The consequences of congress's choice of delegate: Judicial and agency interpretations of title VII

Vanderbilt Law Review

Volumn 63, Issue 2, 2010, Pages 363-435

  Lemos, Margaret H ^a  

^a YESHIVA UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 78650643217     PISSN: 00422533     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (503)

1. See Steven P. Croley, Theories of Regulation: Incorporating the Administrative Process, 98 COLUM. L. REV. 1, 3 (1998) ("In many profound ways, the innumerable activities of everyday life-working, traveling, transacting, recreating, indeed eating, drinking, and breathing-are affected by the work of federal administrative agencies ..")
2. See, e.g., DAVID NSCHOEBROD, POWER WITHOUT RESPONSIBILITY: HOW CONGRESS ABUSES THE PEOPLE THROUGH DELEGATION 3-21 (1993) ("When the lawmakers we elect have others make the law, the people lose.");
3. Peter H. Aranson et al., A Theory of Legislative Delegation, 68 CORNELL L. REV. 1, 7, 63-67 (1982) (proposing to "deprive the legislature of its ability to shift responsibility and to create lotteries in private benefits through regulation" by renewing the nondelegation doctrine);
4. Ernest Gellhorn, Returning to First Principles, 36 AM. U. L. REV. 345, 352 (1987) (calling for a revival of the nondelegation doctrine and a narrower reading of the statutory authority of agencies and the President);
5. Marci A. Hamilton, Representation and Delegation: Back to Basics, 20 CARDOZO L. REV. 807, 819-20 (1999) ("[T]he nondelegation doctrine serves crucial constitutional ends when applied against delegations to the President and against delegations to administrative agencies.");
6. Gary Lawson, Delegation and Original Meaning, 88 VA. L. REV. 327, 333 (2002) (asserting that the Constitution contains a "discernable, textually grounded" nondelegation principle that is "far removed from modern doctrine");
7. David Schoenbrod, Delegation and Democracy: A Reply to My Critics, 20 CARDOZO L. REV. 731, 732 (1999) (responding to critics of his proposition that democracy suffers when Congress evades responsibility by delegating legislative powers to the executive branch).
8. See, e.g., JERRY MASHAW, GREED, CHAOS, AND GOVERNANCE 152-56 (1997) (asserting that the delegation of political authority to administrators is "a device for improving the responsiveness of government to the desires of the general electorate");
9. David Epstein & Sharyn O'Halloran, The Nondelegation Doctrine and the Separation of Powers: A Political Science Approach, 20 CARDOZO L. REV. 947, 950-51 (1999) (describing delegation as a self-regulating "balancing of competing inefficiencies," and arguing that "legislators may well delegate authority to executive actors, but they will rarely, if ever, do so without constraints");
10. Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1 J.L. ECON. & ORG. 81,100 (1985) ("[Delegation to experts [is] a form of consensus building that, far from taking decisions out of politics, seeks to give political choice a form in which potential collective agreement can be discovered and its benefits realized.");
11. David B. Spence, A Public Choice Progressivism, Continued, 87 CORNELL L. REV. 397, 398 (2002) [hereinafter Progressivism] (examining the roots of legal scholars' dissatisfaction with economic models of delegation and detailing the public choice case for delegation);
12. David B. Spence & Frank Cross, A Public Choice Case for the Administrative State, 89 GEO. L.J. 97, 101-02 (2000) (using public choice theory to argue that "agency policymaking autonomy is desirable, constitutionally valid, and practically workable").
13. See infra notes 11-15 and accompanying text.
14. This Article treats statutes that contain substantial gaps or ambiguities, and give courts primary interpretive authority to resolve those uncertainties, as delegations to courts. Some statutes in that category also give courts primary enforcement authority, in the sense that litigation in court is the only route to government-imposed sanctions for statutory violations. Judicial enforcement also involves the efforts of private parties, of course.
15. See Margaret H. Lemos, Special Incentives to Sue 6 (Cardozo Legal Studies Research Paper No. 271, 2010), available at http://papers.ssrn.com/sol3/papers. cfm?abstract-id=1474923 (discussing congressional efforts to encourage private litigation to enforce statutes). It may also involve the efforts of one or more agencies, for example if agencies have been given authority to initiate and conduct litigation to force compliance with statutory requirements or to sanction violators. Such statutes still are properly understood as judicially enforced, as any relief must come from the courts.
16. See Morris P. Fiorina, Legislative Choice of Regulatory Forms: Legal Process or Administrative Process?, 39 PUB. CHOICE 33 (1982) [hereinafter Fiorina, Choice of Regulatory Forms] (focusing on slack-minimization);
17. Morris P. Fiorina, Legislator Uncertainty, Legislative Control, and the Delegation of Legislative Power, 2 J.L. ECON. & ORG. 33 (1986) [hereinafter Fiorina, Delegation] (same);
18. Matthew C. Stephenson, Legislative Allocation of Delegated Power: Uncertainty, Risk, and the Choice Between Agencies and Courts, 119 HARV. L. REV. 1036 (2006) (focusing on stability).
19. For a rare example of commentary on delegations to courts that does not rely on formal modeling, see Eli Salzberger, A Positive Analysis of the Doctrine of Separation of Powers, or: Why Do We Have an Independent Judiciary?, 13 INT'L REV. L. & ECON. 349 (1993).
20. Civil Rights Act of 1964 §§702-718, 42 U.S.C. §§200Oe-1-200Oe-17 (2006).
21. See Fiorina, Delegation, supra note 7, at 34-35 (describing how, in the debates over early railroad legislation, the choice between delegating enforcement authority to the federal courts or to the Interstate Commerce Commission "was viewed by many as the key to [the] substance" of any regulation);
22. Stephenson, supra note 7, at 1036 ("Understanding the conditions under which a rational legislator would prefer delegation to agencies rather than courts, and vice versa, has important implications for both the positive study of legislative behavior and the normative evaluation of legal doctrine ....").
23. It bears emphasis that nothing here turns on the view that Congress always makes a purposeful choice of delegate. Through examination of the consequences of a delegation to courts, this Article illuminates some of the factors that might guide Congress's choice. If Congress does not currently consider such factors, perhaps it should-but that is a question for another day.
24. See Thomas W. Merrill, Rethinking Article I, Section I: From Nondelegation to Exclusive Delegation, 104 COLUM. L. REV. 2097, 2153-54 (2004) (discussing the common argument that the scale of modern government makes delegations necessary);
25. Richard J. Pierce, Jr., Political Accountability and Delegated Power: A Response to Professor Lowi, 36 AM. U. L. REV. 391, 404 (1987) ("Given the nature and level of governmental intervention that Congress now authorizes, it could not possibly make the hundreds, or perhaps thousands, of important policy decisions that agencies make annually.").
26. See Lisa Schultz Bressman, Chevron's Mistake, 58 DUKE L.J. 549, 568 (2009) ("The costs of writing specific legislation are high, indeed wastefully so.");
27. Fiorina, Choice of Regulatory Forms, supra note 7, at 45-46 ("In every session there are hundreds of decisions to be made; time spent on any one competes with opportunities presented by others. Moreover, time spent on decision-making may actually be politically counter-productive as editorialists and interest group spokesmen begin to complain about legislative delays, stalemate, incapacity to govern, and so forth.").
28. See Spence & Cross, supro note 3, at 135-36 ("It would place an enormous burden on Congress to evaluate all the data supplied in a typical notice-and-comment rulemaking process before an agency. The frequency of legislative hearings and the size of legislative staff would have to multiply many times over. Additionally, there is little that could be done to provide Congress with the engineering expertise of OSHA or EPA.").
29. See GRANT GILMORE, THE AGES OF AMERICAN LAW 95 (1977) ("One of the facts of legislative life ... is that getting a statute enacted in the first place is much easier than getting the statute revised so that it will make sense in light of changed conditions.").
30. See Epstein & O'Halloran, supra note 3, at 954 (noting that "one of the primary reasons for delegating" is "the ability of agencies to respond flexibly to changed conditions").
31. Cf. Fiorina, Delegation, supra note 7, at 35 ("[C]omplexity of governmental tasks is not a sufficient explanation for observed patterns of delegation, because legislatures sometimes choose to retain close control over complex policy realms such as taxation . . . while relinquishing close control over many simpler realms.").
32. See Spence, Progressivism, supra note 3, at 432 ("Slender majorities of both houses of Congress may favor legislation aimed at a new policy goal, but different subsets of those slender majorities may oppose some of the particulars in each potential approach to achieving that goal.").
33. See Salzberger, supra note 7, at 361 ("The cases in which all potential voters of a legislator unanimously support a certain arrangement are extremely rare. Usually one will find within a potential voters' group ... a subgroup that will benefit from a certain legislation and thus supports it, and another subgroup that will lose from this arrangement and will naturally oppose it.").
34. See Mark A. Graber, The Nonmajoritarian Difficulty: Legislative Deference to the Judiciary, 7 STUD. AM. POL. DEV. 3, 39 (Spring 1993) ("[M]ainstream politicians do their best to avoid taking firm public stands on those matters that internally divide their coalition.").
35. See Salzberger, supra note 7, at 365 ("One can argue that... it is best for the legislator not to regulate at all. But of course this option can impose costs in the same manner that regulating can .... Thus, not doing anything is not a solution.").
36. See Neal Devins & Michael Herz, The Battle That Never Was: Congress, the White House, and Agency Litigation Authority, 61 LAW & CONTEMP. PROBS. 205, 215-16 (1998) ("As public choice theory suggests, lawmakers often devise legislation at the behest of powerful interest groups. . . . [S]ince interest groups often compete with each other (including industry and environmentalists, unions and business), legislation is often ambiguous.");
37. Joseph A. Grundfest & A.C. Pritchard, Statutes With Multiple Personality Disorders: The Value of Ambiguity in Statutory Design and Administration, 54 STAN. L. REV. 627, 628 (2002) ("Ambiguity serves a legislative purpose. When legislators perceive a need to compromise they can, among other strategies, 'obscur[e] the particular meaning of a statute, allowing different legislators to read the obscured provisions the way they wish.' "
38. (quoting ABNER J. MIKVA & ERIC LANE, LEGISLATIVE PROCESS 779-80 (1997)));
39. Victoria F. Nourse & Jane S. Schacter, The Politics of Legislative Drafting: A Congressional Case Study, 77 N.Y.U. L. REV. 575, 595 (2002) (describing the prevalence of "deliberate [statutory] ambiguity," and reporting that congressional staffers "viewed deliberate ambiguity ... as justified by the felt need for action or the perceived threat that inflexible political positions would thwart passage of any bill at all").
40. See Bressman, supra note 12, at 568 ("Congress might aim to write just enough policy to receive a positive response for its action, while deflecting any negative attention for the burdensome details to the agency.");
41. Jonathan R. Macey & Geoffrey P. Miller, The Canons of Statutory Construction and Judicial Preferences, 45 VAND. L. REV. 647, 666 (1992) ("As interest groups have become more specialized and as more interest groups have succeeded in gaining voice in the policymaking process, consensus has become more difficult to achieve. Congress has adopted, therefore, the strategy of passing increasingly broad and amorphous enabling legislation that delegates controversial matters to administrative agencies.").
42. See generally Margaret H. Lemos, The Other Delegate: Judicially Administered Statutes and the Nondelegation Doctrine, 81 S. CAL. L. REV. 405, 428-34 (2008) (describing delegations to courts);
43. cf. Salzberger, supra note 7, at 359 (noting that, while "[a]lmost all of [the] literature" on delegations focuses on agencies, delegations also may run to the courts).
44. 15 U.S.C. §1 (2004).
45. Bd. of Trade v. United States, 246 U.S. 231, 238 (1918).
46. Texaco Inc. v. Dagher, 547 U.S. 1, 5 (2006)
47. (citing State Oil Co. v. Kahn, 522 U.S. 2, 10 (1997)).
48. Id.
49. See HERBERT HOVENKAMP, THE ANTITRUST ENTERPRISE: PRINCIPLE AND EXECUTION 39 (2005) (describing "much of the legislative history" as "useless");
50. Daniel A. Crane, Antitrust Antifederalism, 96 CAL. L. REV. 1, 5 (2008) (describing legislative history as "notoriously tortured and unhelpful").
51. IX ALEXANDER BICKEL, THE JUDICIARY AND RESPONSIBLE GOVERNMENT 1910-21, at 130 (1984) (describing the method of regulation reflected in the Sherman Act).
52. See Nat'l Soc'y of Prof 1 Eng'rs v. United States, 435 U.S. 679, 688 (1978) ("Congress ... did not intend the text of the Sherman Act to delineate the full meaning of the statute or its application in concrete situations. The legislative history makes it perfectly clear that [Congress] expected the courts to give shape to the statute's broad mandate by drawing on common-law tradition.");
53. 1 PHILIP AREEDA & HERBERT HOVENKAMP, ANTITRUST LAW ¶ 103d2 (3d ed. 2006) (stating that the Sherman Act "investfed] the federal courts with a jurisdiction to create and develop an 'antitrust law' in the manner of the common law courts");
54. Graber, supra note 19, at 50-51 (arguing that the Sherman Act was designed to push off to the courts the difficult questions of antitrust policy).
55. Civil Rights Act of 1964 §703, 42 U.S.C. §2000e-2 (2006). Notably, courts also have primary interpretive authority over all federal criminal law, even though the Department of Justice could, theoretically, perform that role.
56. See Dan M. Kahan, Is Chevron Relevant to Federal Criminal Law?, 110 HARV. L. REV. 469 (1996) (arguing for delegation of criminal law interpretation to the executive branch).
57. 15 U.S.C. §78j(b) (2006).
58. 17 U.S.C. §107 (2006).
59. 20 U.S.C. §1412 (2006).
60. 29 U.S.C. §185(a) (2006);
61. see Textile Workers Union of Am. v. Lincoln Mills of Ala., 353 U.S. 448, 450-51 (1957) (interpreting Section 301 as authorizing courts to develop a substantive law regarding enforcement of collective bargaining agreements).
62. For example, the Sherman Act is subject to public enforcement by the Antitrust Division of the Department of Justice and the Federal Trade Commission. See Lemos, supro note 23, at 462. Similarly, as described in detail below, the EEOC has authority to process, investigate, and conciliate claims of discrimination, and can initiate litigation in federal court on behalf of victims.
63. See infra notes 89, 94,106 and accompanying text.
64. See Paul Frymer, Distinguishing Formal from Institutional Democracy, 65 MD. L. REV. 125, 128 (2006) ("[E]lected officials rely on courts in a myriad of ways to conduct public policy and they frequently authorize legal activism to handle matters precisely because they are incapable of doing it themselves.");
65. Graber, supro note 19, at 44 ("Having a judiciary available to make policy decisions is a particular boon to elected officials whenever they are faced with a strong public demand that the government do something about a pressing problem, but there is no public consensus on a solution.");
66. cf. Howard Gillman, How Political Parties Can Use the Courts to Advance Their Agendas: Federal Courts in the United States, 1871-1891, 96 AM. POL. SCI. REV. 511, 511 (2002) (critiquing the scholarly tendency to "attribute judicial empowerment to factors other than the short-term self-interest of elected power-holders acting on the basis of conventional political agendas").
67. See Stephenson, supra note 7, at 1042 ("Despite the extensive positive literature on legislative delegation and the voluminous normative literature on how courts should allocate interpretive authority between themselves and administrative agencies, there has been relatively little positive analysis of the factors that would influence legislative preferences between delegating to agencies and delegating to courts.").
68. Stephenson, supra note 7, at 1043 (describing the "slack- minimization" theory of delegations as follows: "[L]egislators prefer delegation to an agency rather than a court when the ideological distance between legislator and agency is smaller than that between legislator and court").
69. See Jonathan Bendor & Adam Meirowitz, Spatial Models of Delegation, 98 AM. POL. SCI. REV. 293 (2004) (proposing a delegation model based on risk-aversion);
70. Jonathan Bendor et al., Theories of Delegation, 4 ANN. REV. POL. SCI. 235 (2001) (emphasizing the ally principle within current delegation rationales).
71. Fiorina, Delegation, supra note 7, at 39.
72. Fiorina, Choice of Regulatory Forms, supra note 7, at 57. William Landes and Richard Posner have advanced a similar theory, albeit not directed at Congress's choice of delegate.
73. See William Landes & Richard Posner, The Independent Judiciary in an Interest-Group Perspective, 18 J.L. & ECON. 875 (1975). Landes and Posner argue that an independent judiciary is valuable to Congress because courts will give effect to the "deal" struck by the enacting Congress, even if later Congresses adopt different policies.
74. Id. at 885.
75. Like Fiorina, Landes and Posner assume without explanation that courts will follow the intent of the enacting legislature. As others have pointed out, "[t]his pivotal underlying assumption is supported neither by Landes' and Posnet's own empirical findings, or by others', nor by theoretical proof." Salzberger, supra note 7, at 359.
76. See Einer Elhauge, Preference-Estimating Statutory Default Rules, 102 COLUM. L. REV. 2027, 2084 (2002) (arguing for a default rule that "dynamically tracks the enactable preferences of the current government");
77. William N. Eskridge, Jr., Reneging on History? Playing the Court/Congress/President Civil Rights Game, 79 CAL. L. REV. 613, 643-44 (1991) (arguing that courts consider the potential reactions and beliefs of Congress and the President before interpreting a statute);
78. William N. Eskridge, Jr. & John Ferejohn, The Article I, Section 7 Game, 80 GEO. L.J. 523, 549-52 (1992) (discussing the court's decision-making process in terms of anticipating subsequent congressional action);
79. see also Mario Bergara, Barak Richman & Pablo T. Spiller, Modeling Supreme Court Strategic Decision Making: The Congressional Constraint, 28 LEGIS. STUD. Q. 247, 248 (2003) (concluding that "justices do adjust their decisions to presidential and congressional preferences").
80. Although Fiorina's work is not entirely clear on this point, it appears that he is contrasting broadly worded statutes that delegate significant discretion to agencies with more specific statutes administered by courts. See Fiorina, Choice of Regulatory Forms, supra note 7, at 45 (critiquing one theory of delegations on the ground that "it does not explain why Congress ever passes a specific law rather than hand off the specifics to an agency");
81. id. at 53 (arguing that different consequences follow "[i]f the legislature writes a clear law containing the regulatory decision and charges the courts with enforcement" as opposed to "if the legislature writes a vague law and empowers an agency to interpret and enforce it"). He is not, in other words, comparing delegations to agencies to delegations to courts, but instead is comparing delegations to agencies to decisionmaking by Congress itself (with later enforcement by the courts). That frame of reference may explain why Fiorina assumes that courts will enforce-or at least try to enforce the intentions of the median legislator.
82. On the relationship between agency procedures and congressional control, see generally Lisa Schultz Bressman, Procedures as Politics in Administrative Law, 107 COLUM. L. REV. 1749, 1751-52 (2007) (analyzing the "notion that agencies are answerable to Congress" through congressional use of agency procedure);
83. Mathew D. McCubbins et al, Administrative Procedures As Instruments of Political Control, 3 J.L. ECON. & ORG. 243, 244 (1987) [hereinafter McCubbins et al., Administrative Procedures] (discussing the use of agency procedure to mitigate information disadvantages of politicians and to enfranchise important constituents in agency decision-making processes);
84. Mathew D. McCubbins et al, Structure and Process, Politics and Policy: Administrative Arrangements and the Political Control of Agencies, 75 VA. L. REV. 431, 432-33 (1989) [hereinafter McCubbins et al, Structure and Process] (discussing the use of agency procedure as an ex-ante agreement between Congress and the President to limit their ability to influence agencies).
85. See Peter H. Schuck, Delegation and Democracy: Comments on David Schoenbrod, 20 CARDOZO L. REV. 775, 785 (1999) ("The appropriations process sharply constrains the authority and discretion of agencies.").
86. See id. ("While the nature, quality, and intensity of legislative oversight vary from committee to committee, it is often used to signal congressional preferences on agency policy issues and to extract policy commitments from agency officials.").
87. See Cynthia R. Farina, Statutory Interpretation and the Balance of Power in the Administrative State, 89 COLUM. L. REV. 452, 509-10 (1989) ("Agency action can be publicly castigated on the House or Senate floor, and members of Congress or their staffs can importune agency decision makers.").
88. The President appoints agency heads (subject to the advice and consent of the Senate), and-with the exception of so-called independent agencies-can remove them from their offices. Modern presidents also have exercised control through executive orders requiring review of proposed agency actions and regulatory plans by the executive Office of Management and Budget and the Office of Information and Regulatory Affairs.
89. See Nicholas Bagley & Richard L. Revesz, Centralized Oversight of the Regulatory State, 106 COLUM. L. REV. 1260, 1263-67 (2006) (arguing that the requirements of centralized review provide the President with a "powerful tool" to shape agency policy);
90. Elena Kagan, Presidential Administration, 114 HARV. L. REV. 2245, 2281-82 (2001) (describing how President Clinton used administrative oversight to promote desired policy ends).
91. For a discussion of President Reagan's Executive Order 12,498, which was adopted until recently by subsequent Presidents, see Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory State, 62 U. CHI. L. REV. 1, 3 (1995) (describing the order's "annual regulatory plan" requirement).
92. For a discussion of President George W. Bush's Executive Order 13,422, see Peter L. Strauss, Overseer, or "The Decider"? The President in Administrative Law, 75 GEO. WASH. L. REV. 696, 701-02 (2007) (describing the order's amendments to the provisions respecting the Regulatory Policy Officers).
93. See sources cited supra note 46;
94. see also Kathleen Bawn, Political Control Versus Expertise: Congressional Choices About Administrative Procedures, 89 AM. POL. SCI. REV. 62, 62 (1995) (arguing that agency procedures may be used to maximize competence, increase political control, or achieve a combination of the two);
95. David Epstein & Sharyn O'Halloran, Administrative Procedures, Information and Agency Discretion, 38 AM. J. POL. SCI. 697, 715 (1994) (investigating the ability of Congress to limit agency discretion "when legislators have both ex post agenda control and access to information").
96. See supra notes 18-22 and accompanying text.
97. See Salzberger, supra note 7, at 365 (arguing that courts provide the best opportunity for shifting risk, or responsibility, away from Congress);
98. Stephenson, supra note 7, at 1044. It is worth noting that the President can use the veto power to prevent Congress from tossing a political hot-potato into the executive's lap.
99. Cf. Eric A. Posner & Adrian Vermeule, Interring the Nondelegation Doctrine, 69 U. CHI. L. REV. 1721, 1748 (2002) (critiquing blame-shifting arguments regarding delegations to agencies on the ground that "Congress is accountable when it delegates power-it is accountable for its decision to delegate power to the agency"). Of course, legislators should be concerned about this possibility only to the extent that they anticipate that they will still be in office when any unpopular judicial decisions are handed down.
100. See Thomas W. Merrill & Kristin E. Hickman, Chevron's Domain, 89 GEO. L.J. 833, 861 (2001) ("[F]ederal statutory programs have become so complex that it is beyond the capacity of most federal judges to understand the full ramifications of the narrowly framed interpretational questions that come before them");
101. Spence & Cross, supra note 3, at 140 ("Judges do not possess the technical expertise that justify agency delegations, and courts are the poorest of all government institutions when it comes to independent information-gathering capabilities.");
102. Stephenson, supra note 7, at 1042 ("Perhaps the most common explanation for why a legislator would prefer delegation to an agency rather than a court is that agencies have specialized expertise and better access to relevant information, and they are therefore more likely to 'get it right' than courts.");
103. Cass R. Sunstein, Beyond Marbury: The Executive's Power to Say What the Law Is, 115 YALE L.J. 2580, 2583 (2006) ("For the resolution of ambiguities in statutory law, technical expertise ... [is] highly relevant, and ... the executive has significant advantages over courts.").
104. Bernard W. Bell, Using Statutory Interpretation to Improve the Legislative Process: Can It Be Done in the Post-Chevron Era?, 13 J.L. & POL. 105,144 (1997).
105. Id.
106. Stephen Breyer, Judicial Review of Questions of Law and Policy, 38 ADMIN. L. REV. 363, 377 (1986);
107. Stephenson, supra note 7, at 1042-43.
108. BICKEL, supra note 29, at 130.
109. Breyer, supra note 58, at 397;
110. Stephenson, supra note 7, at 1042-43.
111. See Graber, supra note 19, at 39 (noting that, when courts are called upon to make policy decisions, "[j]udicial efforts to identify the policies favored by the dominant national coalition are .. . likely to prove unavailing because mainstream politicians do their best to avoid taking firm public stands on those matters that internally divide their coalition").
112. 467 U.S. 837 (1984). Under Chevron's famous two-step rule, the reviewing court first must ask whether Congress clearly expressed an intent on the "precise question at issue."
113. Id. at 842. However, if "the statute is silent or ambiguous with respect to the issue," the court must defer to the agency's answer to the question so long as it is reasonable.
114. Id. at 843.
115. Id. at 843 n.9;
116. see also id. at 843 (explaining that an agency's resolution of statutory ambiguity "necessarily requires the formulation of policy and the making of rules to fill any gap left... by Congress").
117. See supra notes 14-15 and accompanying text.
118. See Amy Coney Barrett, Statutory Stare Decisis in the Courts of Appeals, 73 GEO. WASH. L. REV. 317, 327-28 (2004) ("A majority of the circuits has explicitly adopted the super-strong presumption against overruling statutory precedents, and in those circuits that have never explicitly applied the rule, separate opinions assume that it applies.");
119. William N. Eskridge, Jr., Overruling Statutory Precedents, 76 GEO. L.J. 1361, 1362 (1988) (discussing "super strong" statutory stare decisis in the Supreme Court).
120. See Jonathan T. Molot, Reexamining Marbury in the Administrative State: A Structural and Institutional Defense of Judicial Power Over Statutory Interpretation, 96 Nw. U. L. REV. 1239, 1247 (2001) ("[J]udges . . . are subject to strong institutional norms that render judicial interpretation more stable and consistent over time than interpretation by successive political administrations ....").
121. Stephenson, supra note 7, at 1047.
122. See id. at 1058 (explaining that "[a] legislator's interest in intertemporal consistency is likely to be stronger"-and hence delegations to courts more attractive-"when compliance with a statute requires large, irreversible investments-for example, when the interpretive question involves the permissible forms of business organization or the selection of an industry-wide technological standard"). The flip side of this view is that agencies will be the more attractive option when ideological consistency across issues is more important than stability over time.
123. Id. at 1054. Thus, Congress will opt for judicial process when it wishes to diversify the risk of unfavorable decisions across different issues, and will opt for administrative process when it wishes to diversify the risk of unfavorable decisions over time.
124. Id. at 1049.
125. See, e.g., State Oil Co. v. Khan, 522 U.S. 3, 20 (1997) ("[T]he general presumption that legislative changes should be left to Congress has less force with respect to the Sherman Act... .");
126. Eskridge, supra note 65, at 1376-81.
127. See Spence & Cross, supra note 3, at 115 (noting that "the statutory mission locks in agency values over time," even as the values of the public or other political actors evolve).
128. For example, so-called independent agencies, over which the President enjoys only limited removal power, may be less likely than executive agencies to adapt their policies as presidential administrations change.
129. See Stephenson, supra note 7, at 1067 (acknowledging that the assumptions of his model might be less persuasive as applied to independent agencies).
130. See Thomas J. Miles & Cass R. Sunstein, Do Judges Make Regulatory Policy? An Empirical Analysis of Chevron, 73 U. CHI. L. REV. 823, 868-69 (2006) ("[Administrative law already ensures a high degree, and perhaps an excessively high degree, of stability. It is both time consuming and difficult to make a regulation; often the process takes two years or more. To say the least, new presidents cannot immediately change agency policy as they see fit.").
131. That is not to suggest that the choice between delegations to courts and delegations to agencies must be all-or-nothing; Congress can and sometimes does divide power between courts and agencies, for example by giving an agency primary interpretive authority but providing for enforcement through the courts. Congress also can provide for shared enforcement authority, by providing for both judicial and administrative enforcement. But interpretive authority is harder to share, as one institution or the other must have the last word in the case of disagreements as to statutory meaning.
132. As noted, agencies may differ based on the subject matter of their organic statutes and their status as primary or secondary delegates. There also may be important differences among agencies based on their preferred policymaking form-rulemaking or adjudication. Others have suggested that agencies' approaches to statutory interpretation properly may vary depending on policymaking form.
133. See Jerry L. Mashaw, Norms, Practices, and the Paradox of Deference: A Preliminary Inquiry into Agency Statutory Interpretation, 57 ADMIN. L. REV. 501, 525 (2005) (noting that "[i]t would be surprising for agency interpretive methodology to be invariant across [the] different contexts [of rulemaking and adjudication] (although it may be)");
134. Kevin M. Stack, Agency Statutory Interpretation and Policymaking Form, 2009 MICH. ST. L. REV. 225, 226 (arguing that "an agency's approach to statutory interpretation is in part a function of the policymaking form through which it acts"). It follows that the relationship between judicial and administrative process likewise may depend on whether the agency or agencies under consideration typically rely on adjudication or rulemaking to flesh out the meaning of the relevant statute(s).
135. For discussions of the legislative history, see HUGH DAVIS GRAHAM, THE CIVIL RIGHTS ERA 125-52 (1990);
136. CHARLES WHALEN & BARBARA WHALEN, THE LONGEST DEBATE: A LEGISLATIVE HISTORY OF THE 1964 CIVIL RIGHTS ACT 115-19 (1985). For a detailed account of the legislative choice between public and private enforcement of Title VII,
137. see generally Sean Farhang, The Political Development of Job Discrimination Litigation, 1963-1976, 23 STUD. AM. POL. DEV. 23, 58-60 (2009).
138. See Robert C. Lieberman, Ideas, Institutions, and Political Order: Explaining Political Change, 96 AM. POL. SCI. REV. 697, 706 (2002) ("Both Kennedy and then Lyndon Johnson needed to balance the electoral demands of Southern whites and Northern blacks, each of whom was an essential piece of the Democratic coalition. . . . Civil rights posed similar challenges and opportunities for Richard Nixon in his own presidential bids, as he sought to pry the South loose from the Democrats' grip while also competing for minority votes.").
139. See id. at 705-06 ("Ideologically, the debates over civil rights represented the culmination of a long-standing debate in American political and intellectual life between colorblind and race-conscious visions of American society.").
140. Id. at 706-07.
141. Section 703 reads in full: It shall be an unlawful employment practice for an employer (1) to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual's race, color, religion, sex, or national origin; or (2) to limit, segregate, or classify his employees in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual's race, color, religion, sex, or national origin. Civil Rights Act of 1964 §703, 42 U.S.C. §2000e-2(a) (2006).
142. Section 703(j), for example, made clear that Title VII does not require employers to adopt racial quotas; Section 703(h) carved out an exception for differential treatment based on "bona fide seniority system [s]" and authorized employers to rely on "professionally developed ability test[s]" so long as the tests were not "designed, intended, or used to discriminate"; and Section 706(g) stated that judicial relief was available only upon a finding that the employer "has intentionally engaged in or is intentionally engaging in" a prohibited practice, and warned that courts may not require employers to hire, reinstate, or promote employees who were fired or refused employment or promotion "for any reason other than discrimination on account of race, color, religion, sex, or national origin." 42 U.S.C. §2000e-2(j), e-2(h), e-5(g)(2)(A).
143. Lieberman, supra note 77, at 706-07 (citation omitted).
144. See GRAHAM, supra note 76, at 130 (describing the "pro-labor prejudice" of the early NLRB);
145. see also Farhang, supra note 76, at 24 ("[T]he 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.").
146. See HANES WALTON, JR., WHEN THE MARCHING STOPPED: THE POLITICS OF CIVIL RIGHTS REGULATORY AGENCIES 17 (1988) ("[T]he southern congressmen who opposed the creation of the new regulatory agencies felt that such action would lead to (1) bureaucratic tyranny by bigoted bureaucrats who would impose foreign social customs in the South; (2) a more powerful federal government-one that could involve itself in nearly every facet of the individual's life; and (3) a more massive and expensive federal bureaucracy.").
147. See GRAHAM, supra note 76, at 148 (describing "the Republican (and conservative Democratic) principle that in government-business relations, local primacy must prevail over Washington-knows-best"). Objections to a strong enforcement role for the EEOC also might have been linked to a Republican aversion to "big government" and higher taxes.
148. See ROBERT A. KAGAN, ADVERSARIAL LEGALISM: THE AMERICAN WAY OF LAW 50-51 (2001) ("Republicans helped craft regulatory schemes that called for private litigation, rather than public expenditure, to accomplish collective goals.").
149. H.R. REP. NO. 88-914 (1963), reprinted in 1964 U.S.C.C.A.N. 2391, 2515-16 (reasoning that employers and labor unions needed "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");
150. see GRAHAM, supra note 76, at 129-30 (noting that some members of Congress argued that "long-standing principles of American jurisprudence required that final determinations be made by the judiciary rather than by an investigative, prosecuting agency").
151. Lieberman, supra note 77, at 707.
152. To many, this represented a serious setback for the civil rights movement. See Robert Belton, A Comparative Review of Public and Private Enforcement of Title VII of the Civil Rights Act of 1964, 31 VAND. L. REV. 905, 907 (1978) ("The prevailing attitude towards Title VII... was that the civil rights movement had suffered a defeat....").
153. Civil Rights Act of 1964 §706(a), 42 U.S.C. §2000e-5(b) (2006).
154. §707(a).
155. See Albermarle Paper Co. v. Moody, 422 U.S. 405, 415 (1975) (noting that the attorneys fees provision of Title VII, §706(k) or 42 U.S.C. §2000e-5(k), serves the "strong public interest in having injunctive actions brought under Title VII" by plaintiffs acting as "private attorneys general").
156. §706(e), 42 U.S.C. §2000e-5(f).
157. §706(k), 42 U.S.C. §2000e-5(k).
158. §713(a), 42 U.S.C. §2000e-12(a).
159. See Rebecca Hanner White, The EEOC, the Courts, and Employment Discrimination Policy: Recognizing the Agency's Leading Role in Statutory Interpretation, 1995 UTAH L. REV. 51, 56 ("Title VII... expressly delegated to the agency only the power to issue procedural rules. The Supreme Court consequently has interpreted Title VII as denying the EEOC the power to engage in substantive legislative rulemaking ....").
160. For discussion of early proposals to enhance the EEOCs enforcement powers, see GRAHAM, supra note 76, at 253-54; Farhang, supra note 76, at 50-67.
161. See H.R. REP. No. 92-238 (1972), reprinted in 1972 U.S.C.C.A.N. 2137, 2144.
162. See GRAHAM, supra note 76, at 436 ("If there is any group that should not be willing to trust their rights to the federal courts ... it is the women. They have never won.").
163. See Paul Frymer, Acting When Elected Officials Won't: Federal Courts and Civil Rights Enforcement in U.S. Labor Unions, 97 AM. POL. SCI. REV. 483, 493 (2003) ("[B]y the late 1960s,.. . the AFL-CIO was lobbying Congress to shield it from Title VII lawsuits and was willing to increase the power of the EEOC as a compromise."). Unlike most civil rights groups lobbying to increase the EEOCs enforcement authority, labor unions wanted cease-and-desist authority for the EEOC instead of, rather than in addition to, the private right of action that Title VII had created.
164. See Farhang, supra note 76, at 54-55 ("[T]he AFL-CIO . . . conditioned its support for cease-and-desist authority on eliminating the private right of action.").
165. Frymer, supra note 99, at 490;
166. see also H.R. REP. No. 92-238, at 59 (1972), reprinted in 1972 U.S.C.C.A.N. 2137, 2168 (noting that the EEOC had "attained an image as an advocate for civil rights" and as a "mission" agency);
167. GRAHAM, supra note 76, at 425 (describing Justice Rehnquist's opinion that "[a]dministrative agencies . . . lack objectivity and tend to favor one or another of the groups whose interests are protected by their statute").
168. See Neal Devins, Political Will and the Unitary Executive: What Makes an Independent Agency Independent?, 15 CARDOZO L. REV. 273, 294 (1993) (noting that some civil rights interests feared that regulatory capture would reduce the efficacy of the EEOC).
169. ALFRED W. BLUMROSEN, BLACK EMPLOYMENT AND THE LAW 59 (1971).
170. GRAHAM, supra note 76, at 421.
171. Id. at 435.
172. Id. at 429.
173. Civil Rights Act of 1964, § 706(f)(1), 42 U.S.C. § 2000e-5(f)(1) (2006). The EEOC was empowered to bring suit in federal court on behalf of individual claimants, and it also took over the Attorney General's role in bringing "pattern or practice" suits against employers. § 707(c), 42 U.S.C. § 2000e-6(c).
174. § 706(f)(1), 42 U.S.C. § 2000e-5(f)(1).
175. See GRAHAM, supra note 76, at 425 ("[G]iving the ... [EEOC] authority to sue state and local governments over their employment practices would unacceptably interfere with established patterns of federalism that the Republican party had historically defended.").
176. With the help of research assistants, I identified the relevant cases by running a search in Westlaw's "SCT" database for cases involving the terms "Title VII" and/or "employment discrimination." The results of that search were, by design, overinclusive. I read all the cases and omitted those that were not decided after full briefing and oral argument, did not involve Title VII of the Civil Rights Act of 1964, or involved Title VII only tangentially.
177. See, e.g., Hetzel v. Prince William County, Va., 523 U.S. 208 (1998) (holding, in a case that happened to involve a Title VII claim, that the court of appeals violated the plaintiffs Seventh Amendment right to a jury trial by requiring the district court to enter judgment for a lesser amount than that determined by the jury).
178. See 29 C.F.R. §§ 1601-14 (2009) (detailing the agency's procedures and guidelines for issuing rules).
179. Those sources are available on the section of the EEOCs website entitled "Laws & Guidance," located at www.eeoc.gov/laws/index.cfm (last visited January 31, 2010).
180. Recall that the 1972 amendments that enabled the EEOC to seek judicial relief on behalf of the victims of discrimination limited the agency's litigation authority to the district and circuit courts. See supra note 106 and accompanying text. Like most other agencies, the EEOC cannot litigate in the Supreme Court itself but must go through the Solicitor General, who is appointed by the President and serves at his pleasure.
181. See generally Neal Devins, Unitariness and Independence: Solicitor General Control over Independent Agency Litigation, 82 CAL. L. REV. 255 (1994). Not surprisingly, the Solicitor General typically advances the administration's views, sometimes at the expense of the relevant agency.
182. See generally Margaret H. Lemos, The Solicitor General as Mediator Between Court and Agency, 2009 MICH. ST. L. REV. 185, 187 (arguing that the Solicitor General's "role in filtering the arguments presented to the Court may leave insufficient room for a distinctive agency voice").
183. See, e.g., infra note 210 and accompanying text (discussing a case where the Solicitor General essentially "overruled" the EEOC, forcing the agency to abandon its position in the briefs).
184. An example is Yellow Freight System, Inc. v. Donnelly, 494 U.S. 820 (1990), which involved the question whether federal courts have exclusive jurisdiction over Title VII claims. The EEOC did not participate in the case, and had advocated different answers to the question in previous filings.
185. See Brief of Respondent, Yellow Freight Sys., Inc. v. Donnelly, 494 U.S. 820 (1990) (No. 89-431), 1990 WL 10013147 (citing shifts in EEOC's position).
186. See supra note 81 and accompanying text (citing provisions that restrict the reach of Title VII to intentional wrongdoing by employers).
187. See infra notes 151-155 and accompanying text (discussing the Supreme Court's reliance on the EEOC's disparate impact statistics in Griggs v. Duke Power Co., 401 U.S. 424, 436 (1971)).
188. See infra notes 204-212 and accompanying text (explaining the Supreme Court's holding in United Steelworkers of America v. Weber, 443 U.S. 193 (1979), which was consistent with the EEOC's guidelines on affirmative action).
189. See Frymer, supra note 99, at 491-92 ("[C]ourts significantly rewrote the law on Title VII, getting rid of carefully placed loopholes that unions and other civil rights opponents demanded in order to pass the Act....");
190. Lieberman, supra note 77, at 705 ("[T]he act appeared explicitly to rule out an alternative race- and group-conscious approach to recognizing and remedying discrimination in the workplace.. . . And yet within 10 years of the act's passage, the United States had adopted just this approach ....");
191. Julie Chi-Hye Suk, Antidiscrimination Law in the Administrative State, 2006 U. ILL. L. REV. 405, 440 ("In practice, the EEOC's exercise of its own power suggests that it understood its role to be more proactive in the pursuit of equality than Congress ... envisioned.").
192. William N. Eskridge & Lauren E. Baer, The Continuum of Deference: Supreme Court Treatment of Agency Statutory Interpretations from Chevron to Hamdan, 96 GEO. L.J. 1083, 1205 (2008);
193. Jeffrey A. Segal, Separation-of-Powers Games in the Positive Theory of Congress and Courts, 91 AM. POL. SCI. REV. 28, 36 n.27 (1997).
194. I focus here on the Supreme Court's decisions on all 120 Title-VII related issues that the Court resolved after oral argument, rather than the ninety-eight issues on which both the Court and the EEOC took a position, in order to provide the fullest possible view of the Justices' votes on Title VII. Omitting the twenty-two issues that the Court resolved without the benefit of the agency's input would not meaningfully change the results reported in the text. The Court took a liberal position on sixty-three, or 64 percent, of the ninety-eight issues that both the Court and the EEOC decided.
195. See infra note 209 and accompanying text (discussing Sheet Metal Workers v. EEOC, 478 U.S. 421 (1986)).
196. Compare Brief of the United States as Amicus Curiae, Penn. State Police v. Suders, 542 U.S. 129 (2004) (No. 03-95), 2004 WL 121589 (arguing that constructive discharge is not a "tangible employment action" for which an employer is automatically liable even though it results from the acts of supervisory employees),
197. with 2 EQUAL EMPLOYMENT OPPORTUNITY COMM'N, EEOC COMPLIANCE MANUAL § 612.9(a) (2002) ("Respondent is responsible for constructive discharge in the same manner as it is responsible for the outright discriminatory discharge of a charging party.");
198. compare Brief for the United States as Amicus Curiae Supporting Respondent at 45-46, Watson v. Fort Worth Bank & Trust, 487 U.S. 977 (1988) (No. 86-6139), 1987 U.S. S. Ct. Briefs LEXIS 955 (arguing that disparate-impact analysis should not be applied to subjective hiring systems because subjective practices would be so impossibly difficult to validate as job-related that employers would effectively be forced to adopt quotas in order to avoid liability),
199. with 29 C.F.R. §§ 1607.4-1607.6 (1987-2009) (indicating that hiring or promotion systems that could not be formally validated could be challenged successfully under disparate impact analysis-not, as the Solicitor General contended, exempted from disparateimpact analysis for that reason);
200. compare Brief for the United States & the Equal Employment Opportunity Commission as Amici Curiae Supporting Affirmance, Ansonia Bd. of Educ. v. Philbrook, 479 U.S. 60 (1986) (No. 85-495), 1985 WL 670268 (arguing that employer need only provide a reasonable accommodation; it need not consider alternatives proposed by the employee),
201. with 29 C.F.R. § 1605.2(c)(2)(ii) (1986) ("[W]hen there is more than one means of accommodation which would not cause undue hardship, the employer . . . must offer the alternative which least disadvantages the individual with respect to his or her employment opportunities.").
202. See also Brief for the United States & the Equal Employment Opportunity Commission as Amicus Curiae at *15 n.14, Nw. Airlines, Inc. v. Transport Workers Union of Am., 451 U.S. 77 (1981) (No. 79-1086), 1980 WL 339323 ("In the court of appeals, the EEOC as an amicus curiae took the position that contribution [against a union that participated in discrimination] is available under Title VII .... As a result of this Court's invitation to participate on writ of certiorari, the Commission has reconsidered its position and, by formal vote, concluded that an implied right of contribution should not be available under . .. Title VII. . . .").
203. The EEOC also acceded to a somewhat more conservative (but still pro-claimant) position in Meritor Sav. Bank, FSB. v. Vinson, 477 U.S. 57 (1986).
204. Compare Brief for the United States & the Equal Employment Opportunity Commission as Amici Curiae, Meritor Sav. Bank, FSB v. Vinson, 477 U.S. 57 (1986) (No. 84-1979), 1985 WL 670162 (arguing that employers should be liable for hostile work environments created by supervisory employees only if they knew or should have known about the sexually offensive atmosphere),
205. with 29 C.F.R. § 1604.11(c) (1986) ("[A] n employer ... is responsible for its acts and those of its agents and supervisory employees with respect to sexual harassment regardless of whether the specific acts complained of were authorized or even forbidden by the employer and regardless of whether the employer knew or should have known of their occurrence.").
206. See, e.g., EEOC v. Commercial Office Prods. Co., 486 U.S. 107, 115-16 (1988) (explaining that "the EEOC's interpretation of ambiguous language need only be reasonable to be entitled to deference");
207. Local No. 93, Int'l Ass'n of Firefighters, AFL-CIO C.L.C. v. City of Cleveland, 478 U.S. 501 (1986) (reasoning that the EEOC's guidelines can be relied upon because they are a manifestation of experience and informed judgment);
208. Meritor, 477 U.S. at 65 (declaring that the agency guidelines, "while not controlling upon the courts by reason of their authority, do constitute a body of experience and informed judgment to which courts and litigants may properly resort for guidance . . ." (quoting Gen. Elec. Co. v. Gilbert, 429 U.S. 125, 141-42 (1976)));
209. EEOC. v. Shell Oil Co., 466 U.S. 54, 74 (1984) (explaining that "the EEOC's interpretation of its own rules is entitled to deference");
210. McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273, 279 (1976) (noting that the EEOC's interpretations of Title VII are "entitled to great deference");
211. Griggs v. Duke Power Co., 401 U.S. 424, 433-34 (1971) ("The administrative interpretation of [Title VII] by the enforcing agency is entitled to great deference.").
212. 490 U.S. 228, 250 (1989).
213. See also infra text accompanying notes 163-166 (discussing subtle differences between the EEOC's and the Court's approach to claims of sexual harassment by supervisors).
214. The most important counterexample is Griggs, in which the Court endorsed the disparate-impact theory of liability. 401 U.S. at 422. It bears emphasis that the Court there quite self-consciously followed the path marked out by the EEOC. See infra notes 154-155 and accompanying text.
215. See, e.g., Ledbetter v. Goodyear Tire & Rubber Co., 550 U.S. 618, 642 (2007) (refusing to entertain policy arguments against a strict reading of Title VII's statute of limitations, explaining that "it is not our prerogative to change the way in which Title VII balances the interests of aggrieved employees against the interest in encouraging the 'prompt processing of all charges of employment discrimination' ");
216. Faragher v. City of Boca Raton, 524 U.S. 775, 798 (1998) (refusing to adopt a blanket rule of vicarious employer liability for supervisor harassment in part because of an absence of evidence that "Congress wished courts to ignore the traditional distinction between acts falling within the scope and acts amounting to what the older law called frolics or detours from the course of employment");
217. Kremer v. Chem. Constr. Corp., 456 U.S. 461, 476 (1982) (refusing to depart from traditional rules of preclusion without explicit direction from Congress);
218. Nw. Airlines, Inc. v. Transport Workers Union of Am., 451 U.S. 77, 95-98 & n.41 (1981) (refusing to recognize a right of contribution against unions absent evidence of congressional intent, and emphasizing that "[t]he equitable considerations advanced by petitioner are properly addressed to Congress, not to the federal courts");
219. Los Angeles, Dep't of Water & Power v. Manhart, 435 U.S. 702, 721 (1978) (refusing to award class-based retroactive relief against discriminatory pension plans on the ground that "the rules that apply to these funds should not be applied retroactively unless the legislature has plainly commanded that result");
220. Trans World Airlines, Inc. v. Hardison, 432 U.S. 63, 79 (1977) ("Without a clear and express indication from Congress, we cannot agree with . . . the EEOC that an agreed-upon seniority system must give way when necessary to accommodate religious observances.");
221. Gen. Elec. Co. v. Gilbert, 429 U.S. 125, 134 (1976) (refusing to extend Title VII's prohibition on "sex" discrimination to differential treatment of pregnant employers, given Congress's failure to specify "pregnancy" as one of the protected characteristics);
222. see also Steven R. Greenberger, Civil Rights and the Politics of Statutory Interpretation, 62 U. COLO. L. REV. 37, 60 (1991) (arguing that, in civil rights cases, where the text is unclear and "where the background norm necessary to decide a case is controversial, the Court will not extend a statute beyond its uncontroversial meaning unless Congress has made that determination in the statute");
223. cf. William N. Eskridge, Jr., Politics Without Romance: Implications of Public Choice Theory for Statutory Interpretation, 74 VA. L. REV. 275, 279 (1988) ("A court is often tempted to finesse a hard interpretational choice by 'leaving it to the legislature.' ").
224. See GRAHAM, supra note 76, at 249 ("[A] broad global purpose was imputed to Congress in 1964 [by the EEOC] and was then invoked to override the limitations [of the Act]....").
225. For an argument that "purposive" statutory interpretation may be appropriate for agencies even if it is inappropriate for judges, see Michael Herz, Purposivism and Institutional Competence in Statutory Interpretation, 2009 MICH. ST. L. REV. 89, 92.
226. Jerry L. Mashaw, Agency-Centered or Court-Centered Administrative Law? A Dialogue with Richard Pierce on Agency Statutory Interpretation, 59 ADMIN. L. REV. 889, 891 (2007). Mashaw finishes the thought by noting that "courts have no parallel responsibility for implementation." Id. I am not so sure, at least when one focuses on statutes over which Congress has vested the courts with primary interpretive authority. A normative analysis of how courts should act as delegates is outside the scope of this project, however.
227. See Alfred W. Blumrosen, Strangers in Paradise: Griggs v. Duke Power Co. and the Concept of Employment Discrimination, 71 MICH. L. REV. 59, 64 & n.19 (1972) (noting the existence of various seniority systems designed by employers of previously white-only workers to keep minorities from rising above certain advancement levels).
228. See GRAHAM, supra note 76, at 248 ("High on such a list [of targets for the EEOC] were two formidable barriers to black advancement: employment tests and seniority systems.");
229. Frymer, supra note 99, at 489 ("By 1967, the seniority loophole in Title VII became a central concern among EEOC officials ....").
230. GRAHAM, supra note 76, at 252-53.
231. Civil Rights Act of 1964 § 703(h), 42 U.S.C. § 2000e-2(h) (2006).
232. GRAHAM, supra note 76, at 249-51.
233. 431 U.S. 324, 352 (1977).
234. See Am. Tobacco Co. v. Patterson, 456 U.S. 63, 75-76 (1982) (describing the holding in Teamsters).
235. Id. at 76.
236. See Brief for the Equal Employment Opportunity Commission at 25, Am. Tobacco Co. v. Patterson, 456 U.S. 63 (1982) (No. 80-1199), 1981 U.S. S. Ct. Briefs LEXIS 1791 ("Section 703(h) should not apply to a timely challenge to the post-Act adoption of an aspect of a seniority system.").
237. See Patterson, 456 U.S. at 69 ("On its face § 703(h) makes no distinction between preand post-Act seniority systems, just as it does not distinguish between pre- and post-Act merit systems or pre- and post-Act ability tests.").
238. My observations on this point are consistent with William Eskridge's finding that "the Court will sometimes refuse to interpret a statute broadly, especially when such an interpretation would represent a major policy decision that the Court would be more comfortable allowing Congress to make," William N. Eskridge, Jr., Overriding Supreme Court Statutory Interpretation Decisions, 101 YALE L.J. 331, 389 (1991), and with Einer Elhauge's claim that courts in statutory interpretation cases tend to "favor[] middle ground options," Elhauge, supra note 43, at 2081.
239. See supra note 42 and accompanying text (discussing this suggestion by Morris Fiorina).
240. Mashaw, supra note 129, at 891.
241. See generally McCubbins et al., Administrative Procedures, supra note 46 (discussing the relationship between agency procedures and congressional control).
242. See id. at 268 (explaining how legislation enacted in 1981 requires the Consumer Product Safety Commission [CPSC] to invite proposals for voluntary standards from the industry to be regulated before it may issue new rules: "If a feasible voluntary standard is proposed, the CPSC must adopt it and end its own process. CPSC can produce mandatory industry standards only if it finds that voluntary standards are unlikely to reduce risk or would not result in compliance," and must "produce 'substantial evidence' to support this conclusion.").
243. See id. at 268-69 (contrasting the Federal Food, Drug, and Cosmetic Act, which places the burden of proof on pharmaceutical manufacturers to obtain Food and Drug Administration approval of new drugs, with the Toxic Substances Control Act, which places the burden of proof on the Environmental Protection Agency to prove that a new chemical is a risk to human health or the environment).
244. Lieberman, supra note 77, at 707.
245. For a description of how the EEOC was able to impose such requirements in the face of a statutory provision exempting organizations which already were reporting to state or local fair employment practices commissions, see GRAHAM, supra note 76, at 193-97.
246. See THE STORY OF THE UNITED STATES EQUAL EMPLOYMENT OPPORTUNITY COMMISSION: ENSURING THE PROMISE OF OPPORTUNITY FOR 35 YEARS 9-10 (2000) [hereinafter STORY OF THE EEOC] (describing early record-keeping requirements and providing examples of some of the resulting data).
247. GRAHAM, supra note 76, at 244.
248. See Blumrosen, supra note 130, at 64 (observing that "[t]ests and educational requirements were adopted extensively in the early 1960's" in order to perpetuate the subordination of black workers through "seemingly neutral personnel policies");
249. cf. Gene Grove, When a "No. 2" Applies for a Job, N.Y. TIMES, Sept. 18, 1965, at SM32 (describing a job test that asked applicants for an executive job to complete the following sentence: "Crepe suzette is to pancake as Beaujolais is to blank.").
250. 42 U.S.C. § 2000e-2(h) (2006).
251. (quoting EEOC commissioner Samuel C. Jackson, EEOC vs. Discrimination, Inc., CRISIS, Jan. 1968, at 17). The view is reflected as well in the writing of Professor Alfred Blumrosen, who served as the Chief of the EEOC's Office of Conciliation from 1965 to 1967.
252. See Blumrosen, supra note 130, at 73 ("Title VII was intended as a serious response to a major social problem, and, for this reason, the [EEOC] since 1965 has attempted to make the statute effective in dealing with the social problem by giving it the broadest possible construction.").
253. 29 C.F.R. § 1607.3 (1970);
254. see Blumrosen, supra note 130, at 60 & n.5 (describing the impetus for and the original form of the EEOC guidelines, which the author helped formulate).
255. 401 U.S. 424, 436 (1971).
256. Fans and opponents of Griggs tend to agree that the decision is difficult to square with the available indications of congressional intent. See, e.g., GRAHAM, supra note 76, at 387 ("Burger's interpretation in 1971 of the legislative intent of Congress in the Civil Rights Act would have been greeted with disbelief in 1964.");
257. Frymer, supra note 99, at 491-92 (arguing that courts interpreted Title VII broadly, and contrary to legislative will). Despite the evident tension with the intentions of the enacting coalition in 1964, committee reports in both the House and Senate endorsed Griggs when Congress amended Title VII in 1972. H.R. REP. No. 92-238, 1st Sess., at 21-22 (1971), reprinted in 1972 U.S.C.C.A.N. 2137, 2156-57; S. REP. No. 92-415,1st Sess., at 14 (1971).
258. GRAHAM, supra note 76, at 430.
259. Id. at 429 (footnote omitted).
260. See Croley, supra note 1, at 162 ("[A]djudication is simply antithetical to dialogue; it provides adversaries with a chance to persuade a neutral decisionmakers [sic] to side with them.").
261. Cf. Frymer, supra note 99, at 496 (arguing that "[t]he single-mindedness with which many judges and lawyers focused on integrating unions led them to ignore less adversarial ways in which the process might have been resolved").
262. Meritor Sav. Bank FSB v. Vinson, 477 U.S. 57, 66 (1986).
263. See id. at 65 ("In concluding that so-called 'hostile environment' (i.e., non quid pro quo) harassment violates Title VII, the EEOC drew upon a substantial body of judicial decisions and EEOC precedent holding that Title VII affords employees the right to work in an environment free from discriminatory intimidation, ridicule, and insult.").
264. 29 C.F.R. § 1604.11(c) (1986) ("[A]n employer ... is responsible for its acts and those of its agents and supervisory employees with respect to sexual harassment regardless of whether the specific acts complained of were authorized or even forbidden by the employer and regardless of whether the employer knew or should have known of their occurrence.");
265. Discrimination Because of Sex Under Title VII of the Civil Rights Act of 1964, 45 Fed. Reg. 74,676 (Nov. 10, 1980) ("[T]he strict liability imposed in § 1604.11(c) is in keeping with the general standard of employer liability with respect to agents and supervisory employees.").
266. See Burlington Indus., Inc. v. Ellerth, 524 U.S. 742, 759 (1998) ("[A]lthough a supervisor's sexual harassment is outside the scope of employment because the conduct was for personal motives, an employer can be liable, nonetheless, where its own negligence is a cause of the harassment. An employer is negligent with respect to sexual harassment if it knew or should have known about the conduct and failed to stop it.").
267. See id. at 760 (finding "vicarious liability when a discriminatory act results in a tangible employment action").
268. Id. at 765;
269. Faragher v. City of Boca Raton, 524 U.S. 775, 807-08 (1998).
270. See, e.g., Pa. State Police v. Suders, 542 U.S. 129, 140-41 (2004) (grappling with the question whether a constructive discharge constitutes a "tangible employment action" and concluding that it does not).
271. Disparate-impact liability can be understood in either distributive- or corrective-justice terms. See Suk, supra note 118, at 424-26 (discussing antidiscrimination disparate-impact prohibitions in terms of corrective justice and distributive justice). The distributive-justice understanding would treat disparate-impact litigation as a means of achieving equality in the workplace whenever feasible. On that view, policies that have an adverse effect on protected groups would run afoul of Title VII unless they could be shown to be truly necessary to the functioning of the workplace. On the corrective-justice view, by contrast, disparate-impact litigation would serve to smoke out wrongdoing by identifying employment practices that appear to be neutral but really are not, because they exert costs on protected groups without any offsetting benefits. That approach would impose liability on employers only if they adopted policies that were in some respect irrational, bearing no reasonable relationship to the demands of the business. Such policies could be understood as a form of negligence.
272. Although Griggs v. Duke Power Co., 401 U.S. 424 (1971), could be read to adopt the broader view of disparate impact (no surprise, given the Court's heavy reliance on the EEOC's guidelines in that case), later cases tended strongly toward the corrective-justice model.
273. For example, under the Supreme Court's 1989 decision in Words Cove Packing Co. v. Atonio, a disparate-impact plaintiff could prevail only if she could prove that the challenged policy was essentially a pretext for discrimination-in other words, if she could prove wrongdoing on the part of the employer. 490 U.S. 642, 660. Congress overturned aspects of Wards Cove in the Civil Rights Act of 1991,
274. see infra notes 200-203 and accompanying text, but most lower courts have continued to key disparate impact liability to some evidence of irrationality or other blameworthy behavior by defendant employers.
275. See Suk, supra note 118, at 458-59 (noting that "even recent successful claims are often accompanied by some evidence in the record of intentional discrimination").
276. Both the Supreme Court and the lower courts have tended to follow a remedial approach to voluntary affirmative action. Affirmative action creates victims, who come to court alleging wrongdoing on the part of employers and unions. Those claims have a certain force: Title VII prohibits discrimination treatment based on race, and the Court long has held that the prohibition extends to discrimination against whites. See McDonald v. Santa Fe Trail Transp. Co., 427 U.S. 273, 278-79 (1976) (holding that Title VII is "not limited to discrimination against members of any particular race"). The Court has avoided that problem by focusing on a different set of wrongs-the wrongs that would be righted by affirmative action.
277. See United Steelworkers of Am. v. Weber, 443 U.S. 193, 208 (1979) (emphasizing that the affirmative action plan at issue was designed to "eliminate a manifest racial imbalance" and "d[id] not unnecessarily trammel the interests of the white employees");
278. Richard N. Appel et al., Affirmative Action in the Workplace Forty Years Later, 22 HOFSTRA LAB. & EMP. L.J. 549, 564-65 (2005) (reporting that lower courts overwhelmingly have hewed to the remedial approach suggested in Weber).
279. The Court has refused to invalidate seniority systems on the basis of discriminatory effects and absent proof of intentional wrongdoing. See supra notes 136-139 and accompanying text;
280. Belton, supra note 88, at 955 (emphasizing the Court's focus on purposeful discrimination in seniority-systems cases).
281. See, e.g., Price Waterhouse v. Hopkins, 490 U.S. 228, 252-53 (1989) (holding, in conflict with EEOC, that an employer is not liable for sex discrimination if it can show by a preponderance of the evidence that it would have made the same decision regardless of sex),
282. superseded by statute, Civil Rights Act of 1991, Pub. L. No. 102-166, 105 Stat. 1071 (codified in scattered sections of 29 U.S.C. and 42 U.S.C);
283. Ansonia Bd. of Educ. v. Philbrook, 479 U.S. 60, 70-71 (1986) (holding, in conflict with EEOC's guidelines, that an employer's duty to religious employees extends only to providing a reasonable accommodation; employers need not offer the alternative that least disadvantages the employee);
284. Ford Motor Co. v. EEOC, 458 U.S. 219, 240-41 (1982) (holding, in conflict with EEOC, that an employer can toll the accrual of backpay liability by offering the plaintiff the job she was initially denied, even though the offer does not provide for retroactive seniority);
285. Trans World Airlines v. Hardison, 432 U.S. 63, 84-85 (1977) (holding, in conflict with EEOC, that an employer did not violate its duty to provide reasonable accommodations for religious employees by refusing to go against the terms of a valid seniority system).
286. See Suk, supra note 118, at 438 ("[T]he elements of antidiscrimination law that best fit the distribute justice paradigm have emerged largely from administrative agency action ....").
287. See id. at 466 (arguing that agency regulation is more appropriate than litigation in court for addressing practices, like pollution, that have harmful effects but are not blameworthy).
288. Cf. Crane, supra note 28 (explaining that antitrust enforcement has followed a crimetort model rather than a corporate-regulatory model).
289. See McCubbins et al, Administrative Procedures, supra note 46, at 244 ("[E]lected officials can design procedures to . . . enfranchise important constituents in agency decisionmaking processes, thereby assuring that agencies are responsive to their interests.").
290. See Sean Farhang, Private Litigation, Separation of Powers, and the Struggle Over Job Discrimination Enforcement, 1981-1991, at 3 (Jan. 17, 2008) (UC Berkeley, JSP/Center for the Study of Law and Society Faculty Working Papers, available at http://repositories. cdlib.org/csls/fwp/60) ("[W]hen drafting a regulatory statute Congress, if it is going to allow private enforcement litigation at all, has wide latitude in selecting rules that substantially determine" the expected benefits and costs of litigation, as well as the plaintiffs likelihood of success);
291. Lemos, supra note 6 (explaining how statutory mechanisms like attorney's fee shifts can affect the content of courts' dockets).
292. Farhang, supra note 76, at 28.
293. Id. at 91.
294. See, e.g., Thomas W. Merrill, Judicial Deference to Executive Precedent, 101 YALE L.J. 969, 978-79 (1992) ("[A]gency decisionmaking is always more democratic than judicial decisionmaking because all agencies are accountable (to some degree) to the President, and the President is elected by the people.").
295. See Hugh Davis Graham, The Politics of Clientele Capture: Civil Rights Policy and the Reagan Administration, in REDEFINING EQUALITY 103, 106 (Neal Devins & Davison M. Douglas eds., 1998) ("Although environmental and consumer deregulation claimed priority under Reagan, the deregulatory campaign included civil rights components.");
296. see also Drew S. Days, III, The Courts' Response to the Reagan Civil Rights Agenda, 42 VAND. L. REV. 1003, 1008 (1989) ("Harking back to earlier theories of discrimination, the [Reagan] Administration sought to refocus civil rights enforcement on blatant, intentional violations of federal civil rights laws or the Constitution. . . . [T]his shift in policy had a corollary: Enforcement of civil rights laws utilizing concepts of 'discriminatory effect' or 'disparate impact' should be de-emphasized.").
297. Justice Department Seeks to Overturn Promotion Plan for New Orleans Police, Daily Lab. Rep. (BNA) No. 6, at A-8 (Jan. 10, 1983)
298. (quoting Brief of the Department of Justice Civil Rights Division, Williams v. New Orleans, 729 F.2d 1554 (5th Cir. 1984) (No. 82-3435)).
299. Indeed, Reagan was generally opposed to the use of statistics as an indicator of discrimination. During his first term, Reagan's administration pressed the OMB to stop collecting statistical data regarding racial and ethnic groups, and pressed the EEOC to stop using statistical data to determine discrimination. Consistent with that view, Chairman Clarence Thomas "told a House Subcommittee on Employment Opportunities that 'statistics have been misused to charge discrimination against employers.' Thomas testified that 'differences between the proportion of blacks, Hispanics or women at a work site and their proportion in the total work force are not proof of discrimination,' and he urged the government to stop using statistics as an indicator of possible discrimination."
300. WALTON, supra note 84, at 133
301. (quoting Juan Williams, Chairman of EEOC Tells Panel Statistics Misused to Prove Bias, WASH. POST, Dec. 15, 1984, at A4);
302. see also Juan Williams, EEOC Chief Cites Abuse of Racial Bias Criteria, WASH. POST, Dec. 4, 1984, at A13 (quoting Thomas as saying that Griggs had been "overextended and over-applied").
303. See Andrew M. Dansicker, A Sheep in Wolfs Clothing: Affirmative Action, Disparate Impact, Quotas and the Civil Rights Act, 25 COLUM. J.L. & SOC. PROBS. 1, 37 (1991) (explaining that, under Griggs, "the employer is being given a 'request' to hire more minorities, but it is so difficult for the employer to prove that she tried to hire those minorities . . . that the 'request' effectively becomes a 'demand,' and the end results are predetermined");
304. Brief for the United States as Amicus Curiae Supporting Petitioners at 25, Wards Cove Packing Co. v. Atonio, 493 U.S. 802 (1989) (No. 87-1387), 1988 WL 1026056 (arguing that a demanding business necessity test "would threaten to put pressure on employers to avoid disparate impact liability by adopting quotas or otherwise turning their attention away from job qualifications and toward numerical balance").
305. As Neal Devins and David Lewis have explained, Reagan aggressively utilized the appointments process to place ideological allies in positions of agency leadership, "vett[ing] nominees for ideological consistency and intensity" and "emphasiz[ing] the need for appointees to see themselves as part of a unitary administration." Neal Devins & David E. Lewis, Not-So Independent Agencies: Party Polarization and the Limits of Institutional Design, 88 B.U. L. REV. 459, 481 (2008).
306. See infra notes 217-218 and accompanying text.
307. See WALTON, supra note 84, at 157 (quoting the acting Chair of the EEOC from 1981-82 as stating that, "'OMB, Justice and certain recent staff appointees at EEOC have been quietly working to undermine the Employment Selection Guidelines and erode-by oratory-the Griggs Doctrine");
308. Eleanor Holmes Norton, Equal Employment Law: Crisis in Interpretation-Survival Against the Odds, 62 TUL. L. REV. 681, 707 (1988) (noting "frequent, if aborted" threats to revise the guidelines).
309. Notably, the final strategy did not require the EEOC's participation or assent. See supra note 112 (discussing Solicitor General control of litigation in the Supreme Court). Indeed, the 1980s saw a clash between the traditionally pro-claimant EEOC and the Reagan administration. In the years between 1964 and 1981, the EEOC had taken a position on thirty-six issues that reached the Court. On only two of those issues did the Solicitor General adopt a different position. During the Reagan years, the EEOC had an identifiable position on thirty-four issues that reached the Court. The Solicitor General filed a brief without the EEOC's participation that adopted a contrary position with respect to seven of those issues, and persuaded the EEOC to sign briefs representing a change of position with respect to an additional five issues.
310. These findings are consistent with those in Segal, supra note 119, at 34 (empirically testing, and rejecting, the notion that the Supreme Court shifted to the right on civil rights following the election of Ronald Reagan in 1980). William Eskridge, on the other hand, has argued that the Court shifted to the right on civil rights issues after Reagan's election, and has attributed that shift to the fact that "the Court would be protected from congressional overrides by a presidential veto, unless the Court took a position that not even a third of either chamber would accept." Eskridge, supra note 140, at 395. I did not attempt to replicate Eskridge's study, which included Title VII as well as other civil rights statutes. In the Title VII context, however, the Court's decisionmaking was if anything more liberal following Reagan's election than before. The Court rendered a liberal decision on six out of the thirteen issues it considered in the 197779 Terms (46 percent), and on twelve of the twenty-three issues it considered in the 1980-83 Terms (52 percent). But cf. Eskridge, supra note 140, at 396 n.205 (reporting that the Court rendered a liberal decision in ten of the seventeen statutory civil rights cases decided during the 1977-79 Terms, and in eleven of the thirty-four such cases decided during 1980-83).
311. Row percentages are shown in parentheses. Using a significance level of 0.05, no significant relationship exists between the direction (liberal or conservative) of the Supreme Court's decisions on Title VII-related issues and the party of the President in office at the time the decisions were made. In the absence of such a relationship, one would expect to see approximately twenty-two liberal decisions under Democratic administrations, and fifty-one liberal decisions under Republican administrations. A chi-square test with one degree of freedom indicates that there is a 0.2366 probability that random chance alone would have yielded a difference of the type witnessed between actual and expected values.
312. Interestingly, Justice Thomas's votes in Title VII cases seem notably different from his votes in other areas. Justice Thomas consistently is ranked as one of the most conservative Justices on the modern Court, yet he falls close to the middle of the range on Title VII. One possible explanation for that discrepancy is that Justice Thomas's experience at the EEOC heightened his sensitivity to, and understanding of, the many facets of employment discrimination.
313. See generally Lee Epstein et al., Ideological Drift Among Supreme Court Justices: Who, When, and How Important?, 101 Nw. U. L. REV. 1483, 1540-41 (2007) ("The ideological boxes into which Presidents, senators, and the public place Justices at the time of their nominations are not so tightly sealed.").
314. Justice O'Connor replaced Justice Stewart, who had a more conservative voting record on Title VII cases (51 percent liberal) than she would have (57 percent liberal). Justice Scalia was appointed to the Court to fill the vacancy left when Justice Rehnquist was elevated to Chief Justice to replace Chief Justice Burger. Again, Burger's voting record (38 percent liberal) was more conservative than Scalia's proved to be (43 percent liberal). The same pattern holds for Justice Kennedy (49 percent liberal) who replaced Justice Powell (45 percent liberal). The most convulsive shift was Justice Thomas's replacement of Justice Marshall. Justice Thomas has a fairly moderate voting record in Title VII cases (52 percent liberal), but Justice Marshall's voting record was one of the more lopsided (88 percent liberal).
315. Accord Matthew C. Stephenson, Mixed Signals: Reconsidering the Political Economy of Judicial Deference to Administrative Agencies, 56 ADMIN. L. REV. 657, 675 (2004) (charting the ideology of the median Justice following various membership changes on the Court, and finding a significant shift after Thomas replaced Marshall, but scant movement after the additions of O'Connor, Scalia, Kennedy, and Souter).
316. Row percentages are shown in parentheses. Using a significance level of 0.05, a statistically significant relationship exists between the party of the appointing President and the direction of a Justice's Title VII votes (liberal or conservative): Justices appointed by Democratic Presidents are significantly more likely to cast liberal votes on Title VII-related issues than Justices appointed by Republican Presidents. In the absence of a relationship, one would expect to see approximately 148 liberal votes by Justices appointed by Democratic Presidents and 491 liberal votes by Justices appointed by Republican Presidents. A chi-square test with one degree of freedom indicates that there is less than a 0.0001 probability that random chance alone would have yielded a difference as large as the one witnessed between actual and expected values.
317. See supra notes 135-139 and accompanying text.
318. See United Steelworkers of Am. v. Weber, 443 U.S. 193, 214 (1979) (finding that an employer could grant preferential treatment to racial minorities under a private, voluntary affirmative action program);
319. Griggs v. Duke Power Co., 401 U.S. 424, 436 (1971) (holding that under Title VII, if a hiring procedure disparately impacts ethnic minority groups, businesses must demonstrate that such tests are "reasonably related" to the job for which the test is required).
320. 490 U.S. 642 (1989).
321. The Court stated in Griggs that "[t]he touchstone is business necessity," 401 U.S. at 431, but it went on to explain that practices with a disparate impact on minority applications are permissible only if they "can[] be shown to be related to job performance," id., "are demonstrably a reasonable measure of job performance," id. at 436, or have "a manifest relationship to the employment in question," id. at 432. Clearly, "necess[ary]" and "related" are-or at least could be-significantly different standards. Griggs therefore spawned substantial uncertainty in the lower courts as to what "business necessity" entailed, and how it could be proved.
322. See Dansicker, supra note 182, at 16 & nn. 15-16. Subsequent decisions by the Court provided little guidance, as they appeared to waver between a strict and more lenient approach.
323. See id. at 19-20 ("[T]he Supreme Court vacillated between the manifest relationship and business necessity standards.").
324. Compare Albemarle Paper Co. v. Moody, 422 U.S. 405, 431 (1975) (holding that an employment practice with a disparate impact on minority applicants must be " 'predictive of or significantly correlated with important elements of work behavior which comprise or are relevant to the job' " (quoting 29 C.F.R. § 1607.4(c))),
325. and N.Y. Transit Auth. v. Beezer, 440 U.S. 568, 587 n.31 (1979) (concluding that because the defendant employer's goals were "significantly served by-even if they do not require-[the challenged] rule," the record "demonstrat[ed] that [the] rule bears a manifest relationship to the employment in question" (internal quotation marks omitted)),
326. with Dothard v. Rawlinson, 433 U.S. 321, 331 & n.14 (1977) (finding that the defendant prison failed to establish that the physical size of a prison guard is "necessary" or "essential" to success in the job).
327. 490 U.S. at 659 (disclaiming any requirement "that the challenged practice be 'essential' or 'indispensable' to the employer's business").
328. Id. at 659-60.
329. Brief for the United States as Amicus Curiae Supporting Petitioners, Wards Cove, 490 U.S. 642 (No. 87-1387), 1988 U.S. S. Ct. Briefs LEXIS 263, at *29.
330. S. 2104, 101st Cong. §§ 3-4 (as introduced to Senate, Feb. 7, 1990).
331. Eskridge, supra note 43, at 639.
332. The final version of the legislation passed both houses of Congress in 1990 and defined "business necessity" to mean that the challenged practice "must bear a significant relationship to successful performance of the job." S. 2104, § 3 (as passed by Senate, July 18, 1990). Although that language was similar in many respects to the Bush definition, the difference between the two proposals was enough to persuade President Bush to veto the version of the bill that Congress passed in 1990. The Senate fell one vote short of overriding the veto.
333. See Neil A. Lewis, President's Veto of Rights Measure Survives by 1 Vote, N.Y. TIMES, Oct. 25,1990, at A1.
334. Civil Rights Act of 1991, Pub. L. No. 102-166, 105 Stat. 1071 (codified in scattered sections of 42 U.S.C). As enacted, the 1991 Act shifted the burden to the defendant to "demonstrate that the challenged practice is job related for the position in question and consistent with business necessity," § 105, but did not define the key term.
335. 443 U.S. 193, 201-02 (1979).
336. Adoption of Interpretive Guidelines, 44 Fed. Reg. 4422, 4427 (Jan. 19, 1979) (codified at 29 C.F.R. § 1608.4). The Guidelines endorsed voluntary efforts at eradicating discrimination, and concluded that affirmative action is appropriate where an employer, after analyzing its employment practices, finds a reasonable basis for believing that race conscious action is required to bring it into compliance with Title VII or to remedy prior discrimination by the employer or by others.
337. 467 U.S. 561 (1984).
338. Id. at 579-80. Section 706(g) of Title VII provides that "no order of the court shall require the . . . hiring, reinstatement, or promotion of an individual as an employee, or the payment to him of any back pay, if such individual was refused ... employment or advancement or was suspended or discharged for any reason other than discrimination on account of race, color, religion, sex, or national origin." 42 U.S.C. § 2000e-5(g)(2)(A). As the Solicitor General's brief noted, the EEOC rejected the interpretation of § 706(g) pressed by the Reagan administration and adopted by the Court, reasoning that it would "call into question numerous extant consent decrees and conciliation agreements to which the EEOC is a party."
339. Brief for the United States as Amicus Curiae Supporting Petitioners, Firefighters Local Union No. 1784 v. Stotte, 467 U.S. 561 (1984) (No. 82-206), 1983 U.S. S. Ct. Briefs LEXIS 427, at *36 n.23.
340. See Joel L. Selig, The Reagan Justice Department and Civil Rights: What Went Wrong, 1985 U. ILL. L. REV. 785, 824-26 (arguing that the Reagan Administration had misinterpreted Stotts);
341. William Bradford Reynolds, The Reagan Administration and Civil Rights: Winning the War Against Discrimination, 1986 U. ILL. L. REV. 1001, 1015 (writing in reply to Selig, "Stotts appeared to us to hold that quotas or other preferential techniques, which by design benefited non-victims because of race or sex, cannot be a part of relief under Title VII"). Reynolds was the Assistant Attorney General for Civil Rights under President Reagan.
342. 478 U.S. 421 (1986).
343. Interestingly, the EEOC-which was a respondent in the case and had joined the plaintiffs throughout the litigation in arguing that numeric goals were necessary-joined the Solicitor General's brief advocating a strongly anti-affirmative action position. It appears that the Solicitor General effectively overruled the EEOC when the case made it to the Supreme Court and, thus, passed out of the hands of the EEOC and into the hands of the Attorney General. See Devins, supra note 112, at 299-300 ("[W]hen the EEOC explained its position to Solicitor General attorneys, it was flatly told that it... would have to swallow DOJ opposition to affirmative action.").
344. 480 U.S. 616, 641-42 (1987).
345. See Brief for the United States as Amicus Curiae Supporting Petitioner, Johnson v. Transp. Agency, 480 U.S. 616 (1987) (No. 85-1129), 1986 WL 728148, at *13-16.
346. See Reynolds, supra note 208, at 1019 (describing Johnson as "a significant setback").
347. See Graham, supra note 179, at 108 (noting that "appointing youthful conservatives to the federal courts[] was a priority of the Reagan revolution" and that, "[b]y the end of his eight years in the White House, Reagan had filled 338 judgeships on the district and appeals courtsabout half the total.");
348. Stephenson, supra note 191, at 669 (reporting that "the federal circuit courts were liberal from 1977 until about 1985, conservative from 1990 to 1996, and relatively moderate in the 1986-89 and the 1997-2002 periods").
349. See Norton, supra note 185, at 702-03;
350. Selig, supra note 208, at 821-23;
351. see also STORY OF THE EEOC, supra note 148, at 16 (describing successful EEOC effort to obtain a consent decree against the Illinois Central Gulf Railroad (among other employers) that mandated "hiring at a rate 25 percent above the representation of each minority group in the labor force until goals were reached in specific job categories").
352. See supra notes 151-155 and accompanying text;
353. STORY OF THE EEOC, supra note 148, at 31 ("EEOC began the 1980s by continuing to focus on broad, systemic employment practices that operated to discriminate against large classes of individuals. However, the Republican appointees to the Commission wanted to reassess its methods.").
354. STORY OF THE EEOC, supra note 148, at 31-32;
355. see Graham, supra note 179, at 106 (noting that the EEOC "shifted enforcement emphasis from class-action proceedings to conciliation and lawsuits seeking make-whole relief for identified victims of discrimination," and that "the shift in regulatory strategy was accomplished not through formal, notice-and-comment procedures but through internal policy directives").
356. See MICHAEL E. MILAKOVICH & GEORGE J. GORDON, PUBLIC ADMINISTRATION IN AMERICA 373 (10th ed. 2009) ("Ronald Reagan, from the very start of his presidency, used a comprehensive assault on the national government budget as the key to his attempt to reshape the national bureaucracy. Reagan demonstrated convincingly that the most direct way (if not always the easiest politically) to control an agency is to cut-or increase-its budget.");
357. STORY OF THE EEOC, supra note 148, at 45 (explaining that budget cuts forced the agency "to undergo severe staff reductions, leading to an ever larger charge backlog");
358. Graham, supra note 179, at 106 (noting that the Reagan administration "slowed regulatory activity by cutting the agency budgets");
359. see also Norton, supra note 185, at 706 (noting a 70 percent drop in cases filed by the EEOC in court during the 1980s);
360. B. Dan Wood, Does Politics Make a Difference at the EEOC?, 34 AM. J. POL. SCI. 503, 522-23 (1990) (finding a sharp drop in EEOC enforcement under Reagan).
361. STORY OF THE EEOC, supra note 148, at 58;
362. see also Presa Release, E.E.O.C., EEOC Chairman and Commissioners Expected to Take Decisive Actions to Reinvigorate Charge Processing (Apr. 18, 1996), available at http://www.eeoc.gov/eeoc/newsroom/release/archive/4-18-95. html (announcing changes for processing increased charge filings).
363. See Norton, supra note 185, at 706-07. Concededly, the fact that the Court has the last word as to the meaning of Title VII may mean that the history recounted here understates the degree of presidential influence over agencies' interpretive decisions. Had the EEOC's guidelines on employee selection and affirmative action operated with the force of law, Reagan may well have pressed harder for revisions, and those efforts may have met with more success than was possible under Title VII.
364. See Angel Manuel Moreno, Presidential Coordination of the Independent Regulatory Process, 8 ADMIN. L.J. AM. U. 461, 469 (1994) (explaining that, while "no uniform definition of independent agency appears to exist in American scholarship," most definitions "emphasiz[e] that the president cannot remove .. . commissioners");
365. Alan B. Morrison, How Independent are Independent Regulatory Agencies?, 1988 DUKE L.J. 252, 252 ("[A]n independent agency is one whose members may not be removed by the President except for cause, rather than simply because the President no longer wishes them to serve,");
366. see also Strauss, supra note 50, at 717 n,99 (arguing that the President's power to remove independent agency heads for "cause" would not permit removal on the ground that the agency did not interpret the statute as the President wished).
367. See Linda R. Cohen & Matthew L. Spitzer, Judicial Deference to Agency Action: A Rational Choice Theory and an Empirical Test, 69 S. CAL. L. REV. 431, 448 (1995) (noting that, due to their exemption from the requirement of OMB review, "independent agencies d[o] not need to be as immediately attentive to changes in the executive branch as [are] executive agencies"). Predictably, congressional delegations to independent agencies tend to increase during periods of divided government.
368. See DAVID EPSTEIN & SHARYN O'HALLORAN, DELEGATING POWERS: A TRANSACTION COST POLITICS APPROACH TO POLICY MAKING UNDER SEPARATE POWERS 146-47, 158 (1999) (arguing that Congress is less likely to delegate at all during periods of divided government, and, when it does delegate, more likely to choose independent agencies than executive agencies subject to greater presidential control);
369. see also Devins & Lewis, supra note 183, at 464 (explaining that Congress is more likely to delegate authority to independent commissions when they "fear the administrative influence of the current President on policies post-enactment"). Congress also may be more likely to delegate enforcement authority to the courts during times of divided government.
370. In particular, although the President's removal power is constrained with respect to independent agencies, he retains a great deal of leverage through the appointment power. See B. Dan Wood & Richard W. Waterman, The Dynamics of Political Control of the Bureaucracy, 85 AM. POL. SCI. REV. 801, 822 (1991) (finding that "agency outputs shifted immediately after a change in agency leadership" in the EEOC and four other agencies, and concluding that "political appointment... is very important").
371. See generally Devins & Lewis, supra note 183 (discussing the Presidential appointment process). The same is true of federal courts, but the number of judges is much larger and the rate of turnover much slower, making it more difficult for a President quickly to change the makeup of the federal judiciary.
372. See Sean Farhang, Public Regulation and Private Lawsuits in the American Separation of Powers System 8 (Sept. 14, 2007) (Berkeley JSP/Ctr. for Study of Law & Soc'y Faculty Working Papers, available at http://escholarship.org/uc/ item/8m59b989) ("The operation of economic incentives on private litigants and lawyers in statutorily constructed enforcement markets creates an enforcement apparatus with an autopilot character, substantially beyond the reach of presidential influence . . . ."). In this empirical study of legislation designed to facilitate private lawsuits to enforce statutory provisions-i.e., legislation that creates a private right of action and provides for attorney's fees for prevailing plaintiffs and/or enhanced damages-Farhang found that "conflict between the executive and legislative branches causes Congress to rely more heavily upon the mobilization of private litigants for regulatory enforcement."
373. Id. at 33. Although Farhang focuses on congressional empowerment of potential litigants rather than empowerment of courts, it is hard to have the former without the latter. The key point for our purposes is that legislators concerned about the possibility of presidential influence on regulatory policy may prefer "private enforcement" through the courts over an enforcement scheme that leaves all enforcement power in the hands of an agency.
374. In an empirical study of congressional overrides of Supreme Court statutory decisions, William Eskridge found an average of ten overrides per Congress, or five per year. Eskridge, supra note 140, at 338.
375. In an empirical study of the Supreme Court's application of stare decisis to statutory precedents, William Eskridge identified fifty cases decided between 1961 and 1987 in which the Court explicitly or implicitly overruled prior statutory interpretations, for an average of approximately two per year. Eskridge, supra note 65, at 1427-34.
376. See Robert Pear, Congress Passes Bill With Protections for Disabled, N.Y. TIMES, Sept. 18, 2008, at A21 (describing a recent amendment to the Americans with Disabilities Act that overturns several Supreme Court decisions and specifies that "'[t]he definition of disability in this act shall be construed in favor of broad coverage'").
377. For example, the Civil Rights Act of 1991 contained a provision limiting the legislative history that courts could use to supply meaning to the term "business necessity." Pub L. No. 102166, § 105(b), 105 Stat. 1071, 1075 (1991) (codified as amended at 42 U.S.C. 1981 (2006)).
378. The only authorized source of legislative history is an interpretive memorandum stating that "[t]he terms 'business necessity' and 'job related' are intended to reflect the concepts enunciated by the Supreme Court" in Griggs and other cases "prior to" Wards Cove. 137 CONG. REC. 28,680 (daily ed. Oct. 25, 1991) (statement of Sen. Danforth).
379. The Democrats took control of the Senate on May 24, 2001, when Senator Jeffords switched parties, but lost it again in the 2002 elections.
380. One study has found that the number of lawsuits filed by the EEOC dropped dramatically in the mid-1990s, apparently as a result of the "Republican revolution" of 1994. David Hedge & Renee J. Johnson, The Plot That Failed: The Republican Revolution and Congressional Control of the Bureaucracy, 12 J. PUB. ADMIN. RES. & THEORY 333, 344-46 (2002). The cause of the dip in enforcement is uncertain, however, because the numbers rebounded in 1996 and 1997, without a shift in party control of Congress. As the authors acknowledge, "[i]t is difficult to know whether [President Clinton's appointment of a new Chairman] or the waning influence of the Republican Revolution had a stronger influence in reinvigorating the activities of the EEOC, but the change in the EEOC's agenda is clear."
381. Id. at 344.
382. See, e.g., Cohen & Spitzer, supra note 222, at 445 (describing the median Justice at the time of Reagan's inauguration as "a moderate/conservative");
383. Stephenson, supra note 191, at 676 ("The 1977-1981 period was the most liberal [Supreme] Court in the sample. The Court became somewhat more conservative in the 1982-1990 period, and it became sharply more conservative from 1991 to 1993-the most conservative Court in the sample. The Court became somewhat more liberal in the 1994-2002 period, but was not as liberal as it had been in 1991 or before.").
384. Row percentages are shown in parentheses. Using a significance level of 0.05, a statistically significant negative relationship exists between Democratic control of the House of Representatives and liberal decisionmaking by the Supreme Court. In the absence of a relationship, one would expect to see approximately sixty-three liberal decisions under Democratic control and ten liberal decisions under Republican control. A chi-square test with one degree of freedom indicates that there is a 0.0498 probability that random chance alone would have yielded a difference as large as the one witnessed between actual and expected values.
385. Row percentages are shown in parentheses. Using a significance level of 0.05, a statistically significant negative relationship exists between Democratic control of the Senate and liberal decisionmaking by the Supreme Court. In the absence of a relationship, one would expect to see approximately forty-one liberal decisions under Democratic control and thirty-two liberal decisions under Republican control. A chi-square test with one degree of freedom indicates that there is a 0.0428 probability that random chance alone would have yielded a difference as large as the one witnessed between actual and expected values. Because Democrats controlled the House at every time they controlled the Senate, the results for Democratic control of House and Senate are the same as the results for Democratic control of the House.
386. See Civil Rights Act of 1991, Pub. L. No. 102-166, 105 Stat. 1071 (codified in scattered sections of 29 U.S.C. and 42 U.S.C.)
387. (overriding Lorance v. AT&T Tech., Inc., 490 U.S. 900 (1989);
388. Martin v. Wilks, 490 U.S. 755 (1989);
389. Wards Cove Packing Co. v. Antonio, 490 U.S. 642 (1989);
390. Price Waterhouse v. Hopkins, 490 U.S. 228 (1989);
391. and Library of Cong. v. Shaw, 478 U.S. 310 (1986));
392. Pregnancy Discrimination Act, Pub. L. No. 95-555, 92 Stat. 2076 (1978) (codified at 42 U.S.C. § 2000e(k))
393. (overriding Gen. Elec. Co. v. Gilbert, 429 U.S. 125 (1976)). The EEOC had taken a position contrary to the Court's in all the relevant cases except Shaw, in which the Court held that Title VII does not waive the federal government's immunity to paying interest. 478 U.S. at 322. The EEOC had not taken any position on that issue. Congress, especially the Democrat-controlled House, also made active use of oversight hearings during the Reagan and first Bush administration in an effort to correct what legislators saw as a troubling shift in enforcement by the EEOC
394. See Farhang, supra note 175, at 24-25 (reporting that between 1983 and 1991, "Democratic chaired congressional committees conducted no less than 15 oversight hearings examining various aspects of EEOC enforcement efforts").
395. See generally McCubbins et al., Administrative Procedures, supra note 46, at 246-53 (describing means of congressional control of agency policymaking).
396. See id. at 253-64 (describing ex ante means of control).
397. On the many forces that can conspire to block override legislation, see, for example, McNollgast, Legislative Intent: The Use of Positive Political Theory in Statutory Interpretation, LAW & CONTEMP. PROBS., Winter 1994, at 3, 11;
398. Mark Seidenfeld, Bending the Rules: Flexible Regulation and Constraints on Agency Discretion, 51 ADMIN. L. REV. 429, 482 (1999);
399. Kenneth A. Shepsle & Barry R. Weingast, Structure-Induced Equilibrium and Legislative Choice, 37 PUB. CHOICE 503, 513-14 (1981).
400. McCubbins et al, Administrative Procedures, supra note 46, at 252.
401. Cf. McCubbins et al, Structure and Process, supra note 46, at 444-45 (recognizing that the procedural mechanisms Congress can use to control agency decisionmaking before it happens are not available where courts are concerned). Courts might also be more likely to reach seriously unpopular decisions than agencies. As discussed in the previous Section, the adversarial system tends to create well-defined winners and losers rather than forging compromise among competing groups. Agencies, by contrast, may be designed by Congress so as to reflect "a propensity to find compromise, so that in the end the participants will have a blunted incentive to take further political action to alter the policy outcome."
402. McCubbins et al., Administrative Procedures, supra note 46, at 256.
403. See supra note 53 and accompanying text.
404. See JEB BARNES, OVERRULED? LEGISLATIVE OVERRIDES, PLURALISM, AND CONTEMPORARY COURT-CONGRESS RELATIONS 90 (2004) (finding that eighty-nine of 100 randomly sampled overrides followed judicial dissensus).
405. Eskridge, supra note 140, at 344 tbl.4.
406. See Dansicker, supra note 182, at 3 (describing the issue of racial quotas, brought to the fore in the debates leading up to the 1991 overrides of Words Cove and other unpopular judicial decisions, as a "political minefield").
407. Recent controversy over the Supreme Court's decision in Ledbetter v. Goodyear Tire & Rubber Co., 550 U.S. 618, 632 (2007),
408. including the override legislation, Lilly Ledbetter Fair Pay Act of 2009, Pub. L. No. 111-2, 123 Stat. 5 (codified as amended at 29 U.S.C.A. §§ 626, 633a, 794a, and 42 U.S.C.A. §§ 2000a, 2000e-5, 2000e-16 (West 2009)), provides an additional example.
409. Some scholars have used that insight as a way of explaining delegations to agencies. The key premise is that Congress will tend to delegate to agencies the details of statutes that generate diffuse benefits and concentrated costs-e.g. environmental legislation. Such statutes make Congress appear to be addressing the problem, but the agency is left in charge of (potentially critical) details, and can respond to the narrow interests of the regulated community in a way that is less visible to the public. See NEIL K. KOMESAR, IMPERFECT ALTERNATIVES: CHOOSING INSTITUTIONS IN LAW, ECONOMICS, AND PUBLIC POLICY 95-96 (1994) (arguing that politicians can declare support for a policy but quietly undermine it by pressuring the implementing agency). The same logic works to explain delegations to agencies with respect to statutes that benefit powerful, concentrated interests: The agency can hand out the benefits and Congress can avoid the blame.
410. See Aranson et al., supra note 2, at 37-62 (arguing that delegations to agencies facilitates the collective provision of private goods);
411. Eskridge, supra note 126, at 289 ("In conflictual demand situations (concentrated cost measures), legislators will often seek to delegate regulation of the group to an agency. If the legislation distributes benefits at the expense of a concentrated group, the cost payers will tend, over time, to organize themselves effectively to influence the agency.").
412. Cf. McCubbins et al, Structure and Process, supra note 46, at 443-44 (explaining how Congress can design agency procedures to ensure that administrators are responsive to the same groups that the legislation was designed to benefit).
413. But see Salzberger, supra note 7, at 368 (noting in passing that agencies are "more influenced by interest groups" than courts).
414. The foundational work on the subject is MANCUR OLSON, THE LOGIC OF COLLECTIVE ACTION: PUBLIC GOODS AND THE THEORY OF GROUPS (1965). For a sampling of the literature that has built on Olson's insights as applied to agencies,
415. see Croley, supra note 1, at 12-25, 34-56, and the sources cited therein.
416. But see Croley, supra note 1, at 55 ("Numerous scholars have concluded that the empirical evidence often offered in support of the public choice theory is at best inconclusive and at worst inconsistent with the theory.");
417. Jerry L. Mashaw, Explaining Administrative Process: Normative, Positive, and Critical Stories of Legal Development, 6 J.L. ECON. & ORG. (SPECIAL ISSUE) 267, 280 (1990) ("[T]he empirical record of [public choice theory] is one that should induce the utmost caution in its practitioners.").
418. For an overview of the "capture" literature,
419. see Bagley & Revesz, supra note 50, at 1284-92.
420. Others have sought to show that agencies do not, in fact, appear to be captured by interest groups in any consistent or predictable way. See generally, e.g., PAUL J. QUIRK, INDUSTRY INFLUENCE IN FEDERAL REGULATORY AGENCIES 175-77 (1981) (finding no empirical support for the notion that agencies are systematically biased in favor of business interests);
421. Mark Kelman, On Democracy-Bashing: A Skeptical Look at the Theoretical and "Empirical" Practice of the Public Choice Movement, 74 VA. L. REV. 199, 236-68 (1988) (exposing tensions between public choice theory and the actual performance of administrative agencies).
422. Cf. Croley, supra note 1, at 142 ("[T]he regulatory regime's legal process rules do not seem very well designed to facilitate regulatory rent-seeking by special interest groups.").
423. See Richard A. Posner, Theories of Economic Regulation, 5 BELL J. ECON. & MGMT. SCI. 335, 351 (1974) (arguing that judges are more insulated from interest groups than agencies).
424. See Eskridge, supra note 126, at 305 (arguing that life tenure frees judges from the influence of interest groups).
425. See, e.g., id. at 315 (advocating "a more aggressive approach to statutory interpretation" to ameliorate public choice dysfunctions);
426. Jonathan R. Macey, Promoting Public-Regarding Legislation Through Statutory Interpretation: An Interest Group Model, 86 COLUM. L. REV. 223, 227 (1986) (arguing that courts should interpret statutes to enforce their public-regarding purposes, thereby "transforming statutes designed to benefit narrow interest groups into statutes that in fact further the public's interests"). Others have argued that the characteristics that give certain groups a leg up in the political arena also will aid them in court.
427. See generally KOMESAR, supra note 244, at 123-50 (examining the advantages and disadvantages of the adjudicative process);
428. Einer R. Elhauge, Does Interest Group Theory Justify More Intrusive Judicial Review?, 101 YALE L.J. 31, 66-87 (1991) (arguing that the litigation process is subject to interest group pressure);
429. Paul H. Rubin, Common Law and Statute Law, 11 J. LEGAL STUD. 205, 206-07, 217 (1982) (likening spending on lobbying to spending on litigation, especially for cases that have precedential value).
430. See Graham, supra note 179, at 113 ("In the new civil rights regulation of the 1960s,... capture took a different twist. Routinely, newly created regulatory offices, such as the EEOC,... were dominated not by the employers and organizations being regulated but by representatives of the constituencies being served.").
431. See Eskridge, supra note 126, at 320 (acknowledging that public choice theory does not satisfactorily explain the enactment of the Civil Rights Act of 1964);
432. Jonathan Turley, Transnational Discrimination and the Economics of Extraterritorial Regulation, 70 B.U. L. REV. 339, 376-77 (1990) ("Although the Civil Rights Act took many years to enact, few organizations actually lobbied in support of the legislation. Of the major groups lobbying Congress on behalf of the legislation, most were ideological groups organized with civil rights, labor, religious, legal, or political interests. Arguably, the civil rights groups and unions had an economic interest in the legislation, but all of these groups were large national organizations with overtly ideological agendas.").
433. See, e.g., RUSSELL HARDIN, COLLECTIVE ACTION 103-12 (1982) (discussing moral motivations for contributing to a collective good);
434. Elhauge, supra note 251, at 43 ("Others have .. . demonstrate[d] that noneconomic factors such as altruism and ideology play at least some role in political participation and decisionmaking, and that the preferences of regulators and the general public sometimes prevail over the preferences of interest groups." (citing sources)).
435. JAMES Q. WILSON, POLITICAL ORGANIZATIONS 34 (1973) (arguing that individuals derive solidarity benefits-"rewards created by the act of associating"-and expressive benefits"rewards that derive from a sense of satisfaction at having contributed to the attainment of a worthy cause"-from group membership);
436. see also Croley, supra note 1, at 20-21 (discussing "benefits of participation itself");
437. John Mark Hansen, The Political Economy of Group Membership, 79 AM. POL. SCI. REV. 79, 79-82 (1985) (stressing the variety of benefits, both selective and collective, tangible and intangible, that individuals receive from group membership).
438. See Robert C Lieberman, Private Power and American Bureaucracy: The EEOC and Civil Rights Enforcement 21 (unpublished manuscript, on file with author) ("Within days after the EEOC began its operations, NAACP General Counsel Robert L. Carter wrote to EEOC Chairman Franklin D. Roosevelt Jr. expressing the desire to cooperate with the commission in the enforcement of Title VII and requesting a meeting to discuss 'ways and means that we in the Association may best work with you in the ordinary processing of complaints and in their submission and disposition by the Commission.' ").
439. See id. at 21-33.
440. It is not clear that the EEOC's pro-claimant leanings are caused by interest group influence as opposed to merely correlated with it. Indeed, I argued above that the EEOC's procedures may have been a critical factor in cementing the agency as an advocate for the victims of discrimination. See supra Part IV.A. It also bears emphasis that the individuals who would be likely to seek out employment at an agency like the EEOC can be expected to agree with and desire to further the agency's mission.
441. See ANTHONY DOWNS, INSIDE BUREAUCRACY 107 (1967) (arguing that bureaucrats' views are "based upon a 'biased' or exaggerated view of the importance of their own positions 'in the cosmic scheme of things' ");
442. Spence & Cross, supra note 3, at 115 ("We can expect agency bureaucrats to have values that are consistent with the agency's mission; otherwise, the bureaucrats initially would not be attracted to work in the agency.").
443. Concededly, the interest group dynamics apparent in Title VII may be a function of the enforcement scheme that Congress created. If the EEOC had been vested with direct enforcement authority, we might expect to see more lobbying of the agency by business groups and the like.
444. See Farhang, supra note 76, at 59 (quoting Jack Greenberg, Director-Counsel of the NAACP: "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."). According to Greenberg, the private right of action "kept the EEOC honest because it didn't want to be shown up by private enforcers."
445. Id.
446. Stephenson, supra note 7, at 1047.
447. The Court purported to rely on stare decisis more than any other source of statutory meaning, including the text of the statute (which played a substantial role in the Court's reasoning on 53 of the 120 Title VII issues the Court resolved).
448. See, e.g., Faragher v. City of Boca Raton, 524 U.S. 774 (1998);
449. Johnson v. Transp. Agency, Santa Clara County, Cal., 480 U.S. 616 (1987).
450. The EEOC's Guideline on Employment Selection Procedures did undergo changes prior to Reagan's administration, but the amendments do not appear to have been politically motivated. First issued in 1966, the guideline set forth the EEOC's view (later endorsed by the Supreme Court) that an employer violates Title VII if it uses a test or other screening system for selecting employees that excludes disproportionate numbers of minority applicants, unless the employer can demonstrate that the test is job-related. 29 C.F.R. § 1607 (1970). The guideline prescribed a slew of requirements employers must satisfy in order to "validate" tests as jobrelated.
451. See Ronald B. Rubin, Note, The Uniform Guidelines on Employee Selection Procedures: Compromises and Controversies, 28 CATH. U. L. REV. 605, 610-14 (1978) (describing EEOC's guidelines and the validity requirement). Several other federal agencies adopted similar-but not identical-requirements for employers in other contexts.
452. See, e.g., 41 C.F.R. §§ 60-3.1 to 60-3.18 (1976) (Department of Labor guidelines); 5 C.F.R. §§ 300.101 to 300.104 (1976) (Civil Service Commission guidelines). After several failed attempts to settle on a uniform set of guidelines, the agencies in 1978 adopted the Uniform Guidelines on Employee Selection Procedures, which replaced all pre-existing federal employment selection requirements. The Uniform Guidelines cut back on the EEOC's own guidelines in several respects, primarily by liberalizing the means by which employers could validate challenged tests as job related.
453. See Rubin, supra, at 620-30 (describing the differences between the Uniform Guidelines and the prior EEOC's guidelines).
454. For example, the EEOC originally interpreted Title VII's prohibition on sex discrimination to permit claims of wage discrimination only to the extent that such claims would be available under the Equal Pay Act. 29 C.F.R. § 1604.7 (1966) (explaining that "the standards of 'equal pay for equal work' set forth in the Equal Pay Act for determining what is unlawful discrimination in compensation are applicable to Title VII"). In 1972, and without explanation, the agency deleted that limitation from its guidelines. See 37 Fed. Reg. 6837 (1972). When the Supreme Court confronted the issue in 1981, the EEOC took the position that Title VII wage discrimination claims are not limited to those permitted by the Equal Pay Act.
455. Washington v. Gunther, 452 U.S. 161, 162 (1981);
456. Brief for the United States and the Equal Employment Opportunity Commission as Amici Curiae, Washington, 452 U.S. 161 (No. 80-429), 1981 U.S. S. Ct. Briefs LEXIS 1052. The EEOC's approach to pregnancy discrimination followed a similar pattern. Shortly after Title VII was enacted, the EEOC took the position that it did not require employers to provide benefits for pregnancy under the same terms provided for other disabilities.
457. See Gen. Elec. Co. v. Gilbert, 429 U.S. 125, 142-43 (1976) (discussing the EEOC's early and subsequently conflicting interpretations). Again, the EEOC backed away from that opinion in subsequent years, and its 1972 guidelines flatly prohibited discrimination against pregnant employees. 29 C.F.R. § 1604.10 (1973).
458. Stephenson recognizes that the assumptions of his model-particularly the assumption that agency decisionmaking will be more variable over time than judicial decisionmaking-are less realistic as applied to independent agencies. Stephenson, supra note 7, at 1067.
459. Stephenson's model does not address this issue because it assumes that a decision by either court or agency will be rendered within the first time period.
460. Id. at 1053 & n.67 (acknowledging that the assumption detracts from the realism of the model).
461. See Peter L. Strauss, One Hundred Fifty Cases Per Year: Some Implications of the Supreme Court's Limited Resources for Judicial Review of Agency Action, 87 COLUM. L. REV. 1093, 1121 (1987).
462. See Faragher v. City of Boca Raton, 524 U.S. 775, 807-10 (1998) (setting out rules to govern employer liability for supervisor harassment);
463. Burlington Indus., Inc. v. Ellerth, 524 U.S. 742, 764-66 (1998) (same);
464. Tomkins v. Pub. Serv. Elec. & Gas Co., 568 F.2d 1044, 1048-49 (3d Cir. 1977) (holding that employers are liable for supervisor harassment when they have "actual or constructive knowledge" of the problem and do not "take prompt and appropriate remedial action");
465. Barnes v. Costle, 561 F.2d 983, 993 (D.C. Cir. 1977) (holding that employers are liable for supervisor harassment except when "a supervisor contravene[s] employer policy without the employer's knowledge and the consequences are rectified when discovered"). The level of confusion in the lower courts is evidenced by the Court's observation in Ellerth that the lower court, which had considered the issue en banc, "produced eight separate opinions and no consensus for a controlling rationale." 524 U.S. at 749.
466. As explained in the previous Section, Congress adjusted the contours of disparateimpact liability in the 1991 Act. But Congress left the key term "business necessity" undefined. See supra note 205 and accompanying text. The Supreme Court has never clarified matters, so the uncertainty remains. Although the EEOC's guidelines long have provided detailed requirements for test validation-which is to say, justification under the business necessity test, see 29 C.F.R. § 1607,-the lower courts continue to disagree on the appropriate contours of the test.
467. See Christine Nardi, Comment, When Health Insurers Deny Coverage for Breast Reconstructive Surgery: Gender Meets Disability, 1997 WIS. L. REV. 777, 802 n.153 (noting that "courts are split as to whether an employer must provide a legitimate business reason or a compelling justification" (citations omitted)).
468. Stephenson, supra note 7, at 1056.
469. For example, prior to the Supreme Court's decision in General Electric Co. v. Gilbert, the lower courts uniformly had held that employers would violate Title VII if they offered benefits for some disabilities but not for pregnancy. 429 U.S. 125, 147 (1976) (Brennan, J., dissenting). In order to comply with that rule, employers would have to reduce benefits for all employees or expand existing benefits to cover pregnancy. After Gilbert, employers were free to discriminate against pregnant employees, but the freedom lasted only until the enactment of the Pregnancy Discrimination Act, which returned the law to the pre-Gilbert status quo. Congress recognized the potential costs to employers and provided a 180-day window during which employers could adjust their benefit systems without running afoul of the newly amended law. Pub. L. No. 95-555, § 2(b), 92 Stat. 2076 (1978) ("The provisions of the amendment. . . shall not apply to any fringe benefit program or fund, or insurance program which is in effect on the date of enactment of this Act until 180 days after enactment of this Act.").
470. See Stephenson, supra note 7, at 1059 (discussing both inter-issue and interjurisdictional inconsistency).
471. See Strauss, supra note 266, at 1095, 1105-06.
472. Eskridge, supra note 140, at 373-74;
473. see also Jerry L. Mashaw, Norms, Practices, and the Paradox of Deference: A Preliminary Inquiry into Agency Statutory Interpretation, 57 ADMIN. L. REV. 501, 509 (2005) ("Courts repeatedly suggest that interpretation designed to lend coherence to the general legal order is one of their most important responsibilities as custodians of the rule of law.");
474. Molot, supra note 66, at 1298-99 (describing a "hermeneutic tradition . . . which distinguishes legal reasoning from political choice [and] which values the consistent application of interpretive strategies across cases").
475. See, e.g., Burlington N. & Santa Fe Ry. Co. v. White, 548 U.S. 53, 66 (2006) (relying on NLRA analogy to support the conclusion that Title VII prohibits a wide body of retaliatory behavior);
476. EEOC v. Arabian Am. Oil Co., 499 U.S. 244, 251-52 (1991) (relying on NLRA analogy to support the conclusion that Title VII does not apply extraterritorially);
477. Lorance v. AT&T Tech., Inc., 490 U.S. 900, 909 (1989) (relying on case construing NLRA as authority for the proposition that Title VII's statute of limitations runs from the date a discriminatory seniority system is adopted, not when the employee feels the negative effects);
478. Price Waterhouse v. Hopkins, 490 U.S. 228, 249-50 (1989) (relying on NLRA cases to support the conclusion that employers can escape liability under Title VII by showing that they "would have made the same decision in the absence of the unlawful motive");
479. Hishon v. King & Spalding, 467 U.S. 69, 76 n.8 (1984) ("The meaning of this analogous language [in the NLRA] sheds light on the Title VII provision at issue here.");
480. Ariz. Governing Comm. for Tax Deferred Annuity and Deferred Comp. Plans v. Norris, 463 U.S. 1073, 1091 n.21 (1983) (using analogy to NLRA to support the conclusion that employers are responsible for retirement benefits even if they are provided by third parties);
481. Ford Motor Co. v. EEOC, 458 U.S. 219, 226 n.8 (1982) ("The principles developed under the NLRA generally guide . . . courts in tailoring remedies under Title VII. Therefore, throughout this opinion we refer to cases decided under the NLRA as well as under Title VII." (citation omitted));
482. Zipes v. Trans World Airlines, Inc., 455 U.S. 385, 395 n.11 (1982) ("Because the time requirement for filing an unfair labor practice charge under the [NLRA] operates as a statute of limitations subject to recognized equitable doctrines . . . , the time limitations under Title VII should be treated likewise.");
483. McDonnell Douglas Corp. v. Green, 411 U.S. 792, 803-04 (1973) (relying on NRLA analogy to conclude that Title VII does not compel an employer to rehire an employee who engaged in deliberate and unlawful conduct against it).
484. See, e.g., Kolstad v. Am. Dental Ass'n, 527 U.S. 526, 537 (1999) (relying on ADEA analogy to support the conclusion that intentional discrimination may not give rise to punitive damages liability if the employer mistakenly relies on a statutory exception or defense);
485. Burlington Indus., Inc. v. Ellerth, 524 U.S. 742, 754-57 (1998) (reasoning from general agency principles as applied in state tort cases and by federal courts under the Federal Tort Claims Act to conclude that sexual harassment by a supervisor is not conduct within the scope of employment);
486. Int'l Union v. Johnson Controls, Inc., 499 U.S. 187, 201-03 (1991) (observing that the Court has read the bona fide-occupational-qualifications exception to the ADEA "just as narrowly" as the equivalent exception to Title VII and reasoning by analogy from a case applying the ADEA);
487. EEOC v. Commercial Office Prods. Co., 486 U.S. 107, 123-24 (1988) (relying on an ADEA case as support for the proposition that a claimant's failure to file a claim within the state limitations period does not render his claim untimely);
488. Int'l Bhd. of Teamsters v. United States, 431 U.S. 324, 339 (1977) (borrowing from the reasoning in jury-selection cases to support conclusion that statistical proof can be used to establish a prima facie case of discrimination);
489. Gen. Elec. Co. v. Gilbert, 429 U.S. 125, 145 (1976) (relying on reasoning in a constitutional case to conclude that differential treatment of pregnant employees does not constitute discrimination because of sex).
490. Cf. Belton, supra note 88, at 955
491. (arguing that the Court's focus on purposeful discrimination in International Brotherhood of Teamsters v. United States, 431 U.S. 324, is analogous to its approach to discrimination under the Equal Protection Clause).
492. Nicholas S. Zeppos, The Use of Authority in Statutory Interpretation: An Empirical Analysis, 70 TEX. L. REV. 1073, 1093 (1992).
493. 447 U.S. 807, 826 (1980).
494. See id. at 812.
495. Id.
496. Id. at 826.
497. Id. (acknowledging that "the interests of justice might be served in this particular case by a bifurcated construction of [the] word").
498. Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 856 (1984).
499. Id. at 863.
500. Id. at 863-64 ("[T]he agency, to engage in informed rulemaking, must consider varying interpretations and the wisdom of its policy on a continuing basis.");
501. see also Mashaw, supra note 273, at 509 (suggesting that "an agency interpretive posture that seeks to harmonize its actions with the whole of the legal order risks forgetting that agencies are created precisely to carry out special purpose missions").
502. See, e.g., Faragher v. City of Boca Raton, 524 U.S. 775, 805-06 (1998) (using the reasoning of Albemarle to support the adoption of an affirmative defense to liability for supervisory sexual harassment for employers who can show that they have taken steps to try to prevent such harassment);
503. Albemarle Paper Co. v. Moody, 422 U.S. 405, 417 (1975) (explaining, in the context of an application of disparate-impact theory, that the "primary objective" of Title VII is not to provide redress but to avoid harm).