The Ironies of Affirmative Action: Politics, Culture, and Justice in America (Chicago: University of Chicago Press, 1996), 121–22; Stewart, “Civil Rights Act of 1964,” 249–52. 139. Berg, “Equal Employment Opportunity,” 96–97. 140. Stein, Running Steel, Running America, 85. 141. Sovern, Legal Restraints, 79. 142. 81 Stat. 602. 143. The employment provisions are contained in section 504 of the act, 87 Stat. 394. 144. The employment provisions are contained in Title I of the act, 104 Stat. 330. 145. Hearings on the bill had been held in the first session of the Nintieth Congress. See Hearings on S. 1358, S. 2114, and S. 2280 Before a Subcommittee on Housing and Urban Affairs of the Senate Committee on Banking and Currency, 90th Cong., 1st sess., 1967. The bill made it to the floor when offered as an amendment to a bill creating protections for civil rights workers against violence and intimidation (H.R. 2516). Congressional Record, 90th Cong., 2nd sess., February 6, 1968, 2270–73.
NOTES
265
146. Congressional Record, 90th Cong., 2nd sess., February 6, 1968, 2271. 147. Jean E. Dubofsky, “Fair Housing: A Legislative History and a Perspective,” Washburn Law Journal 8 (1969): 149–66, 156. 148. Congressional Record, 90th Cong., 2nd sess., February 20, 1968, 3427. 149. Dubofsky, “Fair Housing,” 156. 150. Dubofsky, “Fair Housing,” 156–57; Charles Lamb, “Congress, the Courts, and Civil Rights: The Fair Housing Act of 1968 Revisited,” Villanova Law Review 27 (1982): 1115–62, 1124–25; Leland B. Ware, “New Weapons for an Old Battle: The Enforcement Provisions of the 1988 Amendments to the Fair Housing Act,” Administrative Law Journal 7 (1993): 59–119, 73–74. 151. Congressional Record, 90th Cong., 2nd sess., February 29, 1968, 4670 (reprinting Washington Post, editorial of February 29, 1968). 152. Graham, The Civil Rights Era, 303. 153. Congressional Record, 90th Cong., 2nd sess., February 28, 1968, 4575. 154. Dubofsky, “Fair Housing,” 156–57. 155. Congressional Record, 90th Cong., 2nd sess., February 28, 1968, 4572. 156. Congressional Record, 90th Cong., 2nd sess., February 28, 1968, 4573. 157. Congressional Record, 90th Cong., 2nd sess., March 4, 1968, 4960. 158. Congressional Record, 90th Cong., 2nd sess., March 6, 1968, 5514. 159. Congressional Record, 90th Cong., 2nd sess., March 6, 1968, 5514. 160. 102 Stat. 1619, Public Law 100-430. 161. See, e.g., Samuel v. Benedict, 573 F.2d 580 (9th Cir. 1978); Crumble v. Blumthal, 549 F.2d 462 (7th Cir. 1977); Clemons v. Runck, 402 F.Supp. 863 (S.D. Oh. 1975). 162. It should be acknowledged, though, that the evidence is still compatible with the notion that budgetary considerations were relevant to some legislators. Concern about controlling government spending has often been expressed as a motivation for opposing growth in the bureaucratic state by conservative Republicans. In addition to the fact that they never mentioned the issue, considering a historical counterfactual also casts doubt on the significance of this factor. If one imagines that proponents of a robustly powerful agency credibly offered to staff it with an army of volunteer bureaucrats, to be led by an appointee of Kennedy or Johnson, it seems doubtful that antiregulation Republicans would have been any more disposed to support the proposal.
Chapter 5 Reverberations: 1965–1976 1. Paul Pierson, “The Study of Policy Development,” Journal of Policy History 17 (2005): 34–51; Robert Jervis, System Effects: Complexity in Political and Social Life (Princeton: Princeton University Press, 1997), chap. 4; Barbara Levitt and James G. March, “Organizational Learning,” Annual Review of Sociology 14 (1988): 319–40, 323. 2. Pierson, “Study of Policy Development,” 38–39; Jervis, System Effects, chap. 4; Howard, The Hidden Welfare State, 93–94.
266
NOTES
3. Paul Pierson, Dismantling the Welfare State? Reagan, Thatcher, and the Politics of Retrenchment (Cambridge: Cambridge University Press, 1994), 41–42; Heclo, Modern Social Politics, 305–16; Edwin Amenta, Elisabeth Clemens, Jefren Olsen, Sunita Parikh, and Theda Skocpol, “The Political Origins of Unemployment Insurance in Five American States,” Studies in American Political Development 2 (1987): 137–82; Jervis, System Effects, chap. 4. 4. Pierson, Dismantling the Welfare State? 40; Margaret Weir and Theda Skocpol, “State Structures and the Possibilities for ‘Keynesian’ Responses to the Great Depression in Sweden, Britain, and the United States,” in Evans, Rueschemeyer, and Skocpol, Bringing the State Back In, 107–63, 143–44; Skocpol, “Origins of Social Policy”; Patashnik, “After Public Interest Prevails.” 5. Hearings on Equal Employment Opportunity Enforcement Act (S. 2453) Before the Subcommittee on Labor of the Senate Committee on Labor and Public Welfare, 91st Cong., 1st sess., 1969 (hereinafter Senate Hearings on S. 2453), 50 (testimony of William Brown, chairman of the EEOC); Graham, The Civil Rights Era, 190, 203, 235–36, 239, 422. 6. James P. Gannon, “Uphill Bias Fight: After Faltering Start, Agency Readies Attack on Job Discrimination,” Wall Street Journal, April 12, 1967, 1. 7. Graham, The Civil Rights Era, 235; 422–23; Chen, The Fifth Freedom, chap. 5. 8. House Report No. 238, 92nd Cong., 1st sess, 1972. 9. Gannon, “Uphill Bias Fight,” 1. 10. Richard P. Nathan, Jobs and Civil Rights: The Role of the Federal Government in Promoting Equal Opportunity in Employment and Training (Washington, DC: U.S. Government Printing Office, 1969), 76. 11. Graham, The Civil Rights Era, 236–37, 422. 12. Nathan, Jobs and Civil Rights, 74–75. 13. Senate Hearings on S. 2453, 40. 14. Senate Hearings on S. 2453, 43 (statement of Senator Ralph Yarborough); 60 (statement of Clifford Alexander, member and former chairman of the EEOC); 66 (statement of Wendell Freeland, member of the board of trustees of the National Urban League); Graham, The Civil Rights Era, 424; Nathan, Jobs and Civil Rights, 47–50, 75; Sovern, Legal Restraints, 79; Chen, The Fifth Freedom, chap. 5. 15. Sape and Hart, “Title VII Reconsidered,” 830–31, 835–41; Chen, The Fifth Freedom, chap. 5. 16. Sape and Hart, “Title VII Reconsidered,” 830–45. It bears noting that, unlike in 1963–64, pivotal Republicans were not choosing between bureaucracy and private litigation; private litigation was now entrenched. They were choosing whether, in addition to a central role for private lawsuits, to have a bureaucratic prosecutor of court actions, or a bureaucratic adjudicator with the power to issue orders. 17. 86 Stat. 106-07, §706(f)(1), and §707(c)–(e). 18. Sape and Hart, “Title VII Reconsidered,” 841–45; Chen, The Fifth Freedom, chap. 5. 19. Senate Hearings on S. 2453, 72 (statement of Jack Greenberg, directorcounsel, NAACP Legal Defense and Education Fund); Hill, “Lichtenstein’s Fictions,” 91–92; Sape and Hart, “Title VII Reconsidered,” 832.
NOTES
267
20. Hill, “Lichtenstein’s Fictions,” 91–92; Frymer, Black and Blue. 21. Hill, “Lichtenstein’s Fictions,” 92. 22. Congressional Record, 90th Cong., 1st sess., February 20, 1967, 3949–52. 23. Hill, “Lichtenstein’s Fictions,” 91–92. 24. See, e.g., Senate Hearings on S. 2453, 72 (statement of Jack Greenberg, director-counsel, NAACP Legal Defense and Education Fund), 75 (statement of Joseph Rauh, general counsel, Leadership Conference on Civil Rights); House Report No. 238, 92nd Cong., 1st sess., 1971, 11–13. 25. Pierson, Dismantling the Welfare State? 41–42; Heclo, Modern Social Politics, 305–16; Amenta et al., “Origins of Unemployment Insurance.” 26. Nathan, Jobs and Civil Rights. 27. Lieberman, Shaping Race Policy, 190. 28. Nathan, Jobs and Civil Rights, 45–46. 29. Nathan, Jobs and Civil Rights, 46 (emphasis added). 30. House Report No. 238, 12. See also Senate Hearings on S. 2453, 72 (statement of Jack Greenberg, director-counsel, NAACP Legal Defense and Education Fund). 31. Skrentny, Ironies of Affirmative Action, 122–23; Nathan, Jobs and Civil Rights, 46. 32. Nathan, Jobs and Civil Rights, 46. 33. Jack Greenberg, correspondence with the author, August 22, 2007. 34. Jack Greenberg, correspondence with the author, August 22, 2007. 35. Alfred W. Blumrosen, Black Employment and the Law (New Brunswick: Rutgers University Press, 1971), 59. 36. Hearings on Equal Employment Opportunity Before the Subcommittee on Employment, Manpower, and Poverty of the Senate Committee on Labor and Public Welfare, 90th Cong, 1st sess., 1967 (hereinafter Senate Hearings on Equal Employment Opportunity, 4–5. 37. Nathan, Jobs and Civil Rights, 45–46, 76. 38. Senate Hearings on Equal Employment Opportunity, 127. 39. Senate Hearings on S. 2453, 70; see also 72. 40. Jack Greenberg, correspondence with the author, August 22, 2007. 41. Reg Murphy and Hal Gulliver, The Southern Strategy (New York: Scribner, 1971); Nadine Cohodas, Strom Thurmond and the Politics of Southern Change (New York: Simon and Schuster, 1993), 385–400; Kotlowski, Nixon’s Civil Rights, 15–43. 42. Kotlowski, Nixon’s Civil Rights, 166–76. 43. Roy Reed, “Job-Rights Chief Quits in Dispute on Nixon’s Goals,” New York Times, April 10, 1969, 1; Staff Reporter, “Government Aids Job Bias by Patronizing Noncompliant Firms, Rights Unit Charges,” Wall Street Journal, May 2, 1969, 4. 44. Reed, “Job-Rights Chief Quits.” 45. Staff Reporter, “Government Aids Job Bias”; Catherine LeRoy, Restoring the Conscience of a Nation: A Report on the U.S. Commission on Civil Rights (Leadership Conference on Civil Rights Education Fund, 2009) (http:// www.civilrights.org/publications/reports/commission/lccref_commission_report_ march2009.pdf), last checked June 30, 2009), 21.
268
NOTES
46. Staff Reporter, “Government Aids Job Bias”; “U.S. Civil Rights Unit Names Acting Staff Director,” Washington Post, September 8, 1968, A7. 47. Senate Hearings on S. 2453, 38–59 (testimony of EEOC chairman William Brown). 48. Senate Hearings on S. 2453, 211 (testimony of Thomas E. Harris, association general counsel, AFL-CIO) (emphasis added). 49. Kotlowski, Nixon’s Civil Rights, 170–71, and see generally 166–76; Rosemary Salomone, “Judicial Oversight of Agency Enforcement: The Adams and WEAL Litigation,” in Justice and School Systems: The Role of the Courts in Education Litigation, ed. Barbara Flicker (Philadelphia: Temple University Press, 1990), 111–81; Stephen C. Halpern, On the Limits of Law: The Ironic Legacy of Title VI of the 1964 Civil Rights Act (Baltimore: Johns Hopkins University Press, 1995), chap. 4; Christopher Bonastia, Knocking on the Door: The Federal Government’s Attempt to Desegregate the Suburbs (Princeton: Princeton University Press, 2006), 108–10, 115–18; Hearings on Enforcement of the Voting Rights Act Before the Civil Rights Oversight Subcommittee of the House Judiciary Committee, 92nd Cong., 1st sess., 1971 (hereinafter 1971 Hearings on Enforcement of the Voting Rights Act), 114 (statement of Joseph L. Rauh, Jr., counsel, Leadership Conference on Civil Rights). 50. 1971 Hearings on Enforcement of the Voting Rights Act, 1–2 (statement of Don Edwards (D-CA)); 114, 120 (statement of Joseph L. Rauh, Jr., counsel, Leadership Conference on Civil Rights). 51. 1971 Hearings on Enforcement of the Voting Rights Act, 3. 52. 1971 Hearings on Enforcement of the Voting Rights Act, 97, 100, 117, 122, 242 (statements of Howard A. Glickstein, staff director, U.S. Commission on Civil Rights; Representative Don Edwards (D-CA); Joseph L. Rauh, Jr., general counsel, Leadership Conference on Civil Rights; John Lewis, director, Voter Education Project). 53. 1971 Hearings on Enforcement of the Voting Rights Act, 114, 120. Later in his testimony Rauh stated that “[t]he reason most of the witnesses here appear political is that most of those who believe deeply in civil rights are now quite concerned at the course of the administration.” 207. 54. Kotlowski, Nixon’s Civil Rights, 27–30; Jonathan Spivak, “Nixon Statement on School Desegregation Stirs Doubts about His Civil Rights Stance,” Wall Street Journal, July 7, 1969, 6; Graham, The Civil Rights Era, 319–20. 55. Halpern, On the Limits of Law, chap. 4; Salomone, “Judicial Oversight of Agency Enforcement.” 56. Graham, The Civil Rights Era, 319. 57. Adams v. Richardson, 351 F.Supp. 636 (D.D.C. 1972) (case filed in 1970); Halpern, On the Limits of Law, chap. 4; Salomone, “Judicial Oversight of Agency Enforcement.” 58. Aberbach, Keeping a Watchful Eye, 27; conclusion of this book (discussing conflict over environmental policy). 59. 86 Stat. 369. The fee shift was codified at 20 U.S.C. § 1617. For a discussion of the winding legislative history of this fee provisions, see Bradley v. School Board of City of Richmond, 416 U.S. 696, 716 n. 23 (1974).
NOTES
269
60. Senate Report No. 61, 92nd Cong., 1st sess., 1971, 26; Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11339, 11345. 61. Senate Report No. 61, 25. 62. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11339; see also 11342. 63. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11345. 64. Kotlowski, Nixon’s Civil Rights, 173. 65. LeRoy, Restoring the Conscience, 21–22. 66. Hearings on the Emergency School Aid Act Before the Subcommittee on Education of the Senate Committee on Labor and Public Welfare, 92nd Cong, 1st sess., 1971 (hereinafter Senate Hearings on the Emergency School Aid Act), 456 (emphasis added). 67. Senate Hearings on the Emergency School Aid Act, 143. 68. Senate Hearings on the Emergency School Aid Act, 397, 405, 407. During committee hearings numerous other civil rights groups, including the ACLU and Common Cause, urged that fee recovery was vitally necessary in order to mobilize adequate enforcement of federal desegregation mandates. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11340. 69. Senate Report No. 61, 25. 70. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11343. 71. Senate Report No. 61, 25; Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11339. 72. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11345. 73. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11338–45. 74. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11343; April 22, 1971, 11521–29; April 23, 1971, 11724–26. 75. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11341–43; April 22, 1971, 11527–29; April 23, 1971, 11725–26. 76. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11345 (ICPSR Senate Roll Call No. 46) (deleting fee shift with public fund); April 23, 1971, 11726 (ICPSR Senate Roll Call No. 49) (inserting plaintiffs fee shift to be paid by losing defendant). 77. Senate Hearings on S. 2453, 69 (statement of Jack Greenberg, directorcounsel, NAACP Legal Defense and Education Fund). 78. Senate Hearings on S. 2453, 69. 79. Senate Hearings on S. 2453, 72. 80. Stein, Running Steel, Running America, 116; see also 113–20; Skrentny, Ironies of Affirmative Action, 161–66; Frymer, “Acting When Elected Officials Won’t.” 81. Eskridge, “Reneging on History?” 646–50; Frymer, “Acting When Elected Officials Won’t.” 82. 401 U.S. 424 (1971). 83. Robert Belton, “The Dismantling of the Griggs Disparate Impact Theory and the Future of Title VII: The Need for a Third Reconstruction,” Yale Law and Policy Review 8 (1990): 223–56; L. Camille Hébert, “Redefining the Burdens of Proof in Title VII Litigation: Will the Disparate Impact Theory Survive Wards
270
NOTES
Cove and the Civil Rights Act of 1990?” Boston College Law Review (1990): 1–94; Eskridge, Dynamic Statutory Interpretation, chap. 1. 84. Belton, “Dismantling of Griggs”; Hébert, “Redefining the Burdens”; Owen M. Fiss, “A Theory of Fair Employment Laws,” University of Chicago Law Review 38 (1971): 235–314; Eskridge, Dynamic Statutory Interpretation, chap. 1 and p. 304; Skrentny, Ironies of Affirmative Action, 166–71. 85. It bears noting here that the federal courts were not alone in supporting disparate impact theory in 1971. Nixon’s Justice Department and solicitor general, and the EEOC under a Nixon-appointed chair, all advocated before the Supreme Court in Griggs for recognizing disparate impact theory under Title VII. Skrentny, Ironies of Affirmative Action, 168–69; Eskridge, Dynamic Statutory Interpretation, 304. 86. Frymer, Black and Blue, 85. 87. George Rutherglen, “Title VII Class Actions,” University of Chicago Law Review 47 (1980): 688–741; Frymer, Black and Blue, 85. 88. Arthur R. Miller, “Of Frankenstein Monsters and Shining Knights: Myth, Reality, and the ‘Class Action Problem,’” Harvard Law Review 92 (1979): 664– 94, 674–75; “Developments in the Law: Class Actions,” Harvard Law Review 89 (1976): 1318–1644, 1353–54. 89. Frymer, Black and Blue, chap. 4. 90. Graham, The Civil Rights Era, 129–30, 430–31. 91. Shapiro, Who Guards the Guardians? 57. 92. Senate Hearings on S. 2453, 70. 93. Jack Greenberg, correspondence with the author, August 22, 2007; Bonfield, “State Civil Rights Statutes,” 1118. 94. Senate Hearings on S. 2453, 70–71. 95. Senate Hearings on S. 2453, 75. 96. Senate Hearings on S. 2453, 64–68 (statement of Wendell Freeland, member of the Board of Directors of the National Urban League), 77–86 (statement of Clarence Mitchell, director of the Washington Bureau of the NAACP). 97. Hill, “Lichtenstein’s Fictions,” 92. 98. Graham, The Civil Rights Era, 431. 99. Senate Hearings on S. 2453, 66–67 (statement of Wendell Freeland, member of the Board of Directors of the National Urban League), 69 (statement of Jack Greenberg, director-counsel, NAACP Legal Defense and Education Fund), 76 (statement of Joseph Rauh, general counsel, Leadership Conference on Civil Rights), 81–82 (statement of Clarence Mitchell, director of the Washington Bureau of the NAACP). 100. Jack Greenberg, correspondence with the author, August 22, 2007. 101. Rutherglen, “Title VII Class Actions,” 713 (emphasis added). 102. Sape and Hart, “Title VII Reconsidered,” 875–78. 103. 90 Stat. 2641, Public Law 95-559. 104. See, e.g., Sanders v. Dobbs Houses, Inc., 431 F.2d 1097 (5th Cir. 1970); Waters v. Wisconsin Steel Workers, 427 F.2d 476 (7th Cir. 1970). 105. Monell v. City of New York Department of Social Services, 436 U.S. 658 (1978).
NOTES
271
106. Harold S. Lewis, Jr. and Elizabeth J. Norman, Civil Rights Law and Practice (Saint Paul: West Group, 2001), 2–19; Employment Discrimination Law and Practice (Saint Paul: West Group, 2001), chap. 1.A, p. 5. 107. 42 U.S.C. § 1988. 108. Pierson, Dismantling the Welfare State? 40. See also Weir and Skocpol, “State Structures,” 143–44. 109. Armand Derfner, “Background and Origin of the Civil Rights Attorney’s Fee Awards Act of 1976,” 8UEDQ /DZ 37 (2005): 653–61, 656; O’Connor and Epstein, “Bridging the Gap,” 241. 110. Rutherglen, “Title VII Class Actions,” 688. 111. Hearings on Legal Fees Before the Subcommittee on Representation of Citizen Interests of the Senate Judiciary Committee, 93rd Cong., 1st sess., 1973 (hereinafter 1973 Hearings on Attorney’s Fees) (testimony of Armand Derfner), 1112–13. 112. O’Connor and Epstein, “Bridging the Gap,” 241. 113. Robert B. McKay, Nine for Equality under Law: Civil Rights Litigation. A Report to the Ford Foundation (New York: Ford Foundation, 1977); O’Connor and Epstein, “Bridging the Gap,” 240. 114. Derfner, “Background and Origin,” 656; Armand Defner, correspondence with the author, July 26, 2007. 115. These were the Age Discrimination in Employment Act of 1967, the Fair Housing Act of 1968, the Emergency School Aid Act of 1972, the Rehabilitation Act of 1973, and the Voting Rights Act Amendments of 1975. 116. Nathan, Jobs and Civil Rights, 74–75. 117. Graham, The Civil Rights Era, 422. 118. The figures for 1970 and after come from Federal Court Cases: Integrated Data Base, 1970–2000, maintained by the Inter-University Consortium for Political and Social Research. Cases coded 442 (“civil rights, employment”) in this dataset include suits filed under Title VII; the Age Discrimination in Employment Act of 1967; the Rehabilitation Act of 1973, which prohibited certain forms of disability discrimination in employment by the federal government or by federal contractors; and employment discrimination claims filed under the Civil Rights Acts of 1866 and 1871. A content analysis of a sample of cases coded “civil rights, employment” during this period shows that approximately 80 percent were Title VII claims. John J. Donohue III and Peter Siegelman, “The Changing Nature of Employment Discrimination Litigation,” Stanford Law Review 43 (1991) 983– 1033, 885 n. 3. This is how I estimated the figures presented. 119. It is possible that the delay was caused, in part, by uncertainty about the legal standard courts would use in deciding whether to award attorney’s fees because the statute stipulated that courts “may” make such an award. In 1968 the Supreme Court held unanimously that winning plaintiffs would automatically get fees unless special circumstances dictated otherwise. This bolstered plaintiffs’ lawyers’ confidence that they would be paid if they won civil rights cases under statutes with fee-shifting provisions. Derfner, “One Giant Step,” 442; Derfner, “Background and Origin,” 654. Amendments in 1966 to the class-action provisions in the Federal Rules of Civil Procedure, which increased judges’ discretion
272
NOTES
to recognize plaintiff classes, and liberal construction of those provisions by federal courts, likely also contributed to the increase, though individual (nonclass) claims have always constituted the overwhelming majority of Title VII claims. Rutherglen, “Title VII Class Actions”; Federal Court Cases: Integrated Data Base, 1970–2000, maintained by the Inter-University Consortium for Political and Social Research (containing data on the number of individual versus class claims). Finally, some of the increase in filings also can be attributed to the Equal Employment Act of 1972 (Public Law No. 92-261, 86 Stat. 103), which extended Title VII coverage to (1) employers with between fifteen and twenty-four employees (the original law only covered employers with twenty-five or more employees); (2) employees of federal and state governments, and (3) employees of educational institutions. It also increased Title VII’s statute of limitations from 90 to 180 days. However, these incremental increases in the covered workforce, and the limitations period window, could only account for a small share of the nearly tenfold increase in total Title VII filings between 1970 and 1975. 120. United States Census of the Population, 1980 Census, Chapter C, “General Social and Economic Characteristics,” table 61 in state-by-state listings (titled “Selected Social and Economic Characteristics by Race: 1980 and 1970”). The estimated figures for 1975 in the text are from linear interpolation between the 1970 and 1980 figures in table 61. Figures for nonwhites are obtained by subtracting figures for “whites” from total workforce figures. 121. Equal Employment Opportunity Commission Tenth Annual Report (Washington, DC: U.S. Government Printing Office, 1976), 55–56; Federal Court Cases: Integrated Data Base, 1970–2000, maintained by the Inter-University Consortium for Political and Social Research. The figure for the percentage of job discrimination lawsuits in the South in 1975 includes all claim types (race, gender, national origin, and religion). Federal data for job discrimination litigation for this period does not differentiate between claim types, and thus it is not possible to ascertain what percentage of federal job discrimination suits specifically based upon race were filed in the South. 122. Derfner, “One Giant Step,” 444–45. 123. Derfner, “One Giant Step,” 445. 124. 1973 Hearings on Attorney’s Fees, 1113. 125. McKay, Nine for Equality, 8, 13. 126. Bill Crider, “Civil Rights Turns to Gold Lode for Southern Lawyers,” Washington Post, April 4, 1976, 59. 127. Hearings on Awarding of Attorneys’ Fees Before the Subcommittee on Courts, Civil Liberties, and the Administration of Justice of the House Judiciary Committee, 94th Cong., 1st sess., 1975 (hereinafter 1975 Hearings on Attorney’s Fees), 136–37 (Bruce Terris, for-profit sector attorney). 128. 1975 Hearings on Attorney’s Fees, 155 (Henry Marsh, for-profit sector attorney). 129. Hearings on Extension of the Voting Rights Act of 1965 Before the Subcommittee on Constitutional Rights of the Senate Judiciary Committee, 94th Congress, 1st sess., 1975, 836 (Charles Morgan, Jr., director, Washington Office, ACLU).
NOTES
273
130. 1975 Hearings on Attorney’s Fees, 165 (Joseph Onek, director, Center of Law and Social Policy). 131. See, e.g., 1973 Hearings on Attorney’s Fees, 854–61, 1112–13 (Mary Frances Derfner and Armand Derfner, Lawyers’ Committee for Civil Rights Under Law); 1128–38 (Sylvia Roberts, chairperson, Legal Defense Fund, National Organization for Women); 1975 Hearings on Attorney’s Fees, 117–34 (Charles Halpern, executive director, Counsel for Public Interest Law); 160–62, 164–66 (Joseph Onek, director, Center of Law and Social Policy); Derfner, “Background and Origin,” 656; Hearings on Extension of the Voting Rights Act of 1965, 836 (Charles Morgan, Jr., director, Washington Office, ACLU). 132. See, e.g., 1973 Hearings on Attorney’s Fees, 788–806 (Anthony Kline, attorney for Public Advocates); 806–31 (Graeme Bell, attorney for Native American Rights Fund); 854–61, 1112–13 (Mary Frances Derfner and Armand Derfner, Lawyers’ Committee for Civil Rights Under Law); 1975 Hearings on Attorney’s Fees, 117–34 (Charles Halpern, executive director, Counsel for Public Interest Law); 81–101 (Mary Frances Derfner and Armand Derfner, Lawyers’ Committee for Civil Rights Under Law); 157–74 (Joseph Onek, director, Center of Law and Social Policy). 133. See, e.g., Bradshaw v. U.S. Dist. Court for Southern Dist. of California, 742 F.2d 515, 516 (9th Cir. 1984); Vymetalik v. E.E.O.C., 1987 WL 19005, 2 n. 3. (D.D.C. 1987); Johnson v. NCT Services, 631 F.Supp. 606, 607 n. 1; (D. Hi. 1986); Sol v. I.N.A. Ins. Co., 414 F.Supp. 29, 30 (E.D. Pa. 1976). 134. In re Nine Applicants, 475 F.Supp. 87 (N.D. Al.1979); Brooks v. Central Bank of Birmingham, 1982 WL 365 (N.D. Al. 1982). 135. See, e.g., Stanford Daily v. Zurcher, 366 F.Supp. 18 (D.C. Ca. 1973) (plaintiff represented by both nonprofit and for-profit lawyers successfully arguing for the availability of attorney’s fees under the CRA of 1871, drawing analogy to the fee provisions in the CRA of 1964); Lee v. Southern Home Sites Corp., 444 F.2d 143 (5th Cir. 1972) (plaintiff represented by both nonprofit and for-profit lawyers successfully arguing for the availability of attorney’s fees under the CRA of 1866, drawing analogy to the fee provisions in the CRA of 1964); Derfner, “One Giant Step.” 136. Stanford Daily v. Zurcher; Lee v. Southern Home Sites Corp.; Derfner, “One Giant Step.” 137. 421 U.S. 240 (1975). 138. 1973 Hearings on Attorney’s Fees; 1975 Hearings on Attorney’s Fees. 139. Derfner, “Background and Origin,” 653–58. Although the hearings cited in the preceding note considered attorney’s fee shifting in a number of different policy areas, civil rights was included among them, and the Senate Report on the Civil Rights Fees Act of 1976 states explicitly that it relied upon these hearings. Senate Report No. 1011, 94th Cong., 2nd sess., 1976, 2. 140. Berg, “Equal Employment Opportunity,” 96–97. 141. Graham, The Civil Rights Era, 424; Greenberg, Race Relations, 16, 101, 138, 271; Nathan, Jobs and Civil Rights, 47–50, 75; Sovern, Legal Restraints, 79; Stein, Running Steel, Running America, 85; Watson, Lion in the Lobby, 362–63; Senate Hearings on S. 2453, 43 (statement of Senator Ralph Yarborough), 60
274
NOTES
(statement of Clifford Alexander, member and former chairman of the EEOC), 66 (statement of Clarence Mitchell, director of the Washington Bureau of the NAACP). 142. Annual Report of the Administrative Office of the United States Courts, 1963, table C-2. 143. I make this statement based upon civil rights cases in the Westlaw database during this period. 144. Senate Report No. 1011, 4. 145. Congressional Record, 94th Cong., 2nd sess., September 21, 1976, 31471. 146. Congressional Record, 94th Cong., 2nd sess., September 29, 1976, 33313. 147. Congressional Record, 94th Cong., 2nd sess., September 29, 1976, 33314. 148. House Report No. 1558, 94th Cong., 2nd sess., 1976, 1. 149. Congressional Record, 94th Cong., 2nd sess., September 21, 1976, 31472. 150. Congressional Record, 94th Cong., 2nd sess., September 29, 1976, 33313. 151. Congressional Record, 94th Cong., 2nd sess., October 1, 1976, 35127. 152. Congressional Record, 94th Cong., 2nd sess., October 1, 1976, 35128. 153. Senate Report No. 1011, 94th Cong., 2nd sess., 1976, 4. 154. Congressional Record, 94th Cong., 2nd sess., September 21, 1976, 31471. 155. Congressional Record, 94th Cong., 2nd sess., September 21, 1976, 31479. 156. Congressional Record, 94th Cong., 2nd sess., September 21, 1976, 31472. 157. Congressional Record, 94th Cong., 2nd sess., September 29, 1976, 33314. 158. The Civil Rights Fees Act was introduced in the Senate on August 1, 1975. Subcommittee on Constitutional Rights, Senate Judiciary Committee, Civil Rights Attorney’s Fees Awards Act of 1976, Source Book: Legislative History, Texts, and Other Documents (Washington, DC: U.S. Government Printing Office, 1976) (hereinafter Legislative History of the Civil Rights Attorney’s Fees Awards Act of 1976), 3. The Voting Rights Act Amendments of 1975 were passed on August 6, 1975. Public Law No. 94-73, 89 Stat. 400. 159. 393 U.S. 544 (1969). Allen found an implied private right of action under section 5 of the Voting Rights Act of 1965, which requires federal authorization prior to implementation of new election rules in certain states (known as “preclearance” requirements). The Court did not make clear whether the implied right of action would apply to other sections of the act. 160. Public Law No. 94-73, 89 Stat. 400, Title III, §§ 401–2; Senate Report No. 295, 94th Cong., 1st sess., 1975, 39–43; House Report No. 196, 94th Cong., 1st sess., 1975, 33–34, 43. The 1975 amendments added a private right of action to section 3 of the Voting Rights Act of 1965, which authorizes courts in voting rights cases to direct appointment of federal examiners to register voters pursuant to section 6 of the act, to suspend tests or devices used as a precondition to voting, and to retain jurisdiction for as long as deemed necessary. 161. Congressional Record, 94th Cong., 1st sess., June 2, 1975, 16268. 162. House Report No. 196, 105–7. For the total number of Republicans on the committee, I rely upon Garrison Nelson, Committees in the U.S. Congress, 1947– 1992 (Washington, DC: Congressional Quarterly Press, 1993–94). The dissenting Republicans argued that the relief available under section 3 of the act, to which the private right of action was attached—judicial appointment of examiners, suspension of voting preconditions, and judicial retention of ongoing jurisdiction—
NOTES
275
was so invasive of state autonomy that it was constitutionally offensive to allow private litigants to obtain it. 163. Derfner, “Background and Origin,” 657. 164. Senate Report No. 295, 39–43; Congressional Record, 94th Cong., 1st sess., July 23, 1975, 24213 (statement of John V. Tunney (D-CA)). 165. Congressional Record, 94th Cong., 1st sess., July 23, 1975, 24213 (statement of John V. Tunney (D-CA)); Robert Malson, “In Response to Alyeska—the Civil Rights Attorney’s Fees Awards Act of 1976,” Saint Lewis University Law Journal 21 (1977) 430–40, 432. 166. Senate Report No. 295, 41; House Report No. 196, 34. 167. Hearings on Extension of the Voting Rights Act of 1965 Before the Subcommittee on Constitutional Rights of the Senate Judiciary Committee, 94th Congress, 1st sess., 1975 (hereinafter Senate Hearings on Extension of the Voting Rights Act of 1965), 132, 134, 137, 839. 168. 1971 Hearings on Enforcement of the Voting Rights Act, 97, 100, 117, 122, 242 (statements of Howard A. Glickstein, staff director, U.S. Commission on Civil Rights; Representative Don Edwards (D-CA); Joseph L. Rauh, Jr., general counsel, Leadership Conference on Civil Rights; John Lewis, director, Voter Education Project). 169. “U.S. Civil Rights Unit Names Acting Staff Director,” Washington Post, September 8, 1968, A7; Nomination of John A. Buggs To Be Staff Director of the Commission on Civil Rights, 92nd Cong., 2nd sess., 1972. 170. Senate Hearings on Extension of the Voting Rights Act of 1965, 219–20. 171. Senate Hearings on Extension of the Voting Rights Act of 1965, 220, 226. 172. The Senate hearings on the Voting Rights Amendments were held in April and May 1975. Senate Hearings on Extension of the Voting Rights Act of 1965. Tunney introduced the Civil Rights Fees Act in the Senate on August 1, 1975. Legislative History of the Civil Rights Attorney’s Fees Awards Act of 1976, 3. 173. Senate Hearings on Extension of the Voting Rights Act of 1965, 133. 174. Senate Hearings on Extension of the Voting Rights Act of 1965, 134 (Tunney statement); 78, 219–20, 1016–17, 1021 (U.S. Commission on Civil Rights recommendations). 175. Senate Hearings on Extension of the Voting Rights Act of 1965, 839. For other indications of Tunney’s lack of faith in the Nixon Justice Department, see also 128–29, 135, 138. 176. See, e.g., Congressional Record, 94th Cong., 1st sess., June 2, 1975, 16254 (statement of Representative Don Edwards (D-CA)); 16268–69 (statement of Representative Robert Drinan (D-MA)); Congressional Record, 94th Cong., 1st sess., July 22, 1975, 24143 (statement of Senator John V. Tunney). 177. Senate Hearings on Extension of the Voting Rights Act of 1965, 835–39; Roy Reed, “Charles Morgan Jr., 78, Dies; Leading Civil Rights Lawyer,” New York Times, January 10, 2009 (obituary). 178. Senate Hearings on Extension of the Voting Rights Act of 1965, 132–33 (statement of Frank R. Parker, assistant chief counsel, Mississippi Office, Lawyers’ Committee for Civil Rights Under Law); 134 (statement of Nicolas deB. Katzenbach, member of the Executive Committee of the Lawyers’ Committee
276
NOTES
for Civil Rights Under Law); Hearings on Extension of the Voting Rights Act Before the Subcommittee on Civil and Constitutional Rights of the House Judiciary Committee, 94th Cong., 1st sess., 1975, Part I (hereinafter House Hearings on Extension of the Voting Rights Act of 1965), 644 (statement of Armand Derfner, Lawyers’ Committee for Civil Rights Under Law). 179. Harper v. Kleindienst, 362 F.Supp. 742 (D.D.C. 1973); Harper v. Levi, 520 F.2d 53 (D.C. Cir. 1975). As early as 1971 Derfner had advocated attorney’s fees in voting rights cases. Hearings on Enforcement of the Voting Rights Act Before Subcommittee No. 4 of the House Judiciary Committee, 92nd Cong., 1st sess., 1971, 273. He also testified in favor of attorneys fees for civil rights cases in general in 1973 and 1975. 1973 Hearings on Attorney’s Fees, 1112–13; 1975 Hearings on Attorney’s Fees, 81–101. 180. Harper v. Kleindienst, 362 F.Supp. 742 (D.D.C. 1973); Harper v. Levi, 520 F.2d 53 (D.C. Cir. 1975). Though the D.C. Circuit’s decision in Harper was not appealed, the logic of the decision was later rejected by the Supreme Court in Morris v. Gressette, 432 U.S. 491 (1977). 181. Senate Hearings on Extension of the Voting Rights Act of 1965, 836, 838–39. 182. Legislative History of the Civil Rights Attorney’s Fees Awards Act of 1976, 3, 235. 183. Congressional Record, 94th Cong., 2nd sess., September 21, 1976, 31473–74. 184. Congressional Record, 94th Cong., 2nd sess., September 28, 1976, 32933–34. 185. Congressional Record, 94th Cong., 2nd sess., October 1, 1976, 35130 (ICPSR House Roll Call No. 1280); Congressional Record, 94th Cong., 2nd sess., September 29, 1976, 33315 (ICPSR Senate Roll Call No. 1298). 186. Congressional Record, 94th Cong., 2nd sess., September 22, 1976, 31835 (ICPSR Senate Roll Call No. 1248). 187. Congressional Record, 94th Cong., 2nd sess., September 24, 1976, 32396 (ICPSR Senate Roll Call No. 1273). 188. Congressional Record, 94th Cong., 2nd sess., September 28, 1976, 32934 (ICPSR Senate Roll Call No. 1289). 189. Congressional Record, 94th Cong., 2nd sess., September 22, 1976, 31858 (ICPSR Senate Roll Call No. 1251). 190. Congressional Record, 94th Cong., 2nd sess., September 27, 1976, 32394 (ICPSR Senate Roll Call No. 1271). 191. Congressional Record, 94th Cong., 2nd sess., September 27, 1976, 32388 (ICPSR Senate Roll Call No. 1270). 192. Congressional Record, 94th Cong., 2nd sess., September 28, 1976, 32927 (ICPSR Senate Roll Call No. 1282). 193. Congressional Record, 94th Cong., 2nd sess., September 27, 1976, 32397–98 (ICPSR Senate Roll Call No. 1274). 194. Congressional Record, 94th Cong., 2nd sess., September 28, 1976, 32391 (ICPSR Senate Roll Call No. 1285). 195. Congressional Record, 94th Cong., 2nd sess., September 28, 1976, 32933 (ICPSR Senate Roll Call No. 1287).
NOTES
277
196. Congressional Record, 94th Cong., 2nd sess., September 21, 1976, 31476 (ICPSR Senate Roll Call No. 1236). 197. Congressional Record, 94th Cong., 2nd sess., September 28, 1976, 32931 (ICPSR Senate Roll Call No. 1285) (Scott (R-VA) amendment to bar the award of attorney’s fees against federal or state governments under Titles II or VII of the CRA of 1964, or the CRA of 1965, as amended in 1975); Congressional Record, 94th Cong., 2nd sess., September 28, 1976, 32933 (ICPSR Senate Roll Call No. 1287) (Helms (R-NC) amendment to create a mandatory award of attorney’s fees to prevailing defendants under Titles II and VII of the CRA of 1964, the CRA of 1965, and the Reconstruction civil rights laws). 198. Congressional Record, 94th Cong., 2nd sess., October 1, 1976, 35130 (ICPSR House Roll Call No. 1280); Congressional Record, 94th Cong., 2nd sess., September 29, 1976, 33315 (ICPSR Senate Roll Call No. 1298). 199. Hearings on the Emergency School Aid Act Before the General Subcommittee on Education of the House Committee on Education and Labor, 92nd Cong., 1st sess., 1971; Senate Hearings on the Emergency School Aid Act. The correspondence from the ACLU, the National Legal Aid and Defender Association, and the law school deans was cited in the floor debates. Congressional Record, 92nd Cong., 1st sess., April 21, 1971, 11340. 200. House Hearings on Extension of the Voting Rights Act of 1965, Part I; Senate Hearings on Extension of the Voting Rights Act of 1965. The four appearances for legal advocacy organizations founded after passage of the CRA of 1964 were on behalf of the Mexican-American Legal Defense and Education Fund (founded in 1968), the Puerto Rican Legal Defense and Education Fund (founded in 1972), and the Center for Civil Rights at Notre Dame University (founded in 1973). 201. 1973 Hearings on Attorney’s Fees; 1975 Hearings on Attorney’s Fees. Although the hearings considered plaintiffs’ ability to recover attorney’s fees in a number of different policy areas, civil rights was primary among them, and the Senate report on the Civil Rights Fees Act of 1976 states explicitly that it relied upon these hearings. Senate Report No. 1011, 94th Cong., 2nd sess., 1976, 2. The hearings also addressed other issues concerning access to and affordability of legal services.
Chapter 6 Escalation: The Civil Rights Act of 1991 1. Olson, Litigation Explosion. 2. Warren E. Burger, “Too Many Lawyers, Too Many Suits; The Litigation Explosion: What Happened When America Unleashed the Lawsuit,” New York Times Book Review, May 12, 1991, 12; “Chief Justice Urges Greater Use of Arbitration,” New York Times, August 22, 1985, A21. 3. Eskridge, “Reneging on History?” 4. Herman Belz, Equality Transformed: A Quarter-Century of Affirmative Action (New Brunswick: Transaction Publishers, 1991), 181; Blumrosen, Modern Law, 268.
278
NOTES
5. Congressional Quarterly Almanac (Washington, DC: Congressional Quarterly News Features, 1980), 58B, 62B. 6. Citizens Commission on Civil Rights, One Nation, Indivisible: The Civil Rights Challenge for the 1990s, ed. Reginald C. Govan and William L. Taylor (Washington, DC: Citizens’ Commission on Civil Rights, 1989), 6. 7. Citizens Commission on Civil Rights, One Nation, Indivisible, 6–19. 8. Betty Friedan, The Second Stage (Cambridge: Harvard University Press, 1981), 237–38. 9. Affirmative Action: Joint Hearings Before the Subcommittee on Civil and Constitutional Rights of the House Judiciary Committee, and the Subcommittee of Employment Opportunities of the House Education and Labor Committee, 99th Cong., 1st sess., 1985, 273–74. 10. Robert R. Detlefsen, Civil Rights under Reagan (San Francisco: Institute for Contemporary Studies Press, 1991), 10. 11. Detlefsen, Civil Rights under Reagan, 10. 12. Blumrosen, Modern Law, 268; Detlefsen, Civil Rights under Reagan, 60– 82; David Rose, “Twenty Five Years Later: Where Do We Stand On Equal Employment Opportunity Law Enforcement?” Vanderbilt Law Review 42 (1989): 1121–82, 1153–55. 13. Detlefsen, Civil Rights under Reagan, 112–31, 156–64. 14. Rose, “Twenty Five Years Later,” 1158–59; Belz, Equality Transformed, 188–89; Blumrosen, Modern Law, 358 n. 28. 15. Belz, Equality Transformed, 189. Congress gave the EEOC the power to issue only procedural, but not substantive, rules. Magill, “Agency Choice,” 1387–88. The EEOC can, however, issue “interpretive guidelines” on substantive matters of law, and although they are not treated with the same deference as substantive rules promulgated pursuant to powers delegated by Congress, courts give some deference to them. Thomas W. Merrill and Kristin E. Hickman, “Chevron’s Domain,” Georgetown Law Journal 89 (2001): 833–921, 857. 16. Belz, Equality Transformed, 189. 17. Rose, “Twenty Five Years Later,” 1158–59. 18. Paul Burstein and Kathleen Monaghan, “Equal Employment Opportunity and the Mobilization of Law,” Law and Society Review 20 (1986): 355–84, 363–65; William Wines, “Title VII Interpretation and Enforcement in the Reagan Years (1980–89): The Winding Road to the Civil Rights Act of 1991,” Marquette Law Review 77 (1994): 645–718, 692; Dan B. Wood, “Does Politics Make a Difference at the EEOC?” American Journal of Political Science 34 (1990): 503–30. 19. Anne Noel Occhialino and Daniel Vail, “Why the EEOC (Still) Matters,” Hofstra Labor and Employment Law Journal 22 (2005): 671–709, 683. 20. Blumrosen, Modern Law, 271, table 17.1; Citizens Commission on Civil Rights, One Nation, Indivisible, 16; see also Wood, “Does Politics Make a Difference?” 21. Frank Dobbin and John R. Sutton, “The Strength of the Weak State: The Employment Rights Revolution and the Rise of Human Resource Management Divisions,” American Journal of Sociology 104 (1998): 441–76, 461; Erin Kelly and Frank Dobbin, “How Affirmative Action Became Diversity Management:
NOTES
279
Employer Response to Antidiscrimination Law, 1961–1996,” American Behavioral Scientist 41 (1998): 960–84, 967. 22. Belz, Equality Transformed, 193; Rose, “Twenty Five Years Later,” 1126; Skrentny, Ironies of Affirmative Action. 23. Blumrosen, Modern Law, 274. 24. Edsall and Edsall, Chain Reaction, 187–88. 25. Citizens Commission on Civil Rights, One Nation, Indivisible, 18. 26. David G. Savage, Turning Right: The Making of the Rehnquist Supreme Court (New York: Wiley, 1992); Sheldon Goldman, “Reagan’s Judicial Appointments at Mid-term: Shaping the Bench in His Own Image,” Judicature 66 (1983): 335–47; Wines, “Title VII Interpretation,” 646–50, 715; Eskridge, “Reneging on History?” 623–38. 27. Jefferson Decker, Lawyers for Reagan: The Conservative Litigation Movement and American Government, 1971-87, Ph.D. dissertation, Columbia University, 2009, 174–88. 28. Michael S. Greve, “Why Defunding the Left Failed,” Public Interest 89 (1987): 91–106, 91. 29. Decker, Lawyers for Reagan, 174–88; Greve, “Why Defunding the Left Failed,” 101–4. 30. Federal Court Cases: Integrated Data Base Series, 1970–2000, maintained by the Inter-University Consortium for Political and Social Research. 31. Annual Report of the Administrative Office of the United States Courts, 1977–1988, table C-2. 32. See the organizational mission description and founding date in the “interest of amicus curie” statement in the following NELA amicus briefs: Braitsch v. EMC Corporation, 1997 WL 33492086 (4th Cir. 1997); Weinstein v. The Board of Trustees of the CWA/ITU Negotiated Pension Plan, 1996 WL 33665449 (2nd Cir. 1996); Rhodes v. Guiberson Oil Tools Division, 1995 WL 17110046 (5th Cir. 1995). 33. Eskridge, “Reneging on History?” 633–41; Louis Fisher and Neal Devins, Political Dynamics of Constitutional Law, 4th ed. (Saint Paul: West Group, 2006), 270–77; Reginald C. Govan, “Honorable Compromises and the Moral High Ground: The Conflict between the Rhetoric and the Content of the Civil Rights Act of 1991,” Rutgers Law Review 46 (1993): 1–242, 7–16; Citizens Commission on Civil Rights, One Nation, Indivisible. 34. See (1) Oversight Hearings on the Federal Enforcement of Equal Employment Opportunity Laws Before the Subcommittee on Employment Opportunities of the House Committee on Education and Labor, 98th Cong., 1st sess., 1983; (2) Hearings on Equal Employment Opportunity Commission’s Handling of Pay Equity Cases Before the Subcommittee on Manpower and Housing of the House Committee on Government Operations, 98th Cong., 2nd sess., 1984; (3) Oversight Hearing on the EEOC’s Enforcement Policies Before the Subcommittee on Employment Opportunities of the House Committee on Education and Labor, 98th Cong., 2nd sess., 1984; (4) Hearing on Equal Employment Opportunity Commission Update: Policies on Pay Equity and Title VII Enforcement Before the Subcommittee on Employment and Housing of the House Committee on Government Operations, 99th Cong., 1st sess., 1985; (5) Oversight Hearing
280
NOTES
on the Equal Employment Opportunity Commission’s Enforcement Policies Before the Subcommittee on Employment Opportunities of the House Committee on Education and Labor, 99th Cong., 1st sess., 1985; (6) Hearing on the Equal Employment Opportunity Commission Collection of Federal Affirmative Action Goals and Timetables and Enforcement of Federal Sector EEO Complaints Before the Subcommittee on Employment Opportunities of the House Committee on Education and Labor, 99th Cong., 1st sess., 1985; (7) Oversight Hearing on EEOC’s Proposed Modification of Enforcement Regulations, Including Uniform Guidelines on Employee Selection Procedures Before the Subcommittee on Employment Opportunities of the House Committee on Education and Labor, 99th Cong., 1st sess., 1985; (8) Hearings on Equal Employment Opportunity Commission Policies Regarding Goals and Timetables in Litigation Remedies Before the Subcommittee on Employment Opportunities of the House Committee on Education and Labor, 99th Cong., 2nd sess., 1986; (9) Hearing on Processing of EEO Complaints in the Federal Sector: Problems and Solutions Before the Subcommittee on Employment and Housing of the House Committee on Government Operations, 100th Cong., 1st sess., 1987; (10) Hearing on Age Discrimination: Quality of Enforcement Before the House Select Committee on Aging, 100th Cong., 2nd sess., 1988; (11) Hearing on EEOC Delays in Processing Age Discrimination Charges Before the Subcommittee on Employment and Housing of the House Committee on Government Operations, 100th Cong., 2nd sess., 1988; (12) Hearing on the EEOC’s Performance in Enforcing the Age Discrimination in Employment Act Before the Senate Special Committee on Aging, 100th Cong., 2nd sess., 1988; (13) Hearing on EEOC’s Reprisal Against District Director for Testimony Before Congress on Age Discrimination Charges Before the Subcommittee on Employment and Housing of the House Committee on Government Operations, 101st Cong., 1st sess., 1989; (14) Joint Oversight Hearing on Equal Employment Opportunity Commission’s Proposed Reform of Federal Regulations Before the Subcommittee on Employment Opportunities of the House Committee on Education and Labor and the Subcommittee on Civil Service of the House Committee on Post Office and Civil Service, 101st Cong., 2nd sess., 1990; (15) Oversight Hearing on the Equal Employment Opportunity Commission’s Implementation of the American’s with Disabilities Act Before the Subcommittee on Employment Opportunities of the House Committee on Education and Labor, 102st Cong., 1st sess., 1991. 35. See note 34, Hearing No. 1, 1–3 (opening statement of Rep. Augustus Hawkins (D-CA); Hearing No. 5, 1–2 (opening statement of Rep. Matthew Martinez (D-CA)); Hearing No. 6, 3 (opening statement of Rep. Pat Williams (D-MT)); Hearing No. 7, 1–3 (opening statement of Rep. Matthew Martinez (D-CA)); Hearing No. 8, 1–3 (opening statement of Rep. Matthew Martinez (DCA)); Hearing No. 10, 1–2 (opening statement of Rep. Edward Roybal (D-CA)); see also A Report on the Investigation of Civil Rights Enforcement by the Equal Employment Opportunity Commission Based on a Study of Selected District Offices, Committee on Education and Labor, 99th Cong., 2nd sess., 1986. 36. Congressional Black Caucus Foundation, “In Opposition to Clarence Thomas: Where We Must Stand and Why,” in Court of Appeal: The Black Community Speaks Out on the Racial and Sexual Politics of Clarence Thomas vs.
NOTES
281
Anita Hill, ed. Robert Chrisman and Robert L. Allen (New York: Ballantine, 1992), 231–54, 234. 37. Congressional Black Caucus Foundation, “In Opposition to Thomas,” 234; see note 34, Hearing No. 9, 1–3 (opening statement of Rep. Tom Lantos (D-CA)); Hearing No. 11, 1–3 (opening statement of Rep. Tom Lantos (D-CA)); Hearing No. 12, 1–4 (opening statement of Senator John Melcher (D-MT)); Hearing No. 14, 1–3 (opening statement of Rep. Matthew Martinez (D-CA)). 38. A Report on the Investigation of Civil Rights Enforcement by theEqual Employment Opportunity Commission Based on a Study of Selected District Offices, Committee on Education and Labor, 99th Cong., 2nd sess., 1986, v–viii. 39. See note 34, Hearing No. 1, 1. 40. See note 34, Hearing No. 8, 2. In 1990 hearings Rep. Martinez characterized the EEOC as a “toothless kitten” responsible for “bankrupt” equal employment opportunity policy. See note 34, Hearing No. 14, p. 1. 41. See note 34, Hearing No. 9, p. 2. 42. Congressional Black Caucus Foundation, “In Opposition to Thomas,” 234. 43. Eskridge, “Reneging on History?” 633–41. 44. 490 U.S. 642 (1989). 45. 490 U.S. 228 (1989). 46. 490 U.S. 755 (1989). 47. 490 U.S. 900 (1989). 48. 491 U.S. 754 (1989). 49. 491 U.S. 164 (1989). 50. 499 U.S. 244 (1991). 51. 499 U.S. 83 (1991). 52. Joan Biskupic, “Groups Look to Capitol Hill for Help on Civil Rights,” Congressional Quarterly Weekly Report, June 17, 1989, 1479. 53. Julie Johnson, “High Court Called Threat to Blacks,” New York Times, July 10, 1989, A14, col. 1. 54. Biskupic, “Groups Look to Capitol Hill.” 55. Arthur S. Hayes, “Job-Bias Litigation Wilts under High Court Rulings,” Wall Street Journal, September 22, 1989; Johnson, “High Court Called Threat”; Charles Rothfeld, “Rulings on Job Bias: Chilling Effect on Lawsuits,” New York Times, October 27, 1989, B7, col. 3. 56. Hayes, “Job-Bias Litigation.” 57. Joan Biskupic, “Congress May Seek to Reverse Narrow Civil Rights Ruling,” Congressional Quarterly Weekly Report, June 10, 1989, 1404; Biskupic, “Groups Look to Capitol Hill.” 58. Govan, “Honorable Compromises.” 59. Bureau of National Affairs, “Omnibus Civil Rights Bill May Include Civil Rights Reform,” Daily Labor Report, December 18, 1989, A1, A9–A10; Govan, “Honorable Compromises,” 31–50. 60. Govan, “Honorable Compromises,” 31. 61. Reginald Govan, correspondence with the author, September 4, 2007; Govan, “Honorable Compromises,” 31–32 and n. 150. 62. Govan, “Honorable Compromises,” 32–50.
282
NOTES
63. 102 Stat. 1619; Hearings on S. 558 (Fair Housing Amendments Act) Before the Subcommittee on the Constitution of the Senate Judiciary Committee, 100th Cong., 1st sess., 1987; Michael Selmi, “Public vs. Private Enforcement of Civil Rights: The Case of Housing and Employment,” UCLA Law Review 45 (1997): 1401–59. 64. Catania, “State Employment Discrimination Remedies,” 807 n. 141; Greenberg, Race Relations, 15–16. 65. Mary Kathryn Lynch, “The Equal Employment Opportunity Commission: Comments on the Agency and Its Role in Employment Discrimination Law,” Georgia Journal of International and Comparative Law 20 (1990): 89–104, 103. 66. Govan, “Honorable Compromises,” 36. Numerous proposals were floated creating new damages without trial by jury. Govan, “Honorable Compromises,” 95, 102, 121, 173–74, 216. 67. Jean F. Rydstrom, “Applicability of 42 U.S.C.A. § 1981 to National Origin Employment Discrimination Cases,” American Law Reports Federal 43 (2007): 103, § 2[a]. 68. Govan, “Honorable Compromises,” 44, 57, 171–73, 181. 69. Alexander, “The EEOC Solution,” A11; Bureau of National Affairs, “Omnibus Civil Rights Bill,” A1, A9–A10; Leroy D. Clark, “Insuring Equal Opportunity in Employment through Law,” in Rethinking Employment Policy, ed. D. Lee Bauden and Felicity Skidmore (Washington, DC: Urban Institute Press, 1989), 173–203, 185–88; Govan, “Honorable Compromises,” 34–37, 47; Judicial Conference of the United States, Federal Courts Study Committee, Tentative Recommendations for Public Comment (1989), 49–51; Judicial Conference of the United States, Federal Courts Study Committee, Report of the Federal Courts Study Committee (1990), 60–62; Eskridge, “Reneging on History?” 681–83. 70. Senate Hearings on S. 2453, 59–61 (statement of Clifford Alexander, member and former chairman of the EEOC). 71. Alexander, “The EEOC Solution.” 72. Govan, “Honorable Compromises,” 34–35. 73. Richard Seymour, correspondence with the author, July 24, 2007. 74. Congressional Black Caucus Foundation, “In Opposition to Thomas,” 234; Mary C. Dunlap, “Are We Integrated Yet? Pursuing the Complex Question of Values, Demographics and Personalities,” University of San Francisco Law Review 29 (1995): 693–704, 695; Paul J. Spiegelman, “Remedies for Victim Group Isolation in the Workplace: Court Orders, Problem Solving and Affirmative Action in the Post-Stotts Era,” Howard Law Journal 29 (1986): 191–258, 199 and n. 35; Richard Seymour, correspondence with the author, July 24, 2007. 75. Richard Seymour, correspondence with the author, July 24, 2007 (emphasis added). 76. Joan Biskupic, “Congress May Seek to Reverse Narrow Civil Rights Ruling”; Joan Biskupic, “Bush Caught in Contradiction over Job-Bias Remedies,” Congressional Quarterly Weekly Report, April 21, 1990, 1196; Joan Biskupic, “New Struggle over Civil Rights Brings Shift in Strategy,” Congressional Quarterly Weekly Report, February 9, 1991, 366; Joan Biskupic, “Democrats Offer
NOTES
283
Compromise on Contentious Bias Bill,” Congressional Quarterly Weekly Report, May 18, 1991, 1286. 77. Janet Hook, “Dole Outburst Shows Frustration over More Than Civil Rights Bill,” Congressional Quarterly Weekly Report, July 21, 1990, 2314; Joan Biskupic and Christine Lawrence, “Civil Rights Bill Advancing Despite GOP Objections,” Congressional Quarterly Weekly Report, May 5, 1990, 1352; Joan Biskupic, “Partisan Rancor Marks Vote on Civil Rights Measure,” Congressional Quarterly Weekly Report, July 21, 1990, 2312; Joan Biskupic, “House Joins in the Standoff over Civil Rights Measure,” Congressional Quarterly Weekly Report, August 4, 1990, 2517. 78. Biskupic, “Bush Caught in Contradiction”; Biskupic, “Civil Rights Act Poised to Clear, but Bush Veto Looks Certain,” Congressional Quarterly Weekly Report, September 29, 1990, 3128; Biskupic, “New Struggle”; Biskupic, “Democrats Offer Compromise.” 79. Dinah Wisenberg, “House Panel Adds Its Stamp to Civil Rights Measure,” Congressional Quarterly Weekly Report, July 28, 1990, 2418; Joan Biskupic, “Expected Bush Veto Looming over Civil Rights Measure,” Congressional Quarterly Weekly Report, October 20, 1990, 2312; Biskupic, “Bush Caught in Contradiction”; Biskupic, “House Joins in Standoff.” 80. In debates over the CRA of 1964, see the comments of Senator Sam Ervin of North Carolina, Congressional Record, 88th Cong., 2nd sess., June 17, 1964, 14201, 14213–14. In the debates over the CRAFAA of 1976, see the comments of Senator James Allen of Alabama, Congressional Record, 94th Cong., 2nd sess., September 21, 1976, 16253–54. 81. Biskupic, “Bush Caught in Contradiction”; Biskupic, “New Struggle”; Joan Biskupic, “Deal on Civil Rights Stymied by ‘Quota’ Issue,” Congressional Quarterly Weekly Report, July 14, 1990, 2225; Biskupic, “Civil Rights Act Poised.” 82. Ronald D. Elving, “Successful Passage of Civil Rights Bill Could Be Political Plus for Both Parties,” Congressional Quarterly Weekly Report, September 2, 1991, 268; Joan Biskupic, “Another Round on ‘Quotas,’” Congressional Quarterly Weekly Report, June 29, 1991, 1760. 83. Biskupic, “House Joins in Standoff”; Biskupic, “Civil Rights Act Poised”; Govan, “Honorable Compromises,” 71, 103–17. 84. House Report No. 856, 101st Cong., 2nd sess., 1990; Joan Biskupic, “Conferees Attempt to Rescue Imperiled Civil Rights Bill,” Congressional Quarterly Weekly Report, October 13, 1990, 3428. 85. Elving, “Successful Passage”; Joan Biskupic, “Failure to Enact Civil Rights Bill Laid to Political Miscalculation,” Congressional Quarterly Weekly Report, October 27, 1990, 3610; Biskupic, “Bush Caught in Contradiction”; Govan, “Honorable Compromises,” 68, 71. 86. Biskupic, “Failure to Enact”; Govan, “Honorable Compromises,” 149–51. 87. Biskupic, “Failure to Enact”; Govan, “Honorable Compromises,” 153–54. 88. Biskupic, “Democrats Offer Compromise”; Biskupic, “New Struggle”; Biskupic, “Another Round on Quotas”; Joan Biskupic, “Danforth’s Persistence Paid Off in Thomas, Civil Rights Battles,” Congressional Quarterly Weekly Report, November 2, 1991, 3202; Govan, “Honorable Compromises,” 204–35.
284
NOTES
89. Biskupic, “New Struggle”; Joan Biskupic, “Supporters of Anti-Job-Bias Bill Need a Winning Strategy,” Congressional Quarterly Weekly Report, March 16, 1991, 683; Govan, “Honorable Compromises,” 171. 90. Pamela Fessler, Joan Biskupic, and Phil Kuntz, “Rights Bill Rises from the Ashes of Senate’s Thomas Fight,” Congressional Quarterly Weekly Report, October 26, 1991, 3124; Joan Biskupic, “Senate Passes Sweeping Measure to Overturn Court Ruling,” Congressional Quarterly Weekly Report, November 2, 1991, 3200; Bureau of National Affairs, “Thomas Hearings Illustrate Problems in Harassment Cases,” Daily Labor Report, October 28, 1991, C1; Govan, “Honorable Compromises,” 224–25. 91. Govan, “Honorable Compromises,” 178–79. 92. Ann Devroy, “Bush Saw Gains in Deal, Officials Say,” Washington Post, October 26, 1991, A1; Govan, “Honorable Compromises,” 145. 93. Fessler, Biskupic, and Kuntz, “Rights Bill Rises.” 94. Govan, “Honorable Compromises,” 144–45. 95. Govan, “Honorable Compromises,” 178–79. 96. Bureau of National Affairs, “White House Announces Civil Rights Compromise Ending Two-Year Long Dispute,” Daily Labor Report, October 28, 1991, A11; Biskupic, “Danforth’s Persistence”; Biskupic, “Senate Passes Sweeping Measure”; Fessler et al., “Rights Bill Rises from Ashes”; Govan, “Honorable Compromises,” 229. 97. Congressional Record, 101st Cong., 2nd sess., October 24, 1990, 16589. 98. Fessler, Biskupic, and Kuntz, “Rights Bill Rises.” 99. Govan, “Honorable Compromises,” 204–35. 100. 105 Stat. 1071, Public Law102-166. Price Waterhouse, Martin, Lorance, Boureslan, Casey, Wards Cove, and Patterson were all overruled. 101. The caps are $50,000 for employers with 15 to 100 employees, $100,000 for employers with 101 to 200 employees, $200,000 for employers with 201 to 300 employees, and $300,000 for employers with more than 300 employees. 42 U.S.C. § 1981a(b)(3). 102. 42 U.S.C. § 1981a(c). 103. Govan, “Honorable Compromises,” 229–38. 104. Congressional Record, 102nd Cong., 1st sess., October 30, 1991, 15503. 105. Congressional Record, 102nd Cong., 1st sess., November 7, 1991, 9557. 106. Govan, “Honorable Compromises,” 235. 107. House Report No. 644, 101st Cong., 2nd sess., 1990, 42, 46; see also 44–45; Senate Report No. 315, 101st Cong., 2nd sess., 1990, 33. 108. Senate Report No. 315, 30; see also House Report No. 644, 39 (in reference to the absence of compensatory damages for pain and suffering under existing law, the House report stated that “the inability of discrimination victims to be made whole for their losses discourages such victims from seeking to vindicate their civil rights,” and this “deficiency in Title VII’s remedial scheme hinders . . . [the goal] of encouraging citizens to act as private attorneys general to enforce the statute”). 109. House Report No. 644, 43 (emphasis added) (quoting Denny v. Westfield State College, 880 F.2d 1464, 1771–72 (1st Cir. 1989)).
NOTES
285
110. Judicial Conference of the United States, Federal Courts Study Committee, Tentative Recommendations for Public Comment (1989), 50; Judicial Conference of the United States, Federal Courts Study Committee, Working Papers And Subcommittee Reports (1990), working paper titled “Employment Discrimination Cases,” 28–29. Senator Edward Kennedy cited this study in his floor statement in favor of amending Title VII to increase economic rewards for enforcement. Congressional Record, 102nd Cong., 1st sess., October 29, 1991, 15341. 111. Federal Courts Study Committee, Working Papers And Subcommittee Reports, “Employment Discrimination Cases,” 28–29. 112. Federal Courts Study Committee, Working Papers And Subcommittee Reports, “Employment Discrimination Cases,” 29. 113. See, e.g., Sangster v. United Air Lines, Inc., 633 F.2d 864 (9th Cir. 1980); Sas v. Trintex, 709 F.Supp. 455 (S.D.N.Y. 1989); E.E.O.C. v. Nutri/System, Inc., 685 F.Supp. 568 (E.D. Va. 1988). 114. Terry, “Eliminating the Plaintiff’s Attorney.” 115. Senate Report No. 315, 33 (quoting Robinson v. Alabama State Department of Education, 727 F.Supp. 1422, 1430–32 (M.D. Ala. 1989)) (internal quotations omitted). 116. Senate Report No. 315, 33 (quoting Terry, “Eliminating the Plaintiff’s Attorney”) (internal quotations omitted). 117. House Report No. 644, 47. 118. Senate Report No. 315, 33 (quoting Robinson v. Alabama State Department of Education, 727 F.Supp. 1422, 1430–32 (M.D. Ala. 1989) (emphasis added)). Senator Edward Kennedy also quotes this language from the opinion of District Judge Myron H. Thompson of the Middle District of Alabama during the floor debates in the Senate. Congressional Record, 102nd Cong., 1st sess., October 29, 1991, 15345. 119. House Report No. 644, 42 (quoting Professor Charles Silver’s testimony during the committee hearings on the bill). 120. House Report No. 644, 42. 121. Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3888. 122. Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3851–52. 123. House Hearings on HR 4000, The Civil Rights Act of 1990, Joint Hearings before the Committee on Education and Labor and the Subcommittee on Civil Rights and Constitutional Rights of the Committee on the Judiciary, 101st Cong., 2nd sess., 1990 (hereinafter House Hearings on HR 4000), vol. 1, February 20, 1990, 383. 124. Senate Hearings on S. 2104, Civil Rights Act of 1990, Hearings before the Committee on Labor and Human Resources, 101st Cong., 2nd sess., 1990 (hereinafter Senate Hearings on S. 2104), March 1, 1990, p. 238; see also Representative Donald Payne (D-NJ), stating that “starting in the 1980s” discrimination was “tolerated . . . from the White House down.” House Hearings on HR 4000, vol. 3, May 21, 1990, 330. 125. See, e.g., House Hearings on HR 4000, vol. 1, February 20, 1990, 219 (testimony of John Buchanan, chairman, People for the American Way); vol. 2, March 20, 1990, 281–82, 286, 310 (testimony of Richard Arrington, Jr., mayor,
286
NOTES
City of Birmingham, Alabama); vol. 2, March 20, 1990, 298 (testimony of William Hundut, III, mayor, City of Indianapolis, Indiana); vol. 2, March 20, 1990, 347, 377 (testimony of Robert Joffee, an attorney who defended the consent decree in Martin). 126. House Hearings on HR 4000, vol. 2, March 13, 1990, 50. 127. Congressional Record, 101st Cong., 2nd sess., August 3, 1990, 6749. 128. Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3852. 129. General Accounting Office, Human Resources Department, “Equal Employment Opportunity: EEOC and State Agencies Did Not Fully Investigate Discrimination Charges,” GAO/HRD-89-11, October 11, 1988. 130. House Hearings on HR 4000, vol. 3, April 25, 1990, 62. 131. House Hearings on HR 4000, vol. 2, March 13, 1990, 63 (testimony of Carol Zabkowicz, of Racine, Wisconsin); vol. 3, May 21, 1990, 326–29 (testimony of Patricia Carroll, of Houston, Texas); vol. 3, May 21, 1990, 398 (testimony of Gerald Birnberg, partner, Williams, Birnberg, and Anderson, Houston, Texas). 132. House Hearings on HR 4000, vol. 1, February 20, 1990, 45. 133. House Hearings on HR 4000, vol. 3, May 21, 1990, 398 (testimony of Gerald Birnberg, partner, Williams, Birnberg, and Anderson, Houston, Texas). 134. House Hearings on HR 4000, vol. 1, February 27, 1990 638 (statement of Antonia Hernandez, President and General Counsel of the Mexican-American Legal Defense and Education Fund); Congressional Record, 102nd Cong., 1st sess., November 7, 1991, 9550 (Rep. Jose Serrano, D-NY). 135. House Hearings on HR 4000, vol. 3, April 25, 1990, 62. 136. House Hearings on HR 4000, vol. 2, March 13, 1990, 51. 137. Congressional Record, 101st Cong., 2nd sess., August 2, 1990, 6798. 138. See, e.g., Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3893 (Rep. Nancy Pelosi, D-CA); Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3896 (Rep. Barbara-Rose Collins, D-MI); Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3850 (Rep. Barbara Kennelly, D-CT); Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3888 (Rep. Major Owens, D-NY). 139. House Hearings on HR 4000, vol. 2, March 13, 1990, 68. 140. Congressional Record, 101st Cong., 2nd sess., August 2, 1990, 6791. 141. Congressional Record, 101st Cong., 2nd sess., August 3, 1990, 6749. 142. House Hearings on HR 4000, vol. 2, March 13, 1990, 172. 143. House Hearings on HR 4000, vol. 2, March 13, 1990, 51, 172. 144. House Hearings on HR 4000, vol. 3, April 25, 1990, 61–62. 145. House Hearings on HR 4000, vol. 2, March 13, 1990, 41, 74 (testimony of Ralph Baxter, Jr., Partner, Orrick, Herrington, Sutcliffe, San Francisco, California). 146. House Hearings on HR 4000, vol. 2, March 20, 1990, 447–48. 147. House Hearings on HR 4000, vol. 1, February 27, 1990, 681–85, 688, 716 (testimony of Lawrence Lorber, on behalf of the National Association of Manufacturers, Washington, DC). 148. Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3847 (Rep. James Sensenbrenner, R-WI); Senate Hearings on S. 2104, February 27, 1990, 103 (Senator Orrin Hatch, R, UT).
NOTES
287
149. House Hearings on HR 4000, vol. 1, February 20, 1990, 360–63 (testimony of Deputy Attorney General Donald Ayer). 150. House Hearings on HR 4000, vol. 2, 221–22 (testimony of David Maddux, the National Retail Federation); vol. 3, April 25, 1990, 16–17 (testimony of Edward Potter, the National Foundation for the Study of Equal Employment Policies). 151. House Hearings on HR 4000, vol. 2, March 13, 1990, 86, 100 (testimony of Victor Schachter, Partner, Schacter, Kristoff, et al., San Francisco, California); Senate Hearings on S. 2104, March 1, 1990, 294 (testimony of James Paras, Senior Partner, Morrison & Foerster, San Francisco, California), 346 (testimony of Cathie Shattuck, Partner, Epstein, Becker & Green, Washington, DC, and former commissioner of the EEOC). 152. House Hearings on HR 4000, vol. 1, February 27, 1990, 639. 153. Congressional Record, 102nd Cong., 1st sess., June 4, 1991, 3843. 154. House Hearings on HR 4000, vol. 2, March 13, 1990, 101, 105 (testimony of Victor Schachter, Partner, Schacter, Kristoff, et al., San Francisco, California). 155. House Hearings on HR 4000, vol. 2, March 13, 1990, 33 (testimony of Ralph Baxter, Jr., Partner, Orrick, Herrington, Sutcliffe, San Francisco, California). 156. House Hearings on HR 4000, vol. 1, February 20, 1990, 424 (testimony of Walter Oi, professor, Department of Economics, University of Rochester, Rochester, New York). 157. These votes were to adopt the Conference Report on S. 2104: 101st Cong., 2nd sess., October 16, 1990 (ICPSR Senate Roll Call No. 598); 101st Cong., 2nd sess., October 17, 1990 (ICPSR House Roll Call No. 831). 158. Southern Democrats voted in favor at 83 percent, as compared to 96 percent of nonsouthern Democrats; southern Republicans voted against at 90 percent, as compared to 77 percent of nonsouthern Republicans. Conference Report on S. 2104: 101st Cong., 2nd sess., October 16, 1990 (ICPSR Senate Roll Call No. 598); 101st Cong., 2nd sess., October 17, 1990 (ICPSR House Roll Call No. 831). 159. These votes were on passage of S. 1745: 102nd Cong., 1st sess., October 30, 1991 (ICPSR Senate Roll Call No. 248); 102nd Cong., 1st sess., November 7, 1991 (ICPSR House Roll Call No. 382). 160. Aggregating across both parties and both chambers, legislators from the eleven former Confederate states voted in favor of the CRA of 1991 at a rate of 88 percent. 102nd Cong., 1st sess., October 30, 1991 (ICPSR Senate Roll Call No. 248); 102nd Cong., 1st sess., November 7, 1991 (ICPSR House Roll Call No. 382). 161. See, e.g., House Hearings on H.R. 4000, vol. 1, 503–18, 546–54; vol. 2, 190–200, 244–50; Senate Hearings on S. 2104, 347–60, 1048–75. 162. See, e.g., House Hearings on H.R. 4000, vol. 2, 31–76; vol. 3, 111–80; Senate Hearings on S. 2104, 605–56. 1012–33. 163. Sean Farhang, “Congressional Mobilization of Private Litigants: Evidence from the Civil Rights Act of 1991,” Journal of Empirical Legal Studies 6 (2009): 1–34.
288
NOTES
164. The federal data to which I refer is Federal Court Cases: Integrated Data Base Series, 1970–2000, maintained by the Inter-University Consortium for Political and Social Research. Beginning in the mid-1990s, the Administrative Office of the United States Courts began recording the statutes under which job discrimination suits were filed, but with such data only beginning in 1995 the effects of the CRA of 1991 on actual litigation rates cannot be tested. 165. When enacted in 1990, the ADA incorporated by reference the enforcement and damages provisions of Title VII of the CRA of 1964, see 42 U.S.C. §12117(a), and thus when the ADA became effective in the summer of 1992, its provisions automatically incorporated the CRA of 1991’s changes to Title VII. Of course, it is impossible to assess how much of the litigation under the ADA can be attributed to the CRA of 1991, as compared to the litigation rate that would have obtained if Title VII’s (and thus the ADA’s) remedial provisions had not been amended. The presence of job discrimination claims as a result of the Age Discrimination in Employment Act, and CRAs of 1866 and 1871, also confound matters. 166. Annual Report of the Administrative Office of the United States Courts, 1977–2005, table C-2. 167. See http://www.eeoc.gov./stats/charges-a.html, and http://www.eeoc.gov./ stats/charges.html, last checked June 26, 2009. In the data cited, while disability charges aggregate together charges under the Rehabilitation Act of 1973 and the ADA of 1990, the overwhelming majority of the claims arise under the ADA. 168. 42 U.S.C. § 2000e-5(e)(1). 169. Robert H. Mnookin and Lewis Kornhauser, “Bargaining in the Shadow of the Law: The Case of Divorce,” Yale Law Journal 88 (1979): 950–77. 170. Farhang, “Congressional Mobilization of Private Litigants.” 171. E.g., Senate Hearings on S. 2104, March 1, 1990, 237–38 (comments of Senator Metzenbaum); House Hearings on H.R. 4000, vol. 1, February 20, 1990, 219 (John Buchanan, chairman, People for the American Way); vol. 3, May 21, 1990, 350, 353 (testimony of Ben Reyes, councilman, Houston City Council). 172. I rely upon the following sources for approximations of NELA membership numbers: Hearings on ERISA Enforcement Before the Subcommittee on Labor of the Senate Committee on Labor and Human Resources, 101st Cong., 2nd sess., 1990, 50 (statement of Jeffrey Lewis, representing the National Employment Lawyers Association) (1990 membership); Hearings on the Nomination of Judge Clarence Thomas to be Associate Justice of the Supreme Court of the U.S. Before the Senate Judiciary Committee, 102nd Cong., 1st sess., 1991, 1021 (statement of Nadia B. Axford, representing the National Employment Lawyers Association) (1991 membership); statements of “interest of amicus curie” in NELA amicus briefs in Cruz v. Local Union Number 3 of the International Brotherhood of Electrical Workers, 1993 WL 13011820 (2nd Cir. 1993) (1993 membership); Rhodes v. Guiberson Oil Tools Division, 1995 WL 17110046 (5th Cir. 1995) (1995 membership); Braitsch v. EMC Corporation, 1997 WL 33492086 (4th Cir. 1997) (1997 membership); Marianne Engelman Lado, “Unfinished Agenda: The Need for Civil Rights Litigation to Address Race Discrimination and Inequalities in Health Care Delivery,” Texas Forum on Civil Liberties and Civil Rights 6 (2001): 1–45, 26 n. 134 (1999 membership, which was said to be 3,400). Start-
NOTES
289
ing in about the year 2000 NELA’s statements of amicus curie simply state that the organization has “more than 3000” members, such as in Rogers v. New York University, 2000 WL 34002295 (2nd Cir. 2000), and Montesano v. Xerox Retirement Income Guarantee Plan, 2001 WL 34113125 (2nd Cir. 2001). The employment provisions of the Americans with Disabilities Act, which became effective in July 1992, by providing new work in the employment discrimination area, likely also contributed to NELA’s growth. However, among discrimination charges handled by the EEOC, disability charges only represented about 19 percent between 1992 and 2000 (http://www.eeoc.gov/stats/charges.html, and http://www.eeoc .gov/stats/charges-a.html, last checked June 26, 2009), and thus can only explain a small part of NELA’s more than 200 percent growth during about the same period. 173. Reiss, “Selecting Strategies.” 174. Kalev and Dobbin, “Enforcement of Civil Rights Law.” 175. Farhang, “Private Lawsuits.” 176. Epp, Making Rights Real. 177. Pierson, “Study of Policy Development,” 43. 178. As I suggested in the discussion of the changes made in the Senate, if one were to look only at those Republican amendments, the case is stronger that Republicans created a much more interventionist enforcement framework than would have occurred if the Senate had just accepted the House-passed version of Title VII. This is because cease-and-desist was already gone by the time the bill reached the Senate, and all Dirksen extracted in exchange for the private enforcement regime that he gave was to move the governmental right to sue from the EEOC to the Justice Department, and to limit the right to initiate suit to pattern-or-practice claims. The right to sue, in both individual and pattern-orpractice claims, was restored to the EEOC in the Equal Employment Act of 1972, so antiregulatory Republicans were unable to retain the regulation-weakening benefits that they had received in exchange for the private enforcement regime that Dirksen created. However, it must be acknowledged as well that the move for cease-and-desist between 1965 and 1972, which nearly succeeded, may well have been stronger in the absence of private litigation to shoulder the enforcement burden. Thus, assessing the ultimate effects of Dirksen’s enforcement bargain is a quite speculative endeavor.
Chapter 7. Conclusion: Private Enforcement Regimes and Themes in American Politics 1. E.g., Stein, Running Steel, Running America, 84–85; Chen, The Fifth Freedom, chap. 5; Skrentny, Ironies of Affirmative Action, 120–25; Rodriguez and Weingast, “Positive Political Theory,” 1489–96. Two recent works deviate from this trend. Robert Lieberman suggests that the EEOC’s hand was actually strengthened by lack of formal powers, which led it to forge effective alliances with civil rights groups, and Paul Frymer emphasizes that private litigation, with fee shifting, proved to be a potent weapon wielded against discriminatory prac-
290
NOTES
tices by labor unions by imposing large financial costs on them. Lieberman, Shaping Race Policy; Frymer, Black and Blue. 2. On the topic of private enforcement regimes in the CRA of 1964 as an influential model that spread to other policy areas, in addition to chapter 5 of this book, see Schudson, The Good Citizen, chap. 6; Yeazell, “Brown”; Burke, Lawyers, Lawsuits, 9. On the topic of union advocates arguing for a private right of action for workers under the NLRA, see Cynthia L. Estlund, “The Ossification of American Labor Law,” Columbia Law Review, 102 (2002): 1527–1612, 1551–59; James J. Brudney, “Isolated and Politicized: The NLRB’s Uncertain Future,” Comparative Labor Law and Policy Journal, 26 (2005): 221–60, 232–34; Benjamin I. Sachs, “Employment Law and Labor Law,” Cardozo Law Review, 29 (2008): 2685–2748, 2690, 2696–97. 3. Epstein and O’Halloran, Delegating Powers; Hamm and Robertson, “Factors Influencing Adoption”; see also Huber and Shipan, Deliberate Discretion? 4. Sean Farhang, “Legislative-Executive Conflict and Private Statutory Litigation: Evidence from Labor, Civil Rights, and Environmental Law,” Goldman School of Public Policy Working Paper (2010). 5. National Labor Relations Board, Legislative History of the Labor Management Relations Act, 1947 (Washington, DC: U.S. Government Printing Office, 1948); Dubofsky, State and Labor; Farhang and Katznelson, “The Southern Imposition”; American Bar Association, Committee on Development of the Law Under National Labor Relations Act, The Developing Labor Law: The Board, the Courts, and the National Labor Relations Act, ed. Patrick Hardin, John E. Higgins, Jr., Christopher T. Hexter, and John T. Neighbours, 4th ed. (Washington, DC: Bureau of National Affairs, 2002), 29–36; Winthrop A. Johns, “Picketing and Secondary Boycotts under the Taft-Hartley Act,” Labor Law Journal 2 (1951): 257–69; Sidney Sherman, “Recognition and Jurisdictional Strikes under Taft-Hartley,” Labor Law Journal 6 (1955): 211–14; continued on 267–71. 6. Dubofsky, State and Labor, 151–61; Gross, Reshaping of National Labor Relations Board; Scher, “Politics of Agency Organization.” 7. Legislative History of the LMRA, 297, 317. 8. Legislative History of the LMRA, 408. 9. Legislative History of the LMRA, 1655. 10. Legislative History of the LMRA, 1201. 11. Legislative History of the LMRA, 1350–55, 1369 (Senator Ball (R-MN)); 1353 (Senator Wherry (R-NE); 1353–54 (Senator Donnell (R-MO)); 460–61 (Supplemental Views in Senate Report). 12. Harry A. Millis and Emily Clark Brown, From the Wagner Act to TaftHartley: A Study of National Labor Relations (Chicago: University of Chicago Press, 1950), 350; Hearings Before the House Committee on Labor on Proposed Amendments to the National Labor Relations Act, 76th Cong., 1st Sess., 1939. For a summary of the bills examined in the cited hearings, see especially Report of the National Labor Relations Board on Proposed Amendments to the National Labor Relations Act, contained in the Supplement to the record of the hearings. 13. Smith, “Congress Opens Courthouse Doors”; Barton H. Thompson, Jr., “The Continuing Innovation of Citizen Enforcement,” University of Illinois Law
NOTES
291
Review 185 (2000): 185–236; Andreen, “Evolving Law”; David T. Buente, “Citizen Suits and the Clean Air Act Amendments of 1990: Closing the Enforcement Loop,” Environmental Law 21 (1991): 2233–52; Stephen Fotes, “Private Enforcement of the Clean Air Act and the Clean Water Act,” American University Law Review 35 (1985): 127–73. 14. Andreen, “Evolving Law,” 98; Robert F. Blomquist, “Rethinking the Citizen as Prosecutor Model of Environmental Enforcement under the Clean Water Act: Some Overlooked Problems of Outcome-Dependent Values,” Georgia Law Review 22 (1988): 337–423, 366–67; Stephen G. Allen, “Environmental Bounty Hunters: Reallocating Enforcement Authority Between the Citizens and the Government under the Clean Water Act,” Journal of Mineral Law and Policy 5 (1989– 90): 257–303, 267. 15. Senate Report No. 1196, 91st Cong., 2nd sess., 1970, 3, 21, 36–37; House Report No. 1146, 91st Cong., 2nd sess., 1970, 5; see also Blomquist, “Rethinking the Citizen,” 367; Allen, “Environmental Bounty Hunters,” 268; Frank Cross, “Rethinking Environmental Citizen Suits,” Temple Environmental Law and Technology Journal 8 (1989): 55–76, 56. 16. Cross, “Rethinking Environmental Citizen Suits,” 56. 17. Andreen, “Evolving Law,” 102 (summarizing the literature making this critique); Farhang, “Legislative-Executive Conflict.” 18. Joel A. Mintz, Enforcement at the EPA: High Stakes and Hard Choices (Austin: University of Texas Press, 1995), 54–58, 78–81, 97–100; Farhang, “Legislative-Executive Conflict”; see also Henry C. Kenski and Margaret Corgan Kenski, “Congress against the President: The Struggle over the Environment,” in Environmental Policy in the 1980s: Reagan’s New Agenda (Washington, DC: Congressional Quarterly Press, 1984), 97–120, 100–101, 104; Sidney A. Shapiro and Robert L. Glicksman, “Congress, the Supreme Court, and the Quiet Revolution in Administrative Law,” Duke Law Journal (1988): 819–79, 824–41. 19. Senate Committee on Environment and Public Works, Legislative History of the Clean Air Act Amendments of 1990 (Washington, DC: U.S. Government Printing Office, 1993) (hereinafter Legislative History of the CAAA of 1990), 6446 (partly quoting, and endorsing, a federal court opinion. 20. Legislative History of the CAAA of 1990, 2776. 21. Legislative History of the CAAA of 1990, 1308 (emphasis added). 22. Henry Waxman, “An Overview of the Clean Air Act Amendments of 1990,” Environmental Law 21 (1991): 1721–1816, 1726. 23. Legislative History of the CAAA of 1990, 2570; Andreen, “Evolving Law,” 102. 24. Cameron, Veto Bargaining, 10–11; Poole and Rosenthal, Congress; James, Presidents, Parties, and the State. 25. Cameron, Veto Bargaining, 259. 26. Gary Jacobson, “Partisan Polarization in Presidential Support: The Electoral Connection,” Congress and the Presidency 30 (2003): 1–36. 27. Nolan McCarty, Keith T. Poole, and Howard Rosenthal, Polarized America: The Dance of Ideology and Unequal Riches (Cambridge: MIT Press, 2006). 28. Moe, “Positive Theory,” 469 (emphasis added).
292
NOTES
29. Moe, “Positive Theory,” 469. 30. Edward Rubin, “Legislative Methodology: Some Lessons from the Truthin-Lending Act,” Georgetown Law Journal 80 (1991): 233–307. 31. Melnick, “From Tax-and-Spend.” 32. Sidney Verba, “American Exceptionalism: A Double-Edged Sword” (book review), American Political Science Review 91 (1997): 92–93, 92. 33. Lipset, American Exceptionalism; Anthony King, “Ideas, Institutions and the Policies of Government,” British Journal of Political Science 3 (1973): 291–314. 34. Lipset, American Exceptionalism, 270. 35. Glendon, Rights Talk; Schudson, The Good Citizen, chap. 6. 36. Kagan, Adversarial Legalism, 6–9 (reviewing literature); Wollschlager, “Exploring Global Landscapes,” 587–88. 37. For criticisms of cultural explanations, grounded in American exceptionalism, for policy outcomes in the United States, see Steinmo, “American Exceptionalism Reconsidered”; Theda Skocpol, Social Policy in the United States: Future Possibilities in Historical Perspective (Princeton: Princeton University Press, 1995). 38. Wilson, Bureaucracy, chap. 16. 39. See Moe, “Political Institutions”; Moe and Caldwell, “Institutional Foundations.” 40. See, e.g., Burke, Lawyers, Lawsuits, 45–51, 185–87; Kagan, Adversarial Legalism, 55. ATLA changed its name to the American Association for Justice in 2006 because of negative public attitudes toward the phrase “trial lawyer.” Al Kamen, “Just Don’t Call Them the Suers,” Washington Post, July 14, 2006, A19. I use the name ATLA here because it will be familiar to more readers and was the organization’s name during the period examined in this book. 41. Burke, Lawyers, Lawsuits, 185–87. 42. See, e.g., Garry, A Nation of Adversaries; Nathan Glazer, “Towards an Imperial Judiciary?” Public Interest 41 (1975): 104–23; Olson, The Excuse Factory; Olson, Litigation Explosion; Ross Sandler and David Schoenbrod, Democracy by Decree: What Happens When Courts Run Government (New Haven: Yale University Press, 2002); see also the collected essays in That Eminent Tribunal: Judicial Supremacy and the Constitution, ed. Christopher Wolfe (Princeton: Princeton University Press). 43. Terri Peretti, In Defense of a Political Court (Princeton: Princeton University Press, 1999). 44. Malcolm M. Feeley and Edward L. Rubin, How the Courts Reformed America’s Prisons: Judicial Policy Making and the Modern State (New York: Cambridge University Press, 1998), 333-34. 45. See, e.g., Susan Lawrence, “Justice, Democracy, Litigation, and Political Participation,” Social Science Quarterly 72 (1991): 464–77; Zemans, “Legal Mobilization”; Busch, Kirp, and Schoenholz, “Taming Adversarial Legalism.”
INDEX
Abourezk, James, 4, 154–55 access to plaintiff status, 25–26 administrative enforcement, 145, 155, 215; in H.R. 405, 98–104 administrative implementation, subversion of, 34–35, 37–38, 40–41 administrative process, as form of command and control regulation, 31–32 adversarial legalism, 57–58 affirmative action, 174, 179, 181 AFL-CIO, 133, 138 Age Discrimination in Employment Act (1967), 118, 200–202 agencies: powers of, 45; single-mission, 109. See also Equal Employment Opportunity Commission; names of other agencies agency bias, 100–101, 104–5, 194–95, 218 agency capture, 40, 50, 136–37, 195 agency policymaking, challenges to, 11, 22 Alexander, Clifford, 137, 164, 184–85 Allen, James, 160 Allen v. State Board of Education, 156, 274n159 allocation of power, and private enforcement regimes, 44–49 Alyeska Pipeline Service Co. v. Wilderness Society (1975), 152, 156 American Bar Association, 69, 75, 116, 126, 183, 199 American Civil Liberties Union (ACLU), 111, 116, 145, 151, 163, 182–83 American exceptionalism, 13, 227–28 American political institutions, and private enforcement regimes, 216–27 American rule, 26–27, 61–62, 97, 110, 152, 156, 241n19 Americans with Disabilities Act (1990), 118, 200–202, 287n165 analytic narratives, 86 antibureaucracy theme, 226–27 antilawyer theme, 186 antilitigation theme, 186 antistatism, 43
appointment of counsel, 110–13, 152 Association of Trial Lawyers of America (ATLA), 69, 75, 232 attorney’s fee awards: and growth of civil rights bar, 149–51; proportionate, 190; and retention of counsel, 190–92. See also fee-shifting attorney’s fees: in Civil Rights Fees Act, 147–55, 160–63; in Title VII enforcement provisions, 110–13; in voting rights cases, 156–60. See also fee-shifting Balkin, Jack M., 32 Ball, Joseph, 218 Bartlett, Steve, 197 bench trial, 63, 184 Berg, Richard, 117, 153 bicameralism, 166 Blumrosen, Alfred, 109–11, 136 Bolling, Richard, 160 Bork, Robert, 178–80 Boureslan v. Arabian American Oil Co., 181 Brookings Institution, 135 Brown, William, 132 budget constraint hypothesis, 71–72, 76, 81, 91–92, 128, 170–71, 211, 265n162 budget constraint variable, 76, 81 bureaucratic behavior, 70 bureaucratic drift hypothesis, 40–42, 57, 60, 69–70, 79, 124–25, 166–67, 207, 225 Burger, Chief Justice Warren, 14, 172 Burke, Thomas, 5, 19, 34, 232 Bush, President George H. W., 186–87, 189–90 Bush, President George W., 215–16 business groups, 107, 113, 186, 196–97, 210 business witnesses variable, 75–76 Byrd, Robert, 120–21 Cameron, Charles, 48, 222 Carpenter, Daniel, 10
294
INDEX
case selection, 89–91 causal narratives, 86 cease-and-desist powers, for EEOC, 99– 101, 104–5, 118, 129–30, 132–35, 138, 145–46, 166, 184–86, 197, 215, 226 Cellar, Emanuel, 134 Center for Law and Social Policy, 151 Chamber of Commerce, U.S., 76, 107, 113, 186 choice to litigate, 21–25 church groups, 116 Citizens’ Commission on Civil Rights, 173 citizen suit provisions, 63, 219 Civil Rights Act (1866), 147–48, 184 Civil Rights Act (1870), 64, 111, 261n93 Civil Rights Act (1871), 97, 148, 153 Civil Rights Act (1957), 96–97 Civil Rights Act (1960), 96–97 Civil Rights Act (1964), 3, 28, 68, 85, 90; passage of, 94–95; public accommodations provisions (Title II), 100; Title VI, 139, 148. See also Title VII Civil Rights Act (1968). See Fair Housing Act (FHA, 1968) Civil Rights Act (1991), 63, 182–90; effects on private enforcement, 199–204; litigant mobilization in, 190–92; as override legislation, 189; and separation of powers conflicts, 192–98; and Title VII claims, 200–202 Civil Rights Attorney’s Fees Awards Act (Civil Rights Fees Act, 1976), 130, 147–56, 160–63, 169–71, 181, 226–27 civil rights bar, 130, 149–54, 168–69, 208–9, 215, 231–32 Civil Rights Commission, 164 Civil Rights Division (Justice Department), 97 Civil Rights Fees Act (1976). See Civil Rights Attorney’s Fees Awards Act civil rights groups, 116, 139, 148, 215; and attorney’s fee awards, 149, 163; and CRA (1991), 208–9; and EEOC, 135–37; and enactment of Title VII, 126–27; enforcement preferences, 167–69; litigation success, 142–47; and private enforcement, 133–34; and private enforcement infrastructure, 151–52 civil rights litigation, in Reagan era, 176–78 civil rights movement, 14, 95–96 class actions, 144–46, 149, 162, 174
Clean Air Act (1970), 217, 219–20 Clean Air Act (1990), 217, 219 coalition drift hypothesis, 37–42, 56–57, 60, 79, 123–24, 166, 207, 224 Collins, Cardiss, 220 command and control system, 31 Commission on Civil Rights, U.S., 97, 135–36, 158; public letter on Nixon administration, 137 committee hearings, 75 committee system, 166 common law default rules on attorney’s fees and damages, 64–65 comparative institutional implications, and private enforcement regimes, 56–58 compromise, 5–6, 20, 42–44, 225–27. See also veto points and compromise hypothesis concessions, in CRA (1991), 187 conciliations, EEOC and, 131–32 Congress. See House of Representatives; legislative-executive conflict; legislativejudicial conflict; Senate Congressional Black Caucus, 179–80 congressional enactment of private enforcement regimes, measure of, 60–68 consent decrees, repudiation of, 179 constituency for private enforcement regimes, legislative, 68 constitutional litigation, 54 constitutional rights, assertion of, 54 Conyers, John, 193 Cook, Marlow, 142 Cooter, Robert D., 24 cost, of private enforcement regimes, 55–56 Counsel for Public Interest Law, 151 court of appeals, 73 courts: regime politics approach to, 32–34; role in the regulatory process, 5, 19; and rule articulation, 46–47, 49–54; and weak American state, 7 credible commitments, 33–34, 39 cross-case analysis, 91 cross-national research, 227–28 Cuban Liberty and Democratic Solidarity (Libertad) Act (1996), 68 cultural transformation hypothesis, 13–14, 16 Dahl, Robert, 32 damages: compensatory, 183, 187, 189,
INDEX 196–97; monetary, 27, 172; punitive, 183, 187, 189, 196–97 damages enhancements, 62–68, 82 damages provisions, in CRA (1991), 183 Danforth, John, 189, 210 Decker, Jefferson, 177 defense lawyers, 209 “defunding the left,” 177 delegation literature, 34–36, 48 democracy, and private enforcement regimes, 233–34 Democratic Party, 70, 80–81, 92–93, 96, 170, 228–30; and Civil Rights Fees Act, 161; and CRA (1964), 94–95; and CRA (1991), 198; criticism of Nixon administration, 164–65; response to Reagan initiative, 178–80; and School Aid Act, 142. See also House of Representatives; Senate; southern Democrats Derfner, Armand, 150, 159 Derfner, Mary, 150 despotic power, 9 Dirksen, Everett, 104, 106–14, 119–21, 137, 151 disability claims, 200 discerning, 86 disparate impact theory, 143–44, 165, 175, 179–80, 270n85 divided government, 244n55; historical patterns of, 221–23. See also legislativeexecutive conflict divided government variable, 72, 76–78 Dobbin, Frank, 204 Dominick, Peter, 142 Drinan, Robert, 156 Duke, David, 188 Eastland, James, 95 economic regulation, 74–75 economic regulation variable, 75 Edelman, Marion Wright, 141 Education Amendments (1972), Title IX, 148 Edwards, Donald, 138, 193 EEOC. See Equal Employment Opportunity Commission EEOC determination, as evidence, 242n32 Eisenhower administration, 222 electoral risk, 73–74 electoral risk variable, 74, 79 electoral uncertainty, 37–40, 166, 207 embedded case, 89–90
295
Emergency Price Control Act (1942), 12, 27 Emergency School Aid Act (1972), 140–41, 168–69 empirical model, for enactment of private enforcement regimes, 17, 68, 72–81, 87–88, 90–92, 121, 123–25, 127–28, 164–70, 205–8, 210, 223–24, 226–27, 230–31 employment discrimination bar, 178, 183, 199, 203, 208–9 employment practices, facially neutral, 143 enactment date, identification of, 82 enforcement. See administrative enforcement; executive enforcement; private enforcement; private enforcement regimes enforcement endowments, 8 enforcement level, 36 Engle, Robert F., 84 English rule, 26–27, 61–62, 241n17, 241n19 environmental policy, 217, 219–20 Environmental Protection Agency (EPA), 219–20 Epp, Charles, 204 Epstein, David, 35, 244n55 Epstein, Lee, 92 Equal Employment Opportunity Act (EEOA, 1972), 131–35, 142–43, 146–47, 150, 226 Equal Employment Opportunity Commission (EEOC), 90; cease-and-desist powers, 99–101, 104–5, 118, 129–30, 132–35, 138, 145–46, 166, 184–86, 197, 215, 226; claims under (1980-2002), 200–202; complaint processing, 179; and executive priorities, 135–37; governmental right to sue, 289n178; interpretive guidelines, 278n15; during Johnson administration, 136; leadership ideology, 202–3; proposed, 98–99, 101–2, 104, 107–10, 118, 122, 212–13; prosecutorial powers, 147; during Reagan era, 174–75, 178–80; and underenforcement, 131–32, 192–98 Equal Pay Act (1963), 27 Erlenborn, John, 146 Ervin, Senator Sam, 114–15 executive enforcement, 151, 164–65, 171, 192–98; rightward shift in, 137–42; and Voting Rights Act, 157–60 executive ideology, 218. See also legislative-executive conflict
296
INDEX
executive power, loss of faith in, 135–37 expected benefits (EB), 27, 51–53, 62 expected costs (EC), 26–27, 51–53, 62 expected value (EV), 27, 51–53, 62 fair employment practice laws (states), 85, 113–14 Fair Housing Act (FHA, 1968), 119–21 Fair Labor Standards Act (1938), 8, 68 False Claims Act (1863), 26 Fawell, Harris, 196 Federal Courts Study Committee (FCSC), 190–91 federalism, and private enforcement regimes, 54–55 federal job discrimination statutes, 3 Federal Rules of Civil Procedure, Rule 23, 144, 146 fee-capping, 177 fee-shifting, 155, 168, 261n93; in Civil Rights Fees Act, 147–55, 160–63; for FHA (1968), 119–21; interest groups and, 162–63; lessons learned, 153–55; in School Aid Act, 140–42; spread of, 130–31; in Title VII, 91–92, 105–6, 110–15; in voting rights cases, 156–60 fee shifts: identification of, 82; plaintiffs’, 62–68 Fein, Bruce, 182 filibuster, 38, 94–96, 102, 104, 106, 112, 118–20, 122, 133, 160, 165–66, 185, 226 filing fees, in Title VII enforcement provisions, 110–13 Filvaroff, David, 108, 112 Findley, Paul, 101 Fiorina, Morris, 31 Florida, 241n17 Force Act (1870), 64, 111, 261n93 Ford Foundation, 150 Foreman, Christopher, 6 Fourteenth Amendment, 97 free lunch, and private enforcement regimes, 55–56 Frelinghuysen, Peter, 101 Frymer, Paul, 92, 236n16 Galanter, Marc, 8, 30 General Accounting Office, 193 general deterrence, 9, 203–4 George, Alexander, 86
Giles, Michael W., 73 Glasser, Ira, 174 Glickstein, Howard A., 138, 140–41, 157, 164 Goodell, Charles, 103 Govan, Reginald, 182–84 governmental prosecutions vs. private litigation, 12 Graham, Hugh Davis, 109 Granger, Clive W., 84 Green, Harold, 108 Greenberg, Jack, 97–98, 111, 126, 133–34, 136–38, 142–46 Greve, Michael S., 177 gridlock interval, 253n48 Griffin, Robert, 101, 103 Griggs v. Duke Power Co. (1971), 143–44, 165, 180–81, 270n85 Hacker, Jacob, 9–10, 43 Hart, Philip, 134 Hawkins, Augustus, 179, 182–83, 193, 195 Health, Education and Welfare Department (HEW), 139 Helms, Jesse, 160 Hettinger, Virginia A., 73 Hickenlooper, Bourke, 115 Hill, Anita, 187 hiring quotas, 186–88 historical evidence: and causal status of legislative-executive conflict, 220–21; and CRA (1991), 190–99; interpretation of, 88–89. See also process-tracing evidence historical institutionalism, 5, 19 holistic case, 89–90 Holtzman, Elizabeth, 155 Hooks, Benjamin, 182, 193–94 Horn, Murray, 42 House of Representatives: Civil Rights Oversight Subcommittee, 138–39, 157; development on Title VII, 98–106; Education and Labor Committee, 100, 103, 116, 179, 185; Judiciary Committee, Subcommittee No. 5, 102–3, 116 Housing and Urban Development Department (HUD), 119–20 Howard, Christopher, 9–10, 43 H.R. 405, 98–104, 121, 134–35, 184, 212–13, 258n30 H.R. 7152, 102–4
INDEX Humphrey, Hubert, 112–14, 126, 262n101 hybrid cease-and-desist/private litigation framework, 130, 134, 142–45, 185, 215 ideology, 72–73. See also executive ideology; judicial ideology; legislative-executive conflict; legislative-judicial conflict Immergut, Ellen, 38 impact litigation, 153 incentives, 31, 53, 58; monetary, 190–92; for passage of civil rights legislation, 95–96; for private litigation to achieve regulatory state goals, 43–44; for the private provision of benefits to achieve welfare state goals, 43–44; in Title VII, 91–92 Independent Fed’n of Flight Attendants v. Zipes, 181 infrastructural power, 9–10 institutional conflict, 122 institutional fragmentation, 34–38, 49, 129, 214 instrumental utility: of judiciary, 32–34; of private enforcement regimes, 34 insulation strategy, 37–38, 50, 55, 61, 123–24, 166, 203, 224–25 interbranch conflict, 172, 216–23; over Title VII enforcement, 173–80. See also legislative-executive conflict; legislativejudicial conflict interbranch realignment, 172 interest groups, 11, 23, 42, 69–70, 75–76, 232, 251n34; and attorney’s fee awards, 149–51; and Civil Rights Fees Act, 162–63; and CRA (1964), 116; and CRA (1991), 198–99; and private enforcement regimes, 228–30. See also civil rights groups; issue group hypothesis; issue groups; names of organizations Interstate Commerce Act (1887), 27–28, 60, 63–65 issue group hypothesis, 69–71, 81, 166–67, 207–8 issue groups, 71, 79, 124–25, 167–68, 254n53. See also civil rights groups issue groups variable, 75 issue witnesses variable, 75, 79 Jeffords, James, 188 job discrimination, “pattern of practice” of, 109
297
job discrimination bar, 178, 183, 199, 203, 208–9 job discrimination laws, as case study, 89–91 job discrimination lawsuits, 3, 272n121; growth of, 186, 198, 200; volume of, 3, 85, 89–90, 177–78 job discrimination prohibitions, 118–19 Johnson, President Lyndon B., 48 Johnson administration, 130, 134, 136 judge-dominated approach to civil law, 8 judges: as neutral adjudicators, 50; voting patterns, 47 judicial bias, 184 Judicial Conference, 144 judicial direction value, 73 judicial direction variable, 78–79 judicial distance value, 73 judicial distance variable, 78–79 judicial drift, 49–50 judicial friendliness, 123, 206, 223–24; in Title VII early years, 142–47 judicial friendliness hypothesis, 51–52, 78, 165, 202 judicial hostility, 123, 223–24 judicial hostility hypothesis, 52–54, 78, 165, 202, 205–6 judicial ideology, 37, 47, 73, 123, 165, 176, 202, 223–24; and CRA (1991), 205–6; and private enforcement regimes, 49–54; in statistical model, 78–79. See also legislative-judicial conflict judicial imperialism, 233 judiciary, instrumental utility of, 32–34 jury trial, 62–63 Justice Department: Civil Rights Division, 131, 173–74; under Reagan administration, 173; and school desegregation, 139; and Title VII enforcement, 109, 136–37; and voting rights enforcement, 157–60 Kagan, Robert, 5, 8, 19, 34, 43, 57–58 Kalev, Alexandra, 204 Katsouyanni, K., 83 Katzenbach, Nicholas, 108, 112, 159 Katzenstein, Peter, 8 Katznelson, Ira, 6, 9 Kempton, Murray, 107 Kennedy, Anthony, 176, 180 Kennedy, Edward, 154–55, 182, 285n110, 285n118
298
INDEX
Kennedy (JFK) administration, 102–3; Civil Rights Division, 158; and civil rights legislation, 104–5, 108–10 Kleindienst, Richard, 159 Krehbiel, Keith, 253n48 Kreiter, Nancy, 194 Ku Klux Klan, 261n93 Labor Department, 175–76 labor policy, 217–19 labor unions, 116, 217–19 Landes, William M., 33, 39, 50 Lantos, Tom, 179–80 law and economics literature, 22–25 law and economics model, 182 laws, expired, repealed, or struck down, 82–83 lawyer-dominated approach to civil law, 8 lawyers associations, 69, 75, 79–80, 198–99, 230; and enactment of Title VII, 125–27; and fee shifting, 163; and lobbying for CRA (1964), 116. See also names of organizations Lawyers’ Committee for Civil Rights Under Law, 116, 149, 151, 158–59, 183; Attorney’s Fees Project, 149 lawyer witnesses variable, 75, 79–80 Leadership Conference on Civil Rights, 117, 119, 133–34, 139, 146, 153, 182–85, 209 Legal Defense Fund (NAACP), 132, 135, 146, 149, 153, 183 legal infrastructure, 30–31 legal process, as form of command and control regulation, 31–32 legislation, passage of, 88 legislative choice, 68–72, 88, 161, 206, 221; general model of, 58–59; in litigant mobilization, 64–65; and private statutory litigation, 3–5 legislative-executive conflict, 34–37, 40–41, 47–48, 60, 87, 91, 101, 167, 172, 216–23, 245n73; and Clean Air Act (1970), 219–20; for control of bureaucracy, 4–5, 19; and CRA (1991), 204–5; and enactment of Title VII, 121–22, 129; and enforcement of school desegregation mandates, 139–40; and historical evidence, 220–21; and issue groups, 208; in parliamentary systems, 56–57; and private enforcement regime, 34–37; and process-tracing evidence, 164–65;
in Reagan era, 178–80; and separation of powers, 216–17; in statistical model, 76–78; and Taft-Hartley Act (1947), 217–19; and voting rights enforcement, 157–60 legislative historical evidence, and CRA (1991), 190–99 legislative instruments of legal mobilization, 21–31 legislative-judicial conflict, 49–54, 172, 180–82 legislative manipulation of expected value, 28–30 Levi, Edward, 159 Levinson, Stanford, 32 Lieberman, Robert, 10, 289n1 limitations period, 26 Lipset, Seymour Martin, 13, 227 litigant mobilization, 19–20, 51, 55, 140, 202; and CRA (1991), 189–92, 200 litigation, threat of, 201 litigation explosion, 3, 12–13, 69, 186, 198, 222–23, 233–34 litigation state, 10–13, 214–16, 228–29, 233–34 litigation success, in Title VII early years, 142–47 Lorance v. AT&T Technologies, 181 “loser pays” rule, 241n17 Luban, David, 8 MacCoun, Robert, 23 Mann, Michael, 9 March on Washington (August 1963), 94 Marshall, Burke, 108, 112 Martin, Dave, 101 Martinez, Matthew, 179 Martin v. Wilks, 181 Mathias, Charles McCurdy, 155 McCarty, Nolan, 72 McCubbins, Mathew, 5, 19, 21, 40–41, 245n73 McCulloch, William, 102, 104 McKeown, Timothy, 86 McNollgast, 5, 19, 21, 40–41, 245n73 Melnick, Shep, 5, 19, 226 Metzenbaum, Howard, 193 Mexican-American Legal Defense Fund, 183 Miller, Jack, 115 Mitchell, Clarence, 112, 119, 141, 153 Mitchell, George, 220
INDEX Mitchell, John, 157 modernization/economic development hypothesis, 13–14, 16 Moe, Terry, 5, 19, 34–35, 37–38, 40, 56–58, 224, 228 Mondale, Walter, 119–21, 140–42, 164 Moorhead, Carlos, 197 Morgan, Charles Jr., 158–59 National Association for the Advancement of Colored Persons (NAACP), 111, 113, 119, 145–46, 153, 169, 182–83, 198–99 National Association of Manufacturers, 76, 107, 186, 196–97, 199 National Employment Lawyers Association, 178, 183, 199, 203, 209 National Gold and Silver Stamping Act (1906), 26 National Labor Relations Act (1935), 3, 99, 216, 218 National Labor Relations Board (NLRB), 94, 100–102, 109, 216–19 National Lawyers Guild, 111 National Legal Aid and Defender Association, 163 National Urban League, 145–46, 183 National Women’s Law Center, 183 Native American Rights Fund, 149 New Deal, 75, 100–102, 111 new institutionalism, 24, 38 Newport News Shipbuilding case, 145 Nixon, President Richard M., 120; “southern strategy,” 137, 158 Nixon administration, 130, 135, 164–65, 222; enforcement record, 157–60; and Title VII enforcement, 137–42 Noll, Roger, 5, 19, 21, 40–41, 245n73 NOMINATE scores, 72, 222 Novak, William, 6, 10 Oakar, Mary Rose, 195 Occupational Safety and Health Act (OSHA, 1970), 3, 25–26, 240n12 O’Connor, Justice Sandra Day, 176, 180 O’Connor, Karen, 92 Office of Federal Contract Compliance Programs (OFCCP), 175–76 O’Halloran, Sharyn, 35, 244n55 Olson, Walter, 172 one-shotters, 30 opposition seat share variable, 72, 76–78 other regulation variable, 75
299
overcoming obstructions scenario, 33–34 override legislation, for 1989 Supreme Court decisions, 182–90 oversight literature, 21–22 Owens, Major Robert Odell, 193 Parker, Frank, 159 parliamentary systems, 56–58, 228 Partial-Birth Abortion Ban Act (2003), 68 partisan identification, 72 partisan seat share variable, 74, 80–81 party alignment: and enactment of Title VII, 127–28; in statistical model, 80–81 party alignment hypothesis, 70–71, 229; and CRA (1991), 210–11; and processtracing evidence, 169–70 party polarization, 5, 186, 198; historical patterns of, 221–23 pattern-or-practice suits, 131, 133 Patterson v. McLean Credit Union, 181 Payne, Donald, 194–95 People for the American Way, 183 Peppers, Todd, 73 Pierson, Paul, 149 plaintiffs bar, 209, 231 policy drift, 224–25; and CRA (1991), 207; and enactment of Title VII, 123–25; and process-tracing evidence, 166–67 policy feedback, 59, 129; self-reinforcing, 129–31, 135, 149, 153, 168, 211 policy instruments, 7–9 policy learning, 149; and fee-shifting, 153–55; as form of feedback, 135 political culture, and American exceptionalism, 227–28 political entrenchment, 33–34, 39 political parties, 70–71, 74; and interest groups, 232; and private enforcement regimes, 228–30. See also Democratic Party; legislative-executive conflict; party alignment; Republican Party Poole, Keith T., 72 Posner, Richard A., 33, 39, 50 Powell, Adam Clayton, 98 predictability assumption, 20–21 presidential distance, 253n48 presidential distance value, 73, 78 presidential veto, 187 presidents, and signing of laws using private enforcement regimes, 48 Price Waterhouse v. Hopkins, 181
300
INDEX
private enforcement, 190, 203–4; in Civil Rights Fees Act, 160–62; CRA (1991) effects on, 199–204; curtailed by Supreme Court, 180–82; for Title VII, 105–14 private enforcement litigation: as form of state intervention, 7; rising rates of, 13–16; vs. tort litigation, 15, 232; volume of, 10–13 private enforcement regimes: and allocation of power, 44–49; and American exceptionalism, 227–28; and American political institutions, 216–27; and bureaucratic drift, 40–42; and comparative institutional implications, 56–58; and compromise, 42–44; cost of, 55–56; and democracy, 233–34; disadvantages of, 55–56; and executive powers, 45–46; and federalism, 54–55; as form of command and control, 31; and free lunch, 55–56; instrumental utility of, 34; and insulation strategy, 38–39; and interest groups, 228–30; and judicial ideology, 49–54; and legislative-executive conflict, 34–37; measure of, 60–68; other causes of, 68–72, 90–91; persistence of, 65–67; policy domains, 67–68; and political parties, 228–30; purposeful use of, 64; and regime politics, 32–34; and state capacity, 214–16; use of term, 4. See also empirical model, for enactment of private enforcement regimes private litigation, 177–78; in early years of Title VII, 131–32; vs. governmental prosecutions, 12; and infrastructural state power, 9–10; policy effects of, 46; role in shaping behavior, 8–9; role in statutory implementation, 3–5; volume of, 150, 153–54, 162 private right of action, express vs. implied, 156 private statutory litigation, and regulatory state capacity, 6–13 probability of prevailing (p), 27–28, 51–53 process-tracing evidence, 86–89, 163–64, 216; and CRA (1991), 204–11; and enactment of Title VII, 121–28; and issue group hypothesis, 230–31; and judicial ideology, 223–24; and party alignment, 229; and rent-seeking lawyer hypothesis, 231; and veto points hypothesis, 226 Progressive era, 75, 111
public enforcement: for civil rights legislation, 96–98; in H.R.405, 98–104 public interest litigation, 151–52 public-private convergence, and American state power, 10 Puerto Rico, 4, 85 Quie, Albert, 103 racial balance, 104–5 ratchet effect, 67 rational choice institutionalism, 5, 19 rationality assumption, 20–21 rational litigant behavior model, 22–25 Rauh, Joseph, 104, 112, 119, 139, 145 Reagan, President Ronald: civil rights record, 193, 217; election of, 205 Reagan administration: and civil rights enforcement, 173–80; demobilization of administrative enforcement, 202–3; and environmental policy, 220 Reconstruction South, 261n93 regime politics, and private enforcement regime, 32–34 regulation, use of term, 6 regulation literature, 74 regulatory productivity, magnitude and nature of, 74–75 regulatory state capacity, and private statutory litigation, 6–13 Rehabilitation Act (1973), 118 Rehnquist, Chief Justice William, 176, 180 Religious Freedom Restoration Act (1993), 148 rent-seeking lawyer hypothesis, 69, 71, 81, 91–92, 231–32; and CRA (1991), 208–9; and enactment of Title VII, 125–27; and process-tracing evidence, 168–69; in statistical model, 79–80 repeat players, 30 Republican Party, 70–71, 96, 168, 186, 228–30; and Civil Rights Fees Act, 161; and CRA (1964), 94–95; and CRA (1991), 197–98, 210–11, 213; and David Duke, 188; and FHA (1968), 119–21; and Hill-Thomas hearings, 188; and private enforcement, 121–22, 148, 162, 170–71, 211–13; and School Aid Act, 142; and Title VII private enforcement compromise, 118. See also House of Representatives; Senate retention of counsel, 190–92
INDEX Reynolds, William Bradford, 173–74 Reynolds v. Sims, 158 rhetoric, in civil rights policy struggles, 88 Richardson, Elliot, 159 rights assertion, 14 rights consciousness, 14, 228 roll calls: for CRA (1991), 198; in Title VII, 114–16 Romano, Roberta, 70–71 Roosevelt, James, 98, 136 Rosenthal, Howard, 72 rule articulation, 165; and courts, 46–47; courts vs. agencies, 49–54; use of term, 44–45 rule enforcement: and agencies, 45; and rule articulation, 54; use of term, 44–45 S. 1732, 114 Saltonstall, Leverett, 114 Saxbe, William, 159 Scalia, Justice Anthony, 176, 180 School Aid Act (1972), 147–48, 166, 170; interest groups and, 162–63 school desegregation, 130, 139 Schwartz, J., 21, 83 Scott, Hugh, 154–55 Securities Industry Association, 76 Segal, Jeffrey A., 50 selection bias, 91 Senate: Committee on Labor and Public Welfare, 140–42; Committee Report on the Clean Water Act, 219–20; and developments on Title VII, 106–16; filibuster rule, 95; Judiciary Committee, 95, 156. See also filibuster separation of powers, 214, 228 separation of powers conflicts, 72–73, 192–98 separation of powers system, 45–46, 56–58. See also legislative-executive conflict; legislative-judicial conflict Seventh Amendment, 282n66 sexual harassment, 187–88 Seymour, Richard, 185 Shapiro, Martin, 32 Shepsle, Kenneth, 37, 42 Sherman Antitrust Act (1890), 27, 64–65 Shulman, Stephen, 131 Sieberling, John, 155 Skocpol, Theda, 6 social regulation, 74–75 social regulation variable, 75
301
Souter, Justice David, 181 South Carolina, 159 southern Democrats, 95, 137; and FHA (1968), 119–21; and Title VII, 114–16, 122 southerners: and Civil Rights Fees Act, 161; and CRA (1991), 198 Sovern, Michael, 117 Spaeth, Harold J., 50 specific deterrence, 9, 203–4 state: strong, 6, 57, 214; weak, 6–7, 34, 44, 57–58, 214 state, American, as weak state, 6–7, 34, 214, 228 state capacity: measures of, 6; and policy instruments, 7–9; and private enforcement regimes, 214–16 state power, American, and public-private convergence, 10 statistical methods, 83–84 statistical model, 225 statutes of limitations, 26, 114, 179, 181, 187 statutory design, 3 statutory interpretation, 51; institutional norms of, 50; as policymaking, 46 statutory specificity, obstacles to, 47 Steinmo, Sven, 42–43, 57–58 Stevens, Ted, 188 Stewart, John G., 108 sticky status quo, 19–20, 38–39, 41, 55, 177, 224–25 subversion, 54; of administrative implementation, 34–35, 37–38, 40–41; threat of, 49, 228 Supreme Court, U.S., 61–62, 143–44, 152, 156, 165, 172–73, 270n85; curtailment of Title VII private enforcement, 180–82; 1989 decisions, 207–8; Reagan appointees, 176. See also override legislation; case names Taft, Robert, 218 Taft-Hartley Act (1947), 68, 217–19 Telemarketing and Consumer Fraud and Abuse Prevention Act (1994), 68 Thirteenth Amendment, 152 Thomas, Justice Clarence, 174–75, 185, 187 Thurmond, Strom, 114, 160 Title VII, 48, 85–86, 90; access to plaintiff status, 25–26; appointment-of-counsel
302
INDEX
Title VI (cont.) provisions, 152, 226; attorney’s fee provision, 231; back-pay limitation, 183–84; claims under (1980-2002), 200–202; disparate impact theory, 144; enforcement provisions, 4, 94–95, 114–19; equal opportunity view of, 143–44; equal treatment view of, 143– 44; fee-shifting provisions, 27, 91–92, 105–6; House developments on, 98–106; incentives in, 91–92; institutional and political context for passage, 95–96; intent and consequences, 211–13; and legislative-executive conflict, 121–22; limitations period, 26; monetary damages, 27, 172; Senate developments on, 106–14; Supreme Court and, 180–82; and underenforcement, 190; and volume of private litigation, 150, 154 Tocqueville, Alexis de, 11 tort litigation, 15, 232 trial by jury, 184, 187 Tunney, John V., 154–55, 158, 160 Ulen, Thomas, 24 underenforcement, 111–12; and EEOC, 131–32, 192–98; in Reconstruction civil rights laws, 97–98; and Title VII, 190 unintended consequences, 211 Unsoeld, Jolene, 194 Vargyas, Ellen, 193, 195–96 Verba, Sidney, 227 veto bargaining, 48 veto pivot model, 253n48
veto points, 5–6, 19, 38, 42–44 veto points and compromise hypothesis, 225–27; and CRA (1991), 206–7; and enactment of Title VII, 122–23; and process-tracing evidence, 165–66 Violence Against Women Act (1994), 148 voting alignments: for Civil Rights Fees Act, 160–62; for CRA (1991), 198; in Title VII, 114–16 voting rights, 130 Voting Rights Act (1965), 147, 156, 274n159–274n160 Voting Rights Act Amendments (1975), 155–60, 168–69 voting rights protections, 261n93 Wards Cove Packing Co. v. Atonio, 180 Warner, John, 188 Washington Legal Foundation, 199 Waxman, Henry, 220 Weingast, Barry, 5, 19, 21, 40–41, 245n73 West Virginia University Hospitals, Inc. v. Casey, 181 White, Justice Byron, 180 Whittington, Keith E., 33 whittling down, 44 Wilkins, Roy, 136, 139 Wilson, James Q., 6, 56, 228 within-case analysis, 86–89 Women’s Law Fund, 149 Women’s Legal Defense Fund, 151, 183, 198–99 Young, Whitney, 136
princeton studies in american politics: historical, international, and comparative perspectives Ira Katznelson, Martin Shefter, and Theda Skocpol, series editors recent titles The Litigation State: Public Regulations and Private Lawsuits in the United States by Sean Farhang Reputation and Power: Organizational Image and Pharmaceutical Regulation at the FDA by Daniel Carpenter Presidential Party Building:Dwight D. Eisenhower to George W. Bush by Daniel J. Galvin Fighting for Democracy: Black Veterans and the Struggle Against White Supremacy in the Postwar South by Christopher S. Parker The Fifth Freedom: Jobs, Politics, and Civil Rights in the United States, 1941–1972 by Anthony Chen Reforms at Risk: What Happens after Major Policy Changes Are Enacted by Eric Patashnik The Rise of the Conservative Legal Movement: The Long Battle for Control of the Law by Steven M. Teles Why Is There No Labor Party in the United States? by Robin Archer Black and Blue: African Americans, the Labor Movement, and the Decline of the Democratic Party by Paul Frymer Political Foundations of Judicial Supremacy: The Presidency, the Supreme Court, and Constitutional Leadership in U. S. History by Keith E. Whittington The Transformation of American Politics: Activist Government and the Rise of Conservatism edited by Paul Pierson and Theda Skocpol
Disarmed: The Missing Movement for Gun Control in America by Kristin A. Goss Filibuster: Obstruction and Lawmaking in the U.S. Senate by Gregory Wawro and Eric Schickler Governing the American State: Congress and the New Federalism by Kimberley S. Johnson What a Mighty Power We Can Be: African-American Fraternal Groups and the Struggle for Racial Equality by Theda Skocpol, Ariane Liazos, and Marshall Ganz When Movements Matter: The Townsend Plan and the Rise of Social Security by Edwin Amenta Shaping Race Policy: The United States in Comparative Perspective by Robert C. Lieberman How Policies Make Citizens: Senior Political Activism and the American Welfare State by Andrea Louise Campbell Dividing Lines: The Politics of Immigration Control in America by Daniel J. Tichenor Managing the President’s Program: Presidential Leadership and Legislative Policy Formulation by Andrew Rudalevige Shaped by War and Trade: International Influences on American Political Development edited by Ira Katznelson and Martin Shefter Dry Bones Rattling: Community Building to Revitalize American Democracy by Mark R. Warren Disjointed Pluralism: Institutional Innovation and the Development of the U.S. Congress by Eric Schickler The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innovations in Executive Agencies, 1862–1928 by Daniel P. Carpenter The Rise of the Agricultural Welfare State: Institutions and Interest Group Power in the United States, France, and Japan by Adam D. Sheingate
In the Shadow of the Garrison State: America’s Anti-Statism and Its Cold War Grand Strategy by Aaron L. Friedberg Stuck in Neutral: Business and the Politics of Human Capital Investment Policy by Cathie Jo Martin Uneasy Alliances: Race and Party Competition in America by Paul Frymer Faithful and Fearless: Moving Feminist Protest inside the Church and Military by Mary Fainsod Katzenstein Forged Consensus: Science, Technology, and Economic Policy in the United States, 1921–1953 by David M. Hart Parting at the Crossroads: The Emergence of Health Insurance in the United States and Canada by Antonia Maioni Bold Relief: Institutional Politics and the Origins of Modern American Social Policy by Edwin Amenta The Hidden Welfare State: Tax Expenditures and Social Policy in the United States by Christopher Howard Morning Glories: Municipal Reform in the Southwest by Amy Bridges Imperiled Innocents: Anthony Comstock and Family Reproduction in Victorian America by Nicola Beisel The Road to Nowhere: The Genesis of President Clinton’s Plan for Health Security by Jacob Hacker The Origins of the Urban Crisis: Race and Inequality in Postwar Detroit by Thomas J. Sugrue Party Decline in America: Policy, Politics, and the Fiscal State by John J. Coleman The Power of Separation: American Constitutionalism and the Myth of the Legislative Veto by Jessica Korn
Why Movements Succeed or Fail: Opportunity, Culture, and the Struggle for Woman Suffrage by Lee Ann Banaszak Kindred Strangers: The Uneasy Relationship between Politics and Business in America by David Vogel From the Outside In: World War II and the American State by Bartholomew H. Sparrow Classifying by Race edited by Paul E. Peterson Facing Up to the American Dream: Race, Class, and the Soul of the Nation by Jennifer L. Hochschild Political Organizations by James Q. Wilson Social Policy in the United States: Future Possibilities in Historical Perspective by Theda Skocpol Experts and Politicians: Reform Challenges to Machine Politics in New York, Cleveland, and Chicago by Kenneth Finegold Bound by Our Constitution: Women, Workers, and the Minimum Wage by Vivien Hart Prisoners of Myth: The Leadership of the Tennessee Valley Authority, 1933–1990 by Erwin C. Hargrove Political Parties and the State: The American Historical Experience by Martin Shefter Politics and Industrialization: Early Railroads in the United States and Prussia by Colleen A. Dunlavy The Lincoln Persuasion: Remaking American Liberalism by J. David Greenstone Labor Visions and State Power: The Origins of Business Unionism in the United States by Victoria C. Hattam