nyamaimule

Electronic copy available at: http://ssrn.com/abstract=1375684

1

THE GENETIC INFORMATION NONDISCRIMINATION ACT OF 2008:

A CASE STUDY OF THE NEED FOR BETTER CONGRESSIONAL

RESPONSES TO FEDERALISM JURISPRUDENCE

Harper Jean Tobin1

Abstract The Genetic Information Nondiscrimination Act of 2008 (GINA) is the first new civil rights statute enacted since the “federalism revolution” of 1995-2001, in which the Supreme Court announced new limitations on congressional authority. Among other things, these decisions invalidated civil rights remedies against states, declaring that Congress had failed to amass sufficient evidence of the need for legislation. Although passed in the shadow of these decisions, GINA’s limited legislative history makes it vulnerable to attack – potentially limiting its protections for millions of state employees. States will likely attack GINA on two grounds: first, that Congress relied only on its commerce power, and not its Fourteenth Amendment remedial power; and second, that Congress failed to identify a sufficient threat to constitutional rights to justify subjecting states to suit. While there are strong grounds for rejecting these challenges, that outcome is far from certain. The risk of invalidation might have been minimized had Congress developed the rationale for GINA’s extension to the states more thoroughly, or alternatively, required states to waive their immunity as a condition of federal grants. These strategies are illustrated by recent proposed civil rights legislation addressing sexual orientation discrimination and racial profiling, as well as by the Voting Rights Act Reauthorization Act of 2006, which is currently before the Supreme Court. To ensure the efficacy of future civil rights legislation, Congress should consistently tailor laws to withstand federalism challenges. Future laws should expressly invoke Congress’s authority and intent to create remedies against states; be accompanied by a strong and targeted legislative record; expressly require waiver of state immunity; and specifically enumerate remedies.

Introduction

In spring of 2008, more than a decade after its initial introduction, Congress

passed the Genetic Information Nondiscrimination Act of 2008 (GINA) with near-

1 Staff Attorney, Herbert Semmel Federal Rights Project, National Senior Citizens Law Center. My colleagues, Simon Lazarus, Ian Millhiser, and especially Rochelle Bobroff provided invaluable comments on this Article, as did Mark Posner and Jennifer Mathis. I am grateful to Christina Crosby, Rebecca Sickenberger, and Lindsay Ruffner for their research and editorial assistance.

Electronic copy available at: http://ssrn.com/abstract=1375684

2

unanimity. Designed to promote genetic research and preventive screening, safeguard

medical privacy, and prevent unfair treatment of individuals based on disease-linked

traits, GINA prohibits the collection and use of genetic information by employers and

insurers. Like other antidiscrimination laws, GINA (which will go into effect in late

2009) applies to private and public employers alike, and enables individuals harmed by

discrimination to seek damages in court. Yet because of GINA’s limited legislative

record, courts could severely limit the new law’s provision for damages actions against

states, cutting back the new law’s protections for more than five million state workers.2

Although Congress surely did not intend this result, it could almost certainly have

prevented it.

The threat to GINA arises from the Supreme Court’s “federalism revolution” of

1995-2001.3 In that period, the conservative majority of the Supreme Court under Chief

Justice Rehnquist took greater strides in limiting the power of Congress than at any time

since the 1930s, invalidating parts of major federal laws on the basis of newly-announced

constitutional rules derived from the Tenth,4 Eleventh and Fourteenth Amendments5 and

the Commerce Clause,6 and purportedly intended to preserve an appropriate balance of

state and federal power. Notable among these were decisions in 2000 and 2001 holding

that two landmark civil rights laws, the Age Discrimination in Employment Act (ADEA)

2 See U.S. CENSUS BUREAU, STATE GOVERNMENT EMPLOYMENT DATA: MARCH 2007 (2009), available at http://ftp2.census.gov/govs/apes/07stus.txt (showing nearly 3.8 million full-time employees and more than 1.4 million part-time employees at that time). 3 See, e.g., Erwin Chemerinsky, The Federalism Revolution, 31 N.M. L. REV. 7, 30 (2001).  4 See Printz v. United States, 521 U.S. 898 (1997). 5 See discussion infra Part I. 6 See United States v. Lopez, 514 U.S. 549 (1995); United States v. Morrison, 529 U.S. 598 (2000).

3

and the Americans with Disabilities Act (ADA), were unconstitutional insofar as they

provided damages remedies against state employers.7

At the beginning of the Roberts Court era, the full implications of these Rehnquist

Court federalism decisions are uncertain, and are being litigated extensively in the lower

courts.8 The Court’s most recent decisions in these areas rejected challenges to

Congressional power,9 but most were decided on narrow grounds.10 Notably, in 2004 and

2006, the Court permitted enforcement of the ADA against states in very specific classes

of cases, leaving undecided the question of whether the ADA’s application to public

services and programs is within Congress’s Fourteenth Amendment power in the

majority of cases.11 At the same time, the Court has indicated it may limit rights and

remedies under a variety of laws enacted under Congress’s spending power.12

These Rehnquist and Roberts Court federalism decisions create standards against

which any major new Congressional action is likely to be judged. Regardless of which

party controls the political branches, these judicial constraints remain, and could limit the

effectiveness of not only existing laws but future legislation as well. Ironically, even as

the Supreme Court places less reliance on legislative history in interpreting statutes than

7 See discussion infra Part I. 8 See, e.g., Rochelle Bobroff, Scorched Earth and Fertile Ground, 2007 CLEARINGHOUSE REV. 298 (discussing lower court treatment of the ADA); Tanya K. Shunnara, Reaction to Raich: The Commerce Clause Counter-Revolution, 37 CUMB. L. REV. 575 (2007) (discussing lower court treatment of Commerce Clause decisions). 9 The 2005 decision in Gonzales v. Raich, 545 U.S. 1 (2005), which upheld federal regulation of the intrastate production of drugs under the Controlled Substances Act, is seen by some as a retreat from sweeping dicta in the Rehnquist Court’s earlier Commerce Clause decisions. See, e.g., Lino A. Graglia, Lopez, Morrison & Raich: Federalism in the Rehnquist Court, 31 HARV. J.L. & PUB. POL'Y 761 (2008); Simon Lazarus, Federalism R.I.P.? Did the Roberts Hearings Junk the Rehnquist Court’s Federalism Revolution?, 56 DEPAUL L. REV. 1 (2006). 10 See Cent. Va. Cmty. Coll. v. Katz, 546 U.S. 356 (2006) (upholding liability of states under bankruptcy laws). 11 See discussion infra Part I. 12 See discussion infra Part III.

4

in the past,13 it has given exacting scrutiny to such history in interpreting Congress’s

power to enact laws, in effect “regularly check[ing] Congress's homework.”14 As the first

new civil rights law enacted since the Court’s “federalism revolution” began15, the

possibility of a federalism challenge to GINA illustrates the continuing challenge this

jurisprudence poses for Congress.

Part I of this Article summarizes in relevant detail the Rehnquist Court’s

splintered, fact-bound and often contradictory decisions on Congress’s power to protect

constitutional rights under Section 5 the Fourteenth Amendment, including its authority

to subject states to private actions for damages. Part II asks how GINA stacks up under

these precedents, concluding that GINA’s damages remedy may be vulnerable to attack

in suits against states. State will argue that the legislative record supporting GINA, and

particularly its prohibition on employment discrimination, is insufficient to comport with

the Court’s precedents. Damages against states are most likely to be upheld in GINA

suits that seek to protect medical privacy rights, or deter race or gender discrimination.

To some extent, the Supreme Court may have doomed GINA’s application to

states by creating constitutional rules that render it difficult, perhaps impossible, for

Congress to respond to emerging threats to constitutional rights. At the same time, it is

clear that Congress did not do all it could to ensure that GINA’s remedies would be

upheld. Part III compares GINA to three other recent pieces of legislation that illustrate

more deliberate responses to the Court’s rulings. In the Voting Rights Act

13 See, e.g., James J. Brudney & Corey Ditslear, Liberal Justices’ Reliance on Legislative History: Principle, Strategy, and the Scalia Effect, 29 BERKELEY J. EMP. & LAB. L. 117, 133-38 (2008). 14 Tennessee v. Lane, 541 U.S. 509, 558 (2004) (Scalia, J., dissenting). 15 While the Voting Rights Act Reauthorization Act of 2006, discussed infra Part III, was the first significant civil rights law to follow the 1990s federalism cases, GINA is the first since that time to create new private causes of action, including against states. Prior to GINA, the most recent laws to create new equal employment rights were the Family and Medical Leave Act of 1993, discussed infra Part I, and the Uniformed Services Employment and Reemployment Rights Act of 1994.

5

Reauthorization Act of 2006 – the constitutionality of which is currently before the

Supreme Court – Congress conducted extensive fact-finding to justify the legislation. The

proposed End Racial Profiling Act takes a similar approach, with statutory findings that

specifically justify regulation of state governments in the name of a variety of

constitutional rights. A different approach is illustrated by the proposed Employment

Non-Discrimination Act, which would require states to waive their immunity from claims

of sexual orientation discrimination in order to remain eligible for federal grants. These

comparisons show that Congress has been inconsistent in addressing the impact of the

Court’s rulings, giving them careful attention in drafting some legislation -- particularly

high-priority or controversial legislation – but little attention in other legislation such as

GINA.

Part IV concludes by suggesting how lawmakers can respond more effectively to

the Court’s federalism jurisprudence. Congress (and policy advocates) should

consistently take the Court’s rulings into account when crafting new civil rights

legislation, by (1) expressly invoking Congress’s authority and intent to create remedies

against states; (2) developing a strong legislative record that focuses on threats to

constitutional rights; (3) laying out arguments for why legislation is needed and complies

with the Court’s precedents; (4) expressly requiring waiver of state immunity in exchange

for federal funds; and (5) specifically enumerating individual remedies. Because even

these steps may not always be adequate, however, Congress should also use its oversight

and confirmation powers and other means to promote a judicial approach that gives more

respect to the constitutional authority and prerogatives of Congress.

I. The Supreme Court’s Section Five Jurisprudence

6

The Intersection of the § 5 Power and Sovereign Immunity

The constitutional questions facing GINA arise from the intersection of two of the

Court’s federalism doctrines. The first is state “sovereign immunity” under the Eleventh

Amendment, which the Rehnquist Court interpreted expansively as a general bar to

damages suits against states.16 In 1996, the Court’s five most conservative justices held –

overruling a six-year-old precedent – that Congress cannot use its commerce power, or its

other Article I powers, to abrogate states’ sovereign immunity.17 The same majority later

held that this immunity extends to all claims for damages, in both federal and state court

as well as in federal agency proceedings.18 The four dissenters characterized this

approach to sovereign immunity as a “shocking…affront to a coequal branch of our

Government,”19 and one without basis in precedent or the constitutional design.20 While

this immunity does not bar injunctive relief, the elimination of damages remedies – such

as back pay for illegal firings – erodes a law’s remedial and deterrent effects.21

The second relevant doctrine is the Rehnquist Court’s restrictive interpretation of

Congress’s authority to protect constitutional rights under section 5 of the Fourteenth

16 The Eleventh Amendment provides that states cannot be sued in federal court by citizens of other states or foreign countries. U.S. Const. Am. XI. But the Rehnquist majority held that “the scope of the States’ immunity from suit is demarcated not by the text of the Amendment alone but by fundamental postulates implicit in the constitutional design.” Alden v. Maine, 527 U.S. 356, 728-29 (1999). 17 Seminole Tribe of Fla. v. Florida, 517 U.S. 44 (1996), overruling Pennsylvania v. Union Gas Co., 491 U.S. 1 (1989). See also Fla. Prepaid Postsecondary Educ. Expense Bd. v. Coll. Sav. Bank, 527 U.S. 627 (1999) (rejecting Patent Clause as basis for abrogation); but see Central Va. Comm. Coll. v. Katz, 546 U.S. 356 (2006) (recognizing Bankruptcy Clause as a narrow exception). The Seminole Tribe majority consisted of Chief Justice Rehnquist, the recently-appointed Justice Thomas, and Justices O’Connor, Kennedy and Scalia. Prior to these decisions, the Court had struggled for decades to define the scope of states’ immunity from suit and legislative exceptions to it. See generally LAURENCE H. TRIBE, 1 AMERICAN CONSTITUTIONAL LAW 522-55 (3d ed. 2000). 18 Alden, 527 U.S. 706; Fed. Mar. Comm’n v. South Carolina State Ports Auth., 535 U.S. 743 (2002). 19 Seminole Tribe, 517 U.S. at 78 (Stevens, J., dissenting). 20 Id. at 100-169 (Souter, J., dissenting). 21 See, e.g., Pamela S. Karlan, The Irony of Immunity: The Eleventh Amendment, Irreparable Injury, and Section 1983, 53 STAN. L. REV. 1311, 1326 (2001) (immunity “deprives private parties of an adequate remedy at law for conceded violations of their rights”); Carlos M. Vazquez, Eleventh Amendment Schizophrenia, 75 NOTRE DAME L. REV. 859, 875 (2000) (noting that availability of only prospective relief can lead to numerous, temporary violations of federal law).

7

Amendment.22 The seeds of this approach lay in the Court’s 1997 decision in City of

Boerne v. Flores, which was primarily a case about the bounds of the Free Exercise

Clause and Congress’s power (or rather, its lack of power) to expand the substantive

scope of constitutional protections. 23 Boerne also held, with relatively little discussion,

that the Religious Freedom Restoration Act (RFRA) was not a proper use of Congress’s §

5 power to remedy and deter constitutional violations. The Court stated generally that

under § 5 “there must be a congruence between the means used and the ends to be

achieved,” and observed that Congress had not documented any instance of deliberate

religious discrimination in state law in modern times.24 But the “lack of support in the

legislative record… [was] not RFRA's most serious shortcoming,” the Court said,

because in general “it is for Congress to determine the method by which it will reach a

decision.”25 Rather, RFRA failed as a § 5 remedy because it was clearly intended to alter

existing constitutional protections, and “not designed to identify and counteract state laws

likely to be unconstitutional.”26 Despite the limited nature of the § 5 analysis in Boerne,

however, the same five-justice majority that expanded the Court’s immunity doctrine

would also employ Boerne’s “congruence and proportionality” formulation as a sharp

limitation on congressional power.27

22 In previous decades, the Supreme Court broadly construed Congress’s power to remedy and deter constitutional violations under § 5 of the Fourteenth Amendment. See Fitzpatrick v. Bitzer, 427 U.S. 445 (1976); South Carolina v. Katzenbach, 383 U.S. 301 (1966); Oregon v. Mitchell, 400 U.S. 112 (1970); City of Rome v. United States, 446 U.S. 156 (1980); Fullilove v. Klutznick, 448 U.S. 448 (1980). 23 521 U.S. 507, 530 (1997). That the Court did not think it was breaking new ground regarding the § 5 power is apparent from the fact that all of the four concurring and dissenting opinions in the case are focused on the substantive scope of the Religion Clauses. Id. at 536-67. 24 Id. at 530. 25 Id. at 531-32. 26 Id. at 534. 27 See Rebecca E. Zietlow, Federalism’s Paradox: The Spending Power and Waiver of Sovereign Immunity, 37 WAKE FOREST L. REV. 141, 153 (2002). Regarding RFRA’s successor statute, the Religious Land Use and Institutionalized Persons Act, see discussion infra Part III.

8

These two doctrines interlock, because the Court’s immunity decisions left § 5 of

the Fourteenth Amendment as the only basis for Congress to abrogate state sovereign

immunity, 28 and thereby enable damages actions under generally-applicable civil rights

laws like GINA.29 The rest of this Part lays out the principles from recent case law under

which GINA’s abrogation of sovereign immunity will likely be tested. These cases have

been decided by shifting majorities over fierce dissents, and display “major

methodological contradictions.”30 Earlier cases appear to set up a strict test, to which later

cases, decided by different majorities, arguably pay only lip service. The Court may well

swing back in the other direction in the future – especially given the replacement of the

swing voters in those cases by the conservative Chief Justice Roberts and Justice Alito31

– and the Court’s impending decision in Northwest Austin Municipal Utility District

Number One (NAMUDNO) v. Mukasey, a challenge to the Voting Rights Act

Reauthorization Act of 2006, presents an opportunity for the Court to do so.32 For present

purposes, I seek to synthesize these conflicting decisions, acknowledging their

considerable indeterminacy and the fact that the Supreme Court may change the game yet

again.33

28 Along with the similar enforcement clauses of the Thirteenth and Fifteenth Amendments. 29 Seminole Tribe, 517 U.S. at 58, citing Fitzpatrick v. Bitzer, 427 U.S. 445 (1976). As noted supra, injunctive remedies are available even in the absence of abrogation. Additionally, as discussed infra Part III, Congress may induce states to waive their immunity in exchange for federal funds. 30 Vikram David Amar, The New “New Federalism,” 6 GREEN BAG 349, 351 (2003). 31 Rochelle Bobroff, The Early Roberts Court Attacks Congress’s Power to Protect Civil Rights, 30 N.C. Cent. L. Rev. 231, 262 (2008); see also Araiza, supra note 66, at 88 (suggesting, prior to 2005 appointments, that whether Lane marks a real shift or only a temporary pause in the Court’s limitations on § 5 authority will depend on changes in the Court’s composition). 32 Northwest Austin Mun. Util. Dist. No. 1 v. Mukasey, 573 F.Supp.2d 221 (D.D.C. 2008), probable jurisdiction noted, 129 S.Ct. 894 (2009). As discussed infra Part III, the case presents the question whether the 2006 reauthorization of the VRA’s preclearance provisions was supported by a sufficient factual predicate. 33 Two points of § 5 doctrine are clear and likely to stay that way: in 2006, the Court unanimously held that Congress may always provide remedies in cases involving actual constitutional violations. United States v. Georgia, 546 U.S. 151 (2006). Under Georgia, courts must decide “on a claim-by-claim basis” whether a

9

Congressional Intent and the Source of Authority for Abrogation

Before deciding whether Congress has properly employed its § 5 power to

abrogate state immunity, it must be clear that Congress intended to abrogate immunity.

The Court has said that this clear-intent rule is generally satisfied when a statute, by its

plain terms, applies to state as well as non-state actors.34 The Court has also said that

Congress does not need to identify the source of constitutional authority for passing § 5

legislation.35 Rather, it is only necessary that a court “be able to discern some legislative

purpose or factual predicate that supports the exercise of that power.”36

Yet the Court has made one statement that seemed to contradict this principle. In

Florida Prepaid Postsecondary Education Expense Board v. College Savings Bank, the

Court refused to even consider whether a patent law was a proper use of the § 5 power to

enforce the Takings Clause of the Fifth Amendment, even though the law expressly

provided for suits against states.37 Noting the absence of any mention of the Fifth

Amendment in the statute or legislative history, the Court stated that because “Congress

was so explicit about [relying on] its commerce power and [its power under § 5 to

protect] due process guarantees as bases for the Act,” Congress’s failure to mention the

violation of a purported § 5 statute would also violate the Constitution, in which case further § 5 analysis is unnecessary. Id. at 160. The Court is also unanimous in the view that § 5 legislation may not respond to purely private action, though it is split on whether Congress has the power to provide remedies against private actors in response to states’ failure to protect them. Compare U.S. v. Morrison, 529 U.S. 598, 624- 27 (2000), with id. at 663-66 (Breyer, J., dissenting). 34 Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 73-74 (2000). See also Dellmuth v. Muth, 491 U.S. 223, 233 (1989) (Scalia, J., concurring) (statutory text need not make “explicit reference to state sovereign immunity or the Eleventh Amendment”). 35 See Fullilove v. Klutznick, 448 U.S. 448, 476-478 (1980) (upholding minority business set-aside under § power in absence of recitation). See also Woods v. Cloyd W. Miller Co., 333 U.S. 138, 144 (1948) (the “constitutionality of action taken by Congress does not depend on recitals of the power which it undertakes to exercise”). 36 EEOC v. Wyoming, 460 U.S. 226, 243 n. 18 (1983). The Wyoming Court specifically rejected the suggestion that such a recitation was required by language in Pennhurst State Sch. v. Halderman, 451 U.S. 1, 15 (1981) (“we should not quickly attribute to Congress an unstated intent to act under its authority to enforce the Fourteenth Amendment”). 37 Fla. Prepaid, v. College Savings Bank, 527 U.S. 627, 642 n. 7 (1999).

10

Fifth Amendment precluded the Court from even considering it.38 Some lower courts

have read the footnote as establishing a broad rule that reliance by Congress on one or

more powers to enact a law precludes the Court from upholding it on the basis of some

other power.39

But there is good reason to read this footnote narrowly. Its briefly stated

conclusion, without citation, should not likely be taken to limit or overrule long-

established principles. Moreover, in the very next term after writing this footnote, the

Court rejected an invitation to extend it. Asserting immunity from liability under the Age

Discrimination in Employment Act, the defendant seized on the Florida Prepaid

footnote, arguing the ADEA could not be upheld as § 5 legislation because its legislative

record and statutory findings focused entirely on interstate commerce effects and did not

invoke the Equal Protection Clause.40 This argument received substantial discussion in

the briefs,41 yet the Court did not even address it in its decision in Kimel v. Florida Board

of Regents, instead holding that the ADEA clearly sought to abrogate immunity and

evaluating the statute as § 5 legislation.42 This strongly suggests that the Court viewed

Prepaid as distinguishable.

38 Id. 39 See Chavez v. Arte Publico Press, 204 F.3d 601, 604 (5th Cir. 2000) (suggesting reliance on commerce power precludes analysis of Copyright Remedy Act under § 5); De Romero v. Inst. of Puerto Rican Culture, 466 F.Supp.2d 410, 416 (D.Puerto Rico 2006) (same); Chittister v. Dep’t of Cmty. & Econ. Dev., 226 F.3d 223, 228 n. 3 (3d. Cir. 2000) (Alito, J.) (refusing to consider Family and Medical Leave Act as remedy for due process rights in light of Congress’s reliance on Equal Protection Clause). 40 Brief for Respondents, 1999 WL 631661, 28-30. 41 See Brief for Petitioners, 1999 WL 503876, 29 n.18 (stating general rule that recitals are unnecessary); Brief for United States, 1999 WL 513848, 18 n.18 (same); Reply Brief for Petitioners, 1999 WL 728345, 7- 8 (characterizing footnote as reflecting a “rule of judicial deference” that “cannot…be leveraged into an affirmative judicial requirement that Congress must state the constitutional predicate of its legislation at the pain of having the courts declare the enactment unconstitutional”); Reply Brief for the United States, 1999 WL 33609325, 17 n.16 (distinguishing Florida Prepaid on ground that the ADEA’s nature as an antidiscrimination statute). 42 Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 73-74 (2000).

11

The best explanation for distinguishing Prepaid comes from the Solicitor

General’s brief in Kimel, which argued that that in Prepaid the Court simply deferred to

Congress’s statements regarding its bases for legislation because no other constitutional

basis was obvious from the legislation itself. By contrast, “the connection between [an]

anti-discrimination statute and the enforcement of the Equal Protection Clause is

obvious.”43 This reading reconciles Prepaid not only with Kimel but with earlier case

law, so that any nondiscrimination law that expressly applies to states – including GINA

– may be evaluated as § 5 legislation.

Legislative Record: the Strict Kimel/Garrett Standard

Under the post-Boerne cases, the first step for determining the validity of § 5

legislation is whether Congress identified a sufficient threat to constitutional rights to

justify a congressional response. In Florida Prepaid, the Court’s majority indicated that

Congress must identify a pattern of “widespread and persisting deprivation of

constitutional rights”44 – language Boerne had used to describe the basis for 1960s civil

rights laws, but which were now framed as the operative § 5 standard. The Court held

that the Patent Remedy Clarification Act did not meet this standard because it was

enacted “in response to a handful of instances of state patent infringement that do not

necessarily violate the Constitution.”45 The Fourteenth Amendment, the Court stated,

cannot be used to address such a “speculative harm.”46

43 Reply Brief for the United States, 1999 WL 33609325, 17 n.16. See also CSX Transp., Inc. v. NY State Office of Real Prop. Servs., 306 F.3d 87, 98 (2d Cir. 2002) (abrogation under Railroad Revitalization and Regulatory Reform Act valid despite express reliance on Commerce Clause and failure to mention § 5 power, where “the discriminatory conduct that the [statute] attempts to regulate can be easily associated with Section 5 powers”). 44 Fla. Prepaid Postsecondary Educ. Expense Bd. V. Coll. Savs. Bank, 527 U.S. 627, 645 (1999). 45 Id. at 646. 46 Id. at 641.

12

The Court appeared to make this standard even stricter in Kimel, ignoring

Boerne’s statement about deferring to “the method by which [Congress] will reach a

decision,” and closely scrutinizing the legislative record supporting the ADEA. Although

this record was substantial, the Court held that § 5 legislation cannot be buttressed by a

pattern of discriminatory action in the private sector. Rejecting abrogation of immunity

under the ADEA, the Court declared that the fact “that Congress found substantial age

discrimination in the private sector is beside the point,” because “Congress made no such

findings with respect to the States.”47 Similarly, in reviewing the ADA’s employment

discrimination provisions in Board of Trustees of the University of Alabama v. Garrett,

the Court faulted Congress for focusing on private-sector discrimination, while producing

only “half a dozen examples” of disability discrimination by states.48 These isolated

incidents, the Court held, fell “far short of even suggesting the pattern of unconstitutional

discrimination on which § 5 legislation must be based.”49

These cases also refused to consider several other categories of evidence relied by

Congress, rejecting as irrelevant: 1) evidence relating to the actions taken by local and

federal government agencies, on the grounds that they were not relevant to abrogation of

state sovereign immunity;50 2) evidence of state action that was not clearly

unconstitutional, on the grounds that Congress lacked the power under § 5 to deter

activity permitted by the Constitution;51 3) evidence unrelated to the specific context at

issue, e.g., evidence of discrimination in public services to support remedies in

47 528 U.S. 62, 90 (2000). 48 531 U.S. 356, 369 (2001). 49 Id. at 370. 50 Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356 (2001). 51 Kimel v. Fla. Bd. of Regents, 528 U.S. 62 (2000); Garrett, 531 U.S. 356.

13

employment;52 and 4) evidence outside the Congressional record.53 Together, these

restrictions on the kinds of evidence that could support § 5 legislation suggested a

dauntingly high bar: Congress could not simply establish a nationwide problem and

extend a solution to public and private sectors alike; instead, it had to prove a widespread

pattern of violations of a particular constitutional right by state governments in a

particular context.54

Legislative Record: the More Flexible Hibbs/Lane Standard In two subsequent cases, however, very different (and largely liberal/moderate)

majorities of the Court would uphold § 5 legislation – at least in part – and indicate that

the exacting Kimel/Garrett standard was not universally applicable.55 In these cases, the

Court emphasized that the laws at issue – or at least particular applications of them –

served to protect constitutional rights such as gender equality and access to the courts,

which are accorded heightened judicial protection. This heightened protection, the Court

said, made it “easier for Congress to show a pattern of state constitutional violations.”56

While not expressly disagreeing with the prior cases, these cases rejected the evidentiary

limitations the Court had previously applied, apparently regarding them as irrelevant in

the context of more strongly-protected rights: the Court accepted evidence regarding

private actors and local governments, evidence of conduct that was not clearly

52 Garrett, 531 U.S. 356. 53 Kimel, 528 U.S. 62; Garrett, 531 U.S. 356. 54 Justice Scalia has gone further, suggesting that Congress must prove a pattern of discrimination with regard to each state. Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 742-43 (Scalia, J., dissenting). But see Tennessee v. Lane, 541 U.S. 509, 537 (Ginsburg, J., concurring) (attacking this view). 55 In Nevada Department of Human Resources v. Hibbs, 538 U.S. 721 (2003), Chief Justice Rehnquist and Justice O’Connor formed a majority with Justices Stevens, Souter, Ginsburg and Breyer, who had dissented in the prior cases. In Tennessee v. Lane, 541 U.S. 509 (2004), Justice O’Connor again joined the Court’s four moderate-liberal justices, but Rehnquist joined the other conservatives in dissent. 56 Id. at 736.

14

unconstitutional and that concerned discrimination in a variety of areas, and evidence not

directly before Congress.57

Thus, in upholding the abrogation of immunity under the Family and Medical

Leave Act (FMLA) in Nevada v. Hibbs, the Court relied primarily on a Bureau of Labor

Statistics survey showing gender disparities in private-sector leave; testimony, based on a

50-state survey, that public-sector policies were similar; and gender differences in states’

parental leave laws.58 The Court also relied on historical evidence of gender

discrimination by states more generally, including its own decisions. 59

Similarly, in Tennessee v. Lane, the Court held that the public services provisions

of the ADA “unquestionably [are] valid § 5 legislation” as they apply to “the class of

cases implicating the fundamental right of access to the courts.”60 Lane (written by

Justice Stevens) explicitly rejected “the mistaken premise that a valid exercise of

Congress’ § 5 power must always be predicated solely on evidence of constitutional

violations by the States themselves” 61 – seemingly contradicting Kimel and Garrett.62

Without distinguishing between state and local actors, the Court observed that the

“overwhelming majority” of examples of disability discrimination collected by Congress

related to public programs and services.63 The Court also pointed to state laws and

57 Id. at 731; Lane, 541 U.S. at 527; Amar, supra note 30, at 351. 58 Hibbs, 538 U.S. at 730-32. 59 Id. at 729-30. Hibbs dealt only with FMLA’s family care provision. The Courts of Appeals addressing the issue to date have held that the FMLA’s self-care provision does not validly abrogate state immunity, on the ground that they are not targeted at gender discrimination See, e.g., Nelson v. Univ. of Texas at Dallas, 535 F.3d 318, 312 (5th Cir. 2008) (citing cases). 60 541 U.S. at 531, 533-34. 61 Id. at 527 n. 16. 62 See Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 89 (2000); Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 368 (2001); Tennessee v. Lane, 541 U.S. 509, 542 (2004) (Rehnquist, J., dissenting). 63 Id. at 526 (quoting Garrett, 531 U.S. at 371 n.7).

15

judicial decisions stretching back a century that had manifested discrimination against

disabled persons in access to courts and the administration of justice.64

Thus, Hibbs and Lane establish that in situations where § 5 legislation protects

constitutional rights afforded heightened protection, the standards Congress must follow

to establish a sufficient threat to those rights are relaxed. Hibbs expressly distinguished

Kimel and Garrett because they dealt with forms of discrimination to which the Court

applies only “rational basis” review.65 While Boerne and Prepaid involved rights

accorded heightened protection (religious exercise and due process), they can also be

distinguished from Hibbs and Lane because the legislative record in the earlier cases was

virtually barren. Some commentators have argued that a distinction based on the level of

judicial scrutiny afforded the relevant constitutional rights cannot fully explain the

different evidentiary standards applied by the Court. 66 Nevertheless, most scholars

accept that this is clearest way to reconcile the cases.67

Balancing Wrongs and Remedies

64 Id. at 525. 65 Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721 (2003). 66 See, e.g., Amar, supra note 30, at 353 (“[E]ach specific governmental action that is unconstitutional because it is irrational should count just as much as a[n] action that is unconstitutional because it fails to survive … intermediate scrutiny”); Yoni Rosenzweig, Tennessee v. Lane: Relaxing the Garrett Requirements for Civil Rights Legislation, 40 HARV. C.R.-C.L. L. REV. 301, 310 (2005) (“Given that the Lane Court accepts evidence of discrimination that would not [violate the constitution under any standard], it should not matter that the standard for finding a … constitutional injury is lower in this case”). See also William D. Araiza, The Section 5 Power After Tennessee v. Lane, 32 PEPP. L. REV. 39 (2004) (“Whether this … distinction can provide a durable basis for these different approaches …is an open question”). 67 See, e.g., Pamela S. Karlan, Section 5 Squared: Congressional Power to Extend and Amend the Voting Rights Act After Tennessee v. Lane, 44 HOUS. L. REV. 1, 13 (2007); Michael E. Waterstone, Lane, Fundamental Rights, and Voting, 56 ALA. L. REV. 793, 808 (2005); Michael J. Pitts, Section 5 of the Voting Rights Act: A Once and Future Remedy?, 81 DENV. U. L. REV. 225, 247 (2003). But see Araiza, supra note 66, at 54-55 (suggesting a distinction between Lane and prior cases based on the regulation of “uniquely governmental” functions). Some courts, without discussing this distinction, have appeared to apply a more liberal § 5 analysis even with regard to discrimination subject to rational-basis review. See, e.g., Toledo v. Sanchez, 454 F.3d 24, 37-39 (1st Cir. 2006) (application of ADA to education).

16

In addition to being based on sufficient evidence of a threat to constitutional

rights, remedies enacted under § 5 must be “congruent and proportional” to the problems

they address.68 Boerne states this as a balancing test: “Strong measures appropriate to

address one harm may be an unwarranted response to another, lesser one.”69 Likewise,

the Kimel Court faulted the ADEA for “prohibiting substantially more state employment

decisions and practices than would likely be held unconstitutional,”70 but also

acknowledged that “[d]ifficult and intractable problems often require powerful

remedies.”71 The Court relied on this latter statement in upholding relatively broad

remedies under the FMLA and Title II of the ADA72 The analysis of the remedy is thus

intertwined with the evidentiary analysis: the stronger the history of discrimination, the

more robust a remedy may be. Kimel and Garrett also suggest that broader remedies may

be appropriate to protect constitutional rights that call for heightened judicial review,

such as freedom from race and gender discrimination.73 Because proportionality is

evaluated in relation to both the strength of the record and the nature of the right, these

factors “largely determine the outcome,”74 making it questionable whether the congruent-

and-proportional test does much independent analytic work. Indeed, in every case so far

68 See Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 372 (2001). 69 City of Boerne v. Flores, 521 U.S. 507, 530 (1997). 70 528 U.S. 62, 86 (2000). 71 Id. at 88. 72 See Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721 (2003); Tennessee v. Lane, 541 U.S. 509 (2004). In Lane, Justice Scalia rejected the congruence and proportionality test outright as too “vague” and “flabby,” contending that outside the context of race discrimination § 5 should not be held to permit anything beyond remedies for actual constitutional violations. 541 U.S. 509, 557-65 (2004) (Scalia, J., dissenting). It is ironic that Scalia’s position is ostensibly based in large part on the risk of “interbranch conflict” caused by “check[ing] Congress’s homework,” as it is hard to see how a drastic restriction of Congress’s powers by the Court would result in less interbranch conflict. 73 See Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 86 (2000) (emphasizing rational-basis standard in finding remedies overbroad); Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 372 (2001) (same). While Hibbs and Lane discuss the importance of heightened scrutiny only with regard to the evidentiary analysis, they are certainly consistent with this view. 74 Araiza, supra note 66, at 63.

17

the Supreme Court has held that a statute either meets both tests or fails both.75 The next

Part considers which outcome is most likely for GINA.

II. Analysis of GINA Employment Provisions

GINA was first introduced in 1995 in response to concerns about the misuse of

information regarding individuals’ possible genetic predispositions to various diseases.

Proponents of the bill sought to prevent insurers from using such information to deny

health care coverage, and to prevent employers from using genetic testing to weed out

individuals with genetic predispositions to potentially costly conditions. Congress

determined that such decisions, based on speculative fears about possible future

conditions, were fundamentally unwarranted and unfair, and would lead to intrusions on

workers’ medical privacy.76 Proponents were also concerned that even the possibility of

genetic discrimination would deter individuals from seeking genetic testing or

participating in genetic research, thereby blunting the societal benefits of this emerging

technology. Numerous versions of the bill and several hearings over a dozen years

culminated in GINA’s final passage by votes of 414-16-1 in the House and 95-0 in the

Senate.77 GINA will go into effect in November of 2009.78

Under GINA, employers are generally prohibited from seeking genetic tests or

family health history, and are completely prohibited from basing employment decisions

75 See Kimel, 528 U.S. 62; Garrett, 531 U.S. 356; Hibbs, 538 U.S. 721; Lane, 541 U.S. 509.In Boerne, the Court said it would find the scope of RFRA improper “[r]egardless of the state of the legislative record.” 521 U.S. 507, 532 (1997). Even if this statement is taken literally, the dramatic breath and stringent requirements of RFRA suggest that the proportionality prong should have independent bite only in extreme cases. 76 H.R. REP. NO. 110-028, Part 1 at 28. 77 Jessica L. Roberts, Preempting Discrimination: Lessons from the Genetic Information Nondiscrimination Act, 63 VAND. L. REV. __ (forthcoming 2009) (manuscript at 7-12, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1342903) (summarizing history). 78 H.R. REP. NO. 110-028, Part 1 at 8.

18

on such information.79 GINA provides employees the same remedies as Title VII of the

Civil Rights Act, including damages against public and private employers alike.80 Given

the Supreme Court precedents outlined above, states are likely to challenge GINA on two

grounds. First, Congress failed to expressly invoke the Fourteenth Amendment to justify

GINA’s abrogation of immunity. Second, and perhaps most significantly, states may

claim that Congress did not assemble a sufficient record of genetic discrimination, and

specifically of discrimination or invasions of privacy by the states. These challenges

could weaken GINA’s protections for the millions of Americans employed by the states.

GINA’s potential vulnerability to these challenges suggests that Congress has yet to

adequately grapple with the § 5 jurisprudence of the last decade.

Failure to Invoke the Section 5 Power

While GINA does not include any express language regarding the abrogation of

state sovereign immunity, the Supreme Court held in Kimel that Congress’s intent to

subject states to federal jurisdiction is sufficiently clear where the statute clearly applies

to the states.81 Congress clearly intended to apply GINA to the states, because its

employment provisions define both “employee” and “employer” to encompass state

employment.82 As discussed above, however, Florida Prepaid raises questions about the

Congress’s identification of the source of its authority for abrogation. The House reports

on GINA cite the Commerce Clause as providing authority for the legislation.83 As

79 42 U.S.C. § 2000ff, 2000ff-1. Title I of GINA adds these protections to several existing statutes to provide protections in the area of insurance. See H.R. REP. NO. 110-28, Part 1, at 14; Roberts, supra note 77, at 12-13. 80 42 U.S.C. § 2000ff-6. 81 Kimel v. Fla. Bd. of Regents, 528 U.S. 62 (2000). 82 42 U.S.C. §§ 2000ff(2)(A)(ii), 2000ff(2)(B)(ii). 83 H.R. REP. NO 110-028, Part 1 at 23. One committee report also identifies the Spending Clause as a basis for GINA, [need cite], apparently based on the statute’s appropriation of funds to create a commission to study whether a “disparate impact” provision should be added. 42 U.S.C. § 2000ff-7(b), (f). While

19

already discussed, however, the Court has repeatedly rejected the Commerce Clause and

other Article I powers as bases for abrogating sovereign immunity.84 That leaves § 5 of

the Fourteenth Amendment as a possible basis for GINA’s application to the states. The

legislative history of GINA does not expressly mention the Fourteenth Amendment

mentioned as a basis for its enactment. However, the reports do contain passing

references indicating that genetic discrimination may violate the Fourteenth

Amendment.85

Accordingly, the key question will be whether the Prepaid footnote precludes

reliance on § 5 to uphold GINA.86 If the footnote is read as establishing a strict rule that

reliance on one congressional power and failure to mention another is determinative,

there can be no abrogation under GINA. I have suggested, however, that the footnote

should be read narrowly since it was implicitly distinguished in Kimel. As an

antidiscrimination statute, GINA has an obvious connection with equal protection. It also

has an obvious connection with the fundamental right to medical privacy.87 Thus, while

the overbroad interpretations of some lower courts would jeopardize GINA’s abrogation

of immunity, the better view is that failure to expressly invoke the Fourteenth

Amendment creates no problem here. GINA’s sparse legislative record, however, creates

another, more serious hurdle.

Congress can condition federal funding on a waiver of state immunity as a condition for receipt of federal funds (as discussed infra, Part III), GINA does not contain such a condition. 84 See Seminole Tribe of Fla. v. Florida, 517 U.S. 44 (1996); Fla. Prepaid Postsecondary Educ. Expense Bd. v. Coll. Savs. Bank, 527 U.S. 627 (1999); Cent. Va. Cmty v. Katz, 546 U.S. 356 (2006). 85 Pub. L. 110-233 § 2(2) (noting that state sterilization laws have been modified to comply with Fourteenth Amendment); S. Rep. 110-48, at 9 (quoting judicial holding that genetic testing implicates Fourteenth Amendment). Compare Nat’l Ass'n of Bds. of Pharmacy v. Bd. of Regents of Univ. Sys. of Georgia, 2008 WL 1805439 (M.D.Ga. 2008) (suggesting a discussion of § 5 power in committee report may suffice to invoke that power for the Copyright Remedy Act, even though report went on to focus on commerce power as basis for that Act). 86 See discussion supra Part I. 87 See discussion infra.

20

Weakness of the Legislative Record

The Rights Protected by GINA

Before evaluating the evidence assembled by Congress, it is important to identify

the constitutional rights that GINA, through its application to state employers, protects.

As discussed supra, the identification of constitutional rights entitled to heightened

judicial review was crucial to recent cases upholding abrogation of sovereign immunity.

Genetic discrimination. The most obvious right protected by GINA is the right,

under the Equal Protection Clause, to be free from discrimination on the basis of genetic

information. While there is no Equal Protection case law on discrimination based on

genetic information, there are substantial reasons for applying heightened judicial

protection to discrimination on the basis of genetic information.88 Similar to suspect or

quasi-suspect characteristics such as race, national origin and gender, possession of a

particular gene, such as one that predisposes one to disease, is clearly an “immutable, or

distinguishing characteristic.”89 Such genetic information by itself also “bears no relation

to ability to perform or contribute to society.”90 Possession of particular genetic marker

may indicate a possibility of future physical or mental impairment, but at present it is a

mere piece of probabilistic data, subject both to chance and, in some cases, individual

88 See Jack M. Balkin, How New Genetic Technologies Will Transform Roe v. Wade, 56 EMORY L.J. 843, 860 (2007); Shannyn C. Riba, The Use of Genetic Information in Health Insurance: Who Will Be Helped, Who Will Be Harmed and Possible Long-Term Effects, 16 S. CAL. REV. L. & SOC. JUST. 469, 484-86 (2007); Kristie E. Deyerle, Genetic Testing in the Workplace: Employer Dream, Employee Nightmare – Legislative Regulation in the United States and the Federal Republic of Germany, 18 COMP. LAB. L.J. 547, 560-62 (1997); George P Smith, II & Thaddeus J. Burns, Genetic Determinism or Genetic Discrimination, 11 J. CONTEMP. HEALTH L. & POL'Y 23, 43-46 (1994). But see Roberts, supra note 77, at 52; Colin S. Diver & Jane Maslow Cohen, Genophobia: What Is Wrong with Genetic Discrimination?, 149 U. PA. L. REV. 1439, 1475-81 (2001). Notably, in passing GINA Congress did not attempt to argue that genetic predisposition resembles suspect or quasi-suspect classifications. 89 Lyng v. Castillo, 477 U.S. 635 (1986). 90 Deyerle, supra note 88 at 561, quoting Frontiero v. Richardson, 411 U.S. 677, 687 (1973).

21

health and lifestyle choices.91 Like gender-based distinctions, decisions based on genetic

predisposition may sometimes be legitimate, but are likely to more often be the result of

unreasoned fears or prejudices.92 Because perceptions of genetic risks may be overblown,

genetic discrimination could readily “have the effect of invidiously relegating the entire

class of [individuals with a particular genetic marker] to inferior legal status without

regard to the actual capabilities of its individual members.”93 If genetic discrimination is

subject to some form of heightened scrutiny, GINA’s ban on discrimination should be

subject to the more liberal analytic framework employed in Hibbs and Lane.

On the other hand, courts might analogize discrimination based on genetic

information to age and disability cases, which are subject only to rational basis review. In

refusing to treat mental retardation as a suspect classification, the Supreme Court said

that “where individuals in the group affected by a law have distinguishing characteristics

relevant to interests the State has the authority to implement,… the Equal Protection

Clause requires only a rational means to serve a legitimate end.”94 Kimel and Garrett

relied on this reasoning to hold that age and disability discrimination are subject only to

rational basis review.95 While genetic markers do not affect current capabilities, they may

indicate a risk of future impairment. This risk itself is relevant to some legitimate state

91 See, e.g., Gabrielle Kohlmeier, The Risky Business of Lifestyle Genetic Testing: Protecting Against Harmful Disclosure of Genetic Information, 2007 UCLA J. L. & TECH. 5, 8-15 (2007) (discussing risks posed by disclosure of information regarding genetic variations linked to diet and lifestyle); Lee M. Silver, The Meaning of Genes and "Genetic Rights,” 40 JURIMETRICS J. 9, 17-18 (1999) (discussing risk of discrimination based on genetic predisposition to addiction despite individuals’ ability to overcome such tendencies). 92 Cf. United States v. Virginia, 518 U.S. 515, 533-34 (1996) (noting that physical differences based on gender must be recognized, but may not be used to reinforce social inferiority). 93 Frontiero, 411 U.S. at 687. 94 Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 445 (1985). See also id. at 440 (heightened scrutiny is appropriate for characteristics that are “so seldom relevant to the achievement of any legitimate state interest that laws grounded in such considerations are deemed to reflect prejudice and antipathy”). 95 Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 83 (2000); Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 366 (2001).

22

interests, including the state’s interest in minimizing the potential costs (in the form of

diminished performance, leaves of absence, and health care) of future illness among state

employees. Thus, much may depend on courts’ assessment of how often genes are likely

to be relevant to state actions, and how often they will be irrelevant. Because there is as

yet no long history of discrimination based on genes as such (nor could there be), courts

may refuse to apply heightened scrutiny.96 If courts hold that genetic discrimination is

subject only to rational-basis review, the strict analysis of the Kimel and Garrett

decisions is likely to guide their treatment of GINA’s employment discrimination

provision.

Privacy. GINA also protects other interests in addition to equal protection.

Whereas § 202(a) of GINA prohibits the discriminatory use of genetic information, §

202(b) prohibits employers from requesting, requiring, or purchasing that information,

with few exceptions. Restricting the acquisition of genetic information certainly furthers

the goal of preventing discriminatory use, but it also protects constitutional privacy

rights, which call for more searching judicial review of state conduct. In Norman-

Bloodsaw v. Lawrence Berkeley Laboratory, the Ninth Circuit held that forced genetic

testing of employees could violate both the Fourth Amendment and the Due Process

Clause.97 Although this is the only reported case precisely on point, there is a judicial

consensus that the constitutional right to privacy extends to medical testing as well as the

privacy of medical information generally,98 including results of past tests,99 and to

96 See Diver & Cohen, supra note 88, at 1476-77; cf. also Virginia, 518 U.S. at 531 (“Today's skeptical scrutiny of official action … based on sex responds to volumes of history”). There is much more to be said on both sides of this constitutional question. The point here is that the outcome is as yet uncertain. 97 Norman-Bloodsaw v. Lawrence, 135 F.3d 1260, 1269 (9th Cir. 1998). 98 See, e.g., U.S. v. Westinghouse Elec. Corp., 638 F.2d 570, 577 (3d Cir. 1980) (“There can be no question that an employee's medical records… are well within the ambit of materials entitled to privacy protection”);

23

government acquisition of medical information from third parties.100 Thus, the full scope

of GINA’s privacy provision implicates the core constitutional right to privacy. Because,

unlike the discrimination provision, GINA’s privacy provision clearly implicates

constitutional privacy rights that are subject to robust judicial protection under current

precedents, it must be analyzed under the more liberal § 5 standard of Hibbs and Lane

rather than Kimel and Garrett. As the rest of this section shows, this more liberal standard

means that GINA’s privacy provision is more likely to survive constitutional challenge.

Race and gender. Congress noted in its findings that because “many genetic

conditions and disorders are associated with particular racial and ethnic groups and

gender… members of a particular group may be stigmatized or discriminated against as a

result of that genetic information.”101 Title VII of the Civil Rights Act already places

some limits on employer decisions regarding medical information; for example,

Lawrence Berkeley Laboratory held that Title VII prohibits screening black employees

for sickle-cell anemia trait.102 GINA supplements that protection by providing a broad

ban on such screening that is not subject to the Title VII defense of business necessity.103

To some extent, then, both of GINA’s employment provisions may be justified as a

response to race and gender discrimination, and benefit thereby from the more liberal

Lankford v. City of Hobart, 27 F.3d 477, 487 (10th Cir. 1994) (rejecting qualified immunity for seizure of medical records). 99 See, e.g., Doe v. City of New York, 15 F.3d 264 (2d Cir. 1994) (“Individuals who are infected with the HIV virus clearly possess a constitutional right to privacy regarding their condition”). 100 See, e.g., Whalen v. Roe, 429 U.S. 589, 599 (1977) (recognizing that state’s acquisition of all prescriptions written for certain drugs implicates privacy right). 101 Pub. L. 110-233 § 2(3). 102 Norman-Bloodsaw, 135 F.3d 1260, 1272 (9th Cir. 1998). It also prohibits the use of medical testing and medical information if it has a disparate impact based on race or gender. See Melinda B. Kaufmann, Genetic Discrimination in the Workplace: An Overview of Existing Protections, 30 LOY. U. CHI. L.J. 393, 424 (1999). Cf. Bradley v. Pizzaco of Neb., Inc., 7 F.3d 795 (8th Cir. 1993). 103 See Griggs v. Duke Power Co., 401 U.S. 424, 431 (1971) (establishing business necessity defense for Title VII disparate-impact claims).

24

framework of Hibbs and Lane. 104 To date, the Supreme Court has always upheld § 5

legislation that combats race or gender discrimination by the states.105 While these

important rights are not implicated by most applications of GINA, the Court’s recent

decisions have indicated a preference for piecemeal analysis of § 5 legislation, according

to the constitutional rights implicated.106 Accordingly, an analysis focused on race and

gender discrimination should benefit GINA with respect to the limited classes of cases

that implicate those concerns.107

Policy goals. While protecting equality and privacy were major justifications for

GINA’s employment provisions, equally important to Congress were concerns about

scientific research and public health.108 Most of the evidence offered to support GINA

relates not to actual discrimination, but to problems caused by public fears of

discrimination. The committee reports cite studies showing that many Americans are

concerned about losing jobs or insurance as a result of genetic testing, and that these fears

may lead them to avoid seeking genetic testing or participating in genetic research.109

This is a classic Commerce Clause rationale.110 Since this kind of evidence does nothing

104 GINA’s employment provisions specifically excludes from coverage “information about the sex or age of any individual.” 42 U.S.C. § 2000ff(4)(C). By its terms, however, this exclusion does not extend to genetic characteristics that are merely correlated with sex. 105 See South Carolina v. Katzenbach, 383 U.S. 301 (1966); Fitzpatrick v. Bitzer, 427 U.S. 445 (1976); Oregon v. Mitchell, 400 U.S. 112 (1970); City of Rome v. United States, 446 U.S. 156 (1980); Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721 (2003). See also Tennessee v. Lane, 541 U.S. 509, 564 (2004) (Scalia, J., dissenting) (§ 5 laws targeting race discrimination need only comply with Necessary and Proper Clause). 106 Lane, 541 U.S. 509; Georgia, 546 U.S. 15. 107 Similarly, under Georgia, GINA’s abrogation of sovereign immunity will be valid in cases presenting actual violations of constitutional rights, whether under the right to privacy or equal protection. Georgia, 546 U.S. 151. 108 See Roberts, supra note 77, at 32-35. 109 H.R. REP. NO. 110-28 Part 1, at 28-29; S. REP. NO. 110-48, at 6-7; Slaughter Testimony at 3; Rothenberg Testimony, at 3 (citing and discussing studies). 110 Compare EEOC v. Wyoming, 460 U.S. 226 (1983) (ADEA valid commerce legislation based on economic effects), with, Kimel v. Fla. Bd. of Regents, 528 U.S. 62 (2000) (ADEA invalid under § 5); see also United States v. Morrison, 529 U.S. 598 (2000).

25

to establish a threat of violations of constitutional rights, there is no reason to expect

courts to give it any weight with regard to abrogating sovereign immunity.

Analysis under the Kimel/Garrett standard

As discussed above, states will argue that genetic discrimination is subject only to

rational basis review and, accordingly, that courts should subject § 202(a) of GINA to the

exacting evidentiary standard of Kimel and Garrett. If analyzed under these precedents,

GINA may be held invalid as applied to the states.111 Under Kimel and Garrett, most of

the evidence collected by Congress would likely be deemed “beside the point” and

disregarded because it concerns discrimination in the private sector, and does not

establish a pattern of unconstitutional discrimination by the states themselves.112 The

studies and reports cited in the record either exclusively describe private-sector activity,

or fail to differentiate between public and private employment.113 The record does

include the case of Lawrence Berkeley Laboratory, a state-operated facility which was

successfully sued for subjecting African-American employees to mandatory medical

testing over several years.114 Although this is clearly an example of unconstitutional use

of genetic information in state employment, it is only a single case.

111 Accord, Roberts, supra note 77, at 52. 112 Kimel, 528 U.S. 62. 113 See S. REP. NO. 100-48, at 7; Slaughter testimony (describing surveys of employers by the American Management Association); Lisa Geller et al., Individual, Family and Societal Dimensions of Genetic Discrimination: A Case Study Analysis, 2-1 SCIENCE & ENGINEERING ETHICS 71 (1996); J.C. Fletcher & D.C. Wertz, Refusal of Employment or Insurance, abstract for presentation made at Annual Meeting of the Am. Soc. of Hum. Genet. (Baltimore, Nov. 1, 1997), cited in Office of Tech. Assessment, Genetic Monitoring and Screening in the Workplace , OTA-BA-455 (1989), and discussed in H.R. REP. NO. 110-28 Part 3, at 27. The Wertz/Fletcher study is more fully described in DOROTHY C. WERTZ & JOHN C. FLETCHER, GENETIC & ETHICS IN GLOBAL PERSPECTIVE 68-71 (2004). Of the two academic studies cited by Congress, one gives no indication of the types of employers reported to have engaged in discrimination, beyond stating that most were in “working-class occupations.” The Geller study is not a survey but a qualitative analysis of case studies. It does not specify how many participants reported discrimination in employment as opposed to other areas, let alone in public versus private employment. 114 The case settled after the Ninth Circuit held that nonconsensual medical testing implicated the constitutional right to privacy, and that singling out Black employees for sickle-cell anemia testing was also

26

This leaves the two historical episodes of state discrimination noted by Congress,

neither of which concerned employment: laws permitting sterilization of people with

disabilities, and mandatory sickle-cell screening in the 1970s. 115 The sterilization laws

were raised in Garrett, and the Court dismissed them in a footnote, saying only that

“there is no indication that any State had persisted in requiring such harsh measures as of

1990 when the ADA was adopted.”116 Accordingly, courts can be expected to give the

evidence of the sterilization laws little, if any, weight. Mandatory sickle-cell testing, too,

has long since been discontinued. If this historical evidence cannot be linked with

evidence of contemporary discrimination in the record, then GINA’s legislative record

may be deemed too weak to support the abrogation of sovereign immunity.117 In sum,

GINA’s record is comparable to the bodies of evidence rejected by the Court in Kimel

and Garrett: it consists primarily of private-sector and historical evidence, with only a

handful of contemporary examples of discrimination by states.118 Under the

Kimel/Garrett standard – which states will argue is the applicable standard – most of the

evidence gathered by Congress is irrelevant, and what little is left is clearly insufficient.

a form of race discrimination under Title VII of the Civil Rights Act (regardless of whether test results were used against the employee). 135 F.3d 1260, 1269-72 (9th Cir. 1998).The court rejected claims under the ADA. Id. at 1273. This case was decided prior to Florida Prepaid, Kimel and Garrett, and the defendants apparently did not assert sovereign immunity. 115 Pub. L. 110-233 §§ 2(2)-(3). 116 Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 369 n.6 (2001). 117 An additional gap in the record is Congress’s failure to identify any specific shortcomings in available state law remedies. While Congress expressly found that the “existing patchwork” of state laws is “confusing and inadequate to protect [Americans] from discrimination,” Pub. L. 110-233 § 2(5), the record contains no evidence to support this contention with respect to state employment laws. The only specific shortcomings of state law discussed in the record relate to the insurance field, and particularly to federal preemption in that field. H.R. REP. NO. 110-28 Part 1, at 30; H.R. REP. NO. 110-28 Part 3, at 28. See also S. REP. NO. 110-48, at 12. But the failure to specifically address state employment laws, at least, should not be a major obstacle. While noting the existence of state laws, Kimel and Garrett did not rely on them, or specifically demand evidence of their inadequacy. Kimel v. Fla. Bd. of Regents, 528 U.S. 62 (2000); Garrett, 531 U.S. 356. 118 In fact GINA is weaker, but in respects that appear to be irrelevant under the standards applied in Kimel and Garrett. See discussion infra.

27

Analysis under Hibbs and Lane

A different picture is presented under the more liberal framework of Hibbs and

Lane, though the result is by no means assured. If courts apply the analysis of Hibbs and

Lane – as they should with respect to, at a minimum, GINA’s privacy provision, and the

class of discrimination cases implicating race and sex equality concerns – the evidence

amassed by Congress should be enough to support application to the states. The record

included recent surveys by the American Management Association which, like the

Bureau of Labor Statistics study relied on by the Court in Hibbs, showed the existence of

a significant, current problem of invasions of genetic privacy in the private sector.119

Specifically, they show that a distinct minority of employers seek genetic information

about applicants and employees. While less than one in twenty employers in these

surveys engaged in testing for specific genetic conditions, the survey also found that one

in five employers sought family medical history,120 which is prohibited under GINA to

the extent that it includes “the manifestation of a disease or disorder in family members

of such individual.”121 A pair of mid-1990s academic studies also provide some evidence

of this problem, despite some methodological shortcomings.122 Additionally, the record

119 But see Patricia Nemeth & Terry W. Bonnette, Genetic Discrimination in Employment, 88-JAN MICH. B.J. 42, 44 (2009) (characterizing record as showing “a dearth of evidence that genetic information discrimination was a widespread problem”). 120 Additionally, most employers who tested for genetic traits said they used that information in personnel decisions, as did a substantial fraction who asked about family medical history. S. REP. No. 100-48, at 7 (describing 2000 survey); Slaughter testimony (describing 2001 survey). 121 42 U.S.C. § 2000ff(4)(A)(iii). 122 The Wertz/Fletcher study found that “almost all” examples of what patients at genetics clinics described as discrimination in both employment and insurance involved symptomatic conditions or were otherwise “characteristic of broad general employment practice…or general insurance practice….None of the patients’ reports pointed to specifically ‘genetic’ discrimination,” in the sense of being based solely on carrier status or predisposition. Although data from medical professionals and the public did indicate that genetic discrimination in employment “exists,” the authors concluded that it was “rare.” WERTZ & FLETCHER, supra note 112, at 69-70. The Geller study is actually an analysis of case studies gathered from interviews; the authors emphasized that it was “not a survey,” had a low response rate, and that “any statistical analysis of the cases would be inappropriate.” The authors provided a breakdown of reports by

28

contains evidence, particularly in a 1989 report by the Office of Technology Assessment,

that genetic screening by employers has been a continuing phenomenon since the 1970s,

and has included some of the nation’s largest employers.123 The legislative record also

included several individual stories of discrimination, including two handled by the Equal

Employment Opportunity Commission.124

Although the GINA record contains only a few instances of recent state

violations,125 the Court’s precedents nowhere demand a certain number of recent

constitutional violations by states. More importantly, they do not say that recent actions

are the only kind of evidence relevant to § 5 analysis. While the cases since Boerne have

focused heavily on recent evidence of constitutional violations by states, the Court has

never said that this is the only kind of acceptable factual predicate for § 5 legislation.126

Each of the post-Boerne cases involved new congressional responses to longstanding

problems; nothing in them mandates that Congress wait years for conclusive proof that

past violations are being perpetuated through innovative means.127 Rather, the Court,

through its reliance on pre-Boerne decisions, “has recognized that Congress has utilized a

type of genetic condition, but did not specify how many reports related to employment. Geller et al., supra note 112, at 83, 88. Finally, both studies relied on self-reports and provided little to no detail about the cases, with one noting that it was “difficult to determine to what extent reports of genetic discrimination are of actual rather than perceived discrimination.” Id. at 83. If this collection of “anecdotal” evidence constituted the only or the primary evidence supporting abrogation, these shortcomings could be very problematic. See Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 370 (2001); Tennessee v. Lane, 541 U.S. 509, 542 (2004) (Rehnquist, C.J., dissenting). In light of the other, stronger evidence of private- sector use of genetic information, however, these studies simply add a modicum of support to the record. 123 Office of Tech. Assessment, Genetic Monitoring and Screening in the Workplace, OTA-BA-455 (1989), cited in S. REP. No. 100-48, at 7. 124 See Roberts, supra note 77 at 27-28. One case, settled by the EEOC, involved systematic testing of employees. See S. REP. No. 110-48, at 9. 125 These include the Lawrence Berkeley Laboratory case, as well as reports in the academic studies of instances of discrimination based on family genetic history in education and state adoption services. Geller, et al., supra note 112 at 77-78. 126 Mark A. Posner, Time Is Still On Its Side, 10 N.Y.U. J. LEGIS. & PUB. POL'Y 51, 99 (2007). 127 Cf. Johnson v. Florida, 405 F.3d 1214, 1244 (11th Cir. 2005) (Wilson, J., dissenting in part) (arguing Congress was not constrained to extend Voting Rights Act coverage only to particular forms of discrimination it had documented, because “[i]f this were the standard, states would always have one free bite at the apple”).

29

variety of approaches when establishing the historical predicate for enacting prophylactic

legislation.”128 Notably, the Court’s first modern § 5 decision, upholding the language-

discrimination section of the Voting Rights Act, stated that Congress has the authority to

assess “the risk or pervasiveness of the discrimination” by state actors.”129

Consider the reauthorization of the preclearance provisions of the Voting Rights

Act.130 The existence of the VRA preclearance provisions for four decades has effectively

prevented and deterred much discrimination, making it more difficult to prove that its

requirements are still needed.131 Congress’s rationale for periodic reauthorization

therefore has begun with the long history of unconstitutional discrimination, and “then

appropriately [sought] to link those violations to the present day through what Congress

has concluded is a continuing special risk of discriminatory decision making engendered

by those violations.”132 The Court embraced this approach in City of Rome v. United

States, in which the pre-Boerne Court upheld the 1975 VRA extension.133 Rome primarily

relied not on evidence of recent constitutional violations but on the evidence of frequent

preclearance denials for voting changes that would have a disparate impact on minorities,

which cumulatively indicated the risk that purposeful discrimination would recur without

128 Posner, supra note 125, at 99. 129 Katzenbach v. Morgan, 384 U.S. 641, 653 (1966). Kimel even notes that review of legislative history is itself only “[o]ne means” of evaluating § 5 remedies, suggesting that these cases’ focus on recent state violations is not exclusive. Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 88 (2000). 130 The preclearance provisions, otherwise known as (the other) Section 5, provide for mandatory federal review of all changes in election procedures in certain jurisdictions with a history of pervasive voting discrimination. See 42 U.S.C. § 1973c (2000). 131 See, e.g., Karlan, supra note 67, at 20-27; Nathan Persily, The Promise and Pitfalls of the New Voting Rights Act, 117 YALE. L.J. 174, 191-207 (2007). 132 Posner, supra note 125, at 100. Cf. also Ass’n for Disabled Americans v. Fla. Int’l Univ., 405 F.3d 954, 959 (11th Cir. 2005) (holding with regard to the ADA that “[i]n light of the long history of state discrimination against students with disabilities, Congress reasonably concluded that there was a substantial risk for future discrimination”). 133 City of Rome v. United States, 446 U.S. 156, 180-82 (1980); Posner, supra note 125, at 101.

30

an extension.134 Thus, Rome stands for a principle of “historical predicate flexibility,”

unchanged by subsequent cases.135 Under this principle, at least when protecting

fundamental rights Congress may rely not only on a recent pattern of constitutional

violations, but alternatively on other evidence, both recent and older, that establishes a

special risk that historical violations are likely to be perpetuated.136

Like the VRA, GINA presents a “unique constitutional quandary,”137 and as with

the VRA, “it is the problem of time that lies at the heart of the constitutional question.”138

The VRA’s time problem arises from the fact that the passage of time and the

effectiveness of the law itself have both served to lessen a historically severe problem.

GINA presents the flip-side of this problem: rapid technological developments have

enabled previously impossible forms of discrimination. A sensible extension of Rome’s

flexible approach would be that, where the passage of time has resulted in the emergence

of qualitatively new or vastly expanded opportunities for invasion of constitutional rights;

where the problem is proven to exist in the private sector; and where states engaged in

analogous violations in the past, the threshold for recent evidence should be relatively

modest.

Under this view, GINA’s abrogation of immunity could and should be upheld.

It is only in recent years that scientists have discovered genetic links to a wide variety of

medical conditions, and that accurate and affordable testing for a wide variety of traits

134 Rome, 446 U.S. at 180-82. Posner also describes how earlier decisions upholding aspects of the VRA employed a similar “historical predicate flexibility.” Supra note 125 at 99-100. 135 See Posner, supra note 125, at 99-100. Earlier decisions upholding parts of the VRA displayed a similar flexibility. See id. 136 Consistent with this approach, the district court in the current VRA challenge looked to both the past record and “the risk of future constitutional harm.” Northwest Austin Mun. Util. Dist. No. 1 v. Mukasey, 573 F.Supp.2d 221, 269 (D.D.C. 2008). 137 Persily, supra note 130, at 192. 138 Posner, supra note 125, at 94.

31

has become available.139 Opportunities for scrutiny of workers’ possible genetic

predispositions were far fewer even a decade ago, when most of the research in the record

was conducted.140 Given the emergent nature of the problem (and the relatively small

fraction of American jobs located in state governments), it might have been impossible

for Congress to uncover a large body of evidence regarding use of genetic information by

state employers. Yet the evidence Congress did assemble suggests that states are not

likely to be exempt from the practices of the private sector. Congress enacted the

National Sickle Cell Anemia Control Act of 1972 to halt states’ use of mandatory sickle-

cell testing – precisely the sort of invasion of medical privacy that GINA prohibits by

state employers.141 Despite the 1972 Act, the aforementioned Lawrence Berkeley

Laboratory case revealed a years-long pattern of coerced sickle cell testing.142

Recognizing the risk that these practices would multiply rapidly in the coming years,

GINA’s sponsors argued that Congress “[could] not possibly afford to wait any longer”143

for this problem “to flourish [and] take root.”144, 145 Although not relied on by Congress,

139 Compare, e.g., U.S. Office of Tech. Assessment, The Role of Genetic Testing in the Prevention of Occupational Disease (Apr. 1983) (describing genetic testing as “in its infancy,” with screening available for only a few traits, and not reliable for use in the general population), with, Amanda K. Sarata, CRS Report for Congress: Genetic Testing: Scientific Background for Policy makers (Jan. 2007) (describing rapid growth in genetic testing, with over 1,000 tests available for clinical use). 140 See, e.g., Andrew E. Rice, Eddy Curry and the Case for Genetic Privacy in Professional Sports, 6 VA. SPORTS & ENT. L.J. 1, 5 (2006) (noting more than six-fold increase in number of testable disease-related traits since 1997). 141 See S.REP. NO. 110-48 at 9; 42 U.S.C. § 300b. The sterilization laws, which persisted for longer than mandatory sickle cell screening and involved an even more egregious invasion of privacy, would also be relevant to this analysis. See Tennessee v. Lane, 541 U.S. 509, 534 (2004) (Souter, J., concurring) (citing these laws as supporting the ADA). 142 See Norman-Bloodsaw v. Lawrence, 135 F.3d 1260 (1998). 143 110 Cong. Rec. H4095 (daily ed. Apr. 23, 2007) (statement of Rep. Slaughter). 144 Roberts, supra note 77 at 31 (quoting Sen. Snowe). 145 The Court’s rejection of what it called “speculative harms” in Florida Prepaid is easily distinguished. There, the Court dismissed Congress’s concern that “patent infringement by States might increase in the future,” calling it merely “speculative.” 527 U.S. 627, 541 (1999). This concern was based on the trend in state universities toward commercializing the results of research. H.R. REP. NO. 101-960, pt. 1, at 38. While Congress had evidence of a pattern of patent violations in the private sector, unlike with GINA there was no historical evidence to suggest that states would be likely to engage in this sort of behavior. Moreover,

32

the history of disability discrimination identified in Lane further establishes this risk,

because genetic discrimination is based essentially on the perception of an individual’s

propensity to become disabled.146

States will argue, however, that such a flexible approach to GINA is inconsistent

with the Court’s recent precedents. While the post-Boerne cases have not squarely held

that a showing of widespread and recent state violations is always required, their general

approach, as well as some isolated language, can be interpreted as adopting that

approach.147 Accordingly, states will argue that the flexible, risk-based analysis of earlier

cases like Rome has been abandoned, or limited to the context of race and/or statutory

reauthorization. Under this view, section 5, as the Supreme Court has construed it, simply

does not permit Congress to act with dispatch in the face of emerging threats to

constitutional rights.148

In the event courts do hold that evidence of recent constitutional violations is

always required for § 5 legislation, the GINA record may not pass muster even under

Hibbs. As noted above, in Hibbs the key link between the substantial evidence of private-

sector discrimination and the abrogation of state sovereign immunity was a 50-state

survey showing that public sector policies “differ[ed] little from those [of] private sector

the supposed changed circumstances relied upon by Congress in Florida Prepaid amounted to the fact that states’ own choices gave them increased incentives to violate the law. By contrast, GINA responds to changes in objective circumstances that provide new opportunities and incentives for constitutional violations. 146 See, e.g., Steve Lash, Maryland employers work to comply with Genetic Information Nondiscrimination Act, DAILY RECORD (Balt. Mar. 2, 2009) (quoting EEOC chair Stuart Ishimaru describing GINA as extending protection to individuals with a propensity to become disabled). 147 See City of Boerne v. Flores, 521 U.S. 507, 530 (1997)(“RFRA's legislative record lacks examples of modern instances…”); Bd. of Trs. of the Univ. of Ala v. Garrett, 531 U.S. 356, 369 n.6 (2001) (“there is no indication that any State had persisted in requiring such harsh measures as of 1990 when the ADA was adopted”). This view is most clearly stated, however, in the Hibbs and Lane dissents, which criticize the Court’s reliance on “outdated” evidence. Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 744 (Kennedy, J., dissenting); Tennessee v. Lane, 541 U.S. 509, 540 (Rehnquist, C.J., dissenting). 148 Moreover, the pending challenge to the 2006 VRA reauthorization presents an opportunity for the Supreme Court to clarify, and possibly restrict, the application of § 5 in time-sensitive contexts.

33

employers.”149 The GINA record contains no such evidence that current practices in

private and public employment are actually comparable. Instead there is a record similar

to Kimel and Garrett, with little evidence of a link between past or private actions and

contemporary state actors. Even assuming those cases would have turned out differently

under the Hibbs/Lane standard, GINA is weaker: the record in Kimel included relatively

recent evidence of age discrimination in California,150 while Garrett record contained

substantial evidence of disability discrimination in non-employment contexts.151

Accordingly, while strong arguments exist for upholding abrogation under GINA’s

privacy provisions, that outcome is by no means certain.

GINA’s narrow remedies may not compensate for a weak record It is worth reiterating that congruence and proportionality is best understood as a

balancing test. GINA provides a well-tailored remedy, placing limited burdens on states

that largely track constitutional principles. Its privacy provision is “narrowly targeted” 152

at the unnecessary, covert, or coercive collection of information, and contains no less

than six exceptions, ensuring that states are not punished for inadvertent or innocuous

collection of information.153 It is thus considerably less burdensome than the affirmative

duties upheld in Hibbs and Lane. GINA’s discrimination provision is also more narrowly

149 See Hibbs, 538 U.S. at 730 n. 3; Lane, 542 U.S. at 529 n.17 (citing this as a key piece of evidence in Hibbs). 150 Kimel, 528 U.S. at 90. 151 Garrett, 531 U.S. at 369-72. A skeptical court could also invoke the (unsubstantiated) assertion of the Congressional Budget Office, in its cost estimate for GINA, that states are unlikely to engage in prohibited conduct. CBO Statement on H.R. 493, in H.R. REP. NO. 110-28 Part 1, at 49. 152 Cf. Hibbs, 538 U.S. at 738. 153 Collection of genetic information is permitted if it is inadvertent; is used anonymously and with permission for purposes of a health benefit program; consists of family medical history needed to comply with the FMLA; appears in publicly or commercially available documents; is appropriately used to monitor the potential effects of toxic substances; or if the employee works in DNA testing and a sample is needed to detect contamination. 42 U.S.C.A. § 2000ff-1(b)(1)-(6). See also S. REP. NO. 110-48, at 29 (stressing that Congress carefully drafted exceptions in response to business input to avoid unintended burdens or conflicts with other laws).

34

tailored than the ADA’s.154 To the extent that they prevail under the evidentiary analysis

of Hibbs and Lane, both provisions should be deemed sufficiently tailored as well. Yet

just as the case law indicates that a stronger the record of discrimination will justify a

more robust remedy, a weak record will not support even a modest remedy.155 Thus, to

the extent that the record supporting GINA is deemed insufficient to support § 5

legislation, GINA’s modest scope may not suffice to save it.

III. What Went Wrong with GINA?

Based on the above analysis, GINA’s employment provisions may be vulnerable

to attack under the Supreme Court’s Section 5 jurisprudence insofar as they provide

remedies for state employees. First, unless courts apply heightened constitutional scrutiny

to genetic discrimination, states may be successful in arguing that the scant evidence

collected is insufficient to permit suits for damages against states under GINA’s

employment provision in most cases. Second, there is a strong case for abrogating

sovereign immunity in those cases where employers rely on genetic information linked to

particular racial or ethnic – but this will only be a limited slice of GINA cases. Third, the

case for upholding GINA’s privacy provision as § 5 legislation is also stronger and

should prevail, but this outcome is far from certain. Thus, while in general GINA

represents a story of bipartisan legislative success – providing a nuanced solution to a

complex emerging problem – in the context of state employment it risks falling short, just

as have previous civil rights laws. But it didn’t have to be this way. Legislation

contemporaneous with GINA demonstrates how Congress how can more effectively

154 Compare 42 U.S.C.A. § 2000ff-1(a) (employers may not “discriminate…because of genetic information” or “limit, segregate or classify” on such basis); 42 U.S.C. § 2000ff-7 (no disparate-impact claims), with Bd. of Trs. of the Univ. of Ala v. Garrett, 531 U.S. 356, 372-73 (2001) (emphasizing reasonable-accommodation and disparate-impact provisions). 155 See Kimel v. Fla. Bd. of Regents, 528 U.S. 62 (2000).

35

respond to the Court’s § 5 jurisprudence through a combination of diligent fact-finding

and careful drafting.

Building a Supporting Record: the Voting Rights Act Reauthorization Act

The Voting Rights Act Reauthorization Act of 2006 (VRARA) was passed 22

months before GINA, by similarly overwhelming margins.156 The Act reauthorized for an

additional 25 years several temporary provisions set to expire in 2007, most importantly

the “preclearance” provisions, which require certain jurisdictions with a history of racial

discrimination in voting to obtain prior approval from the Attorney General or a federal

court for any changes in voting procedures.157 The Supreme Court upheld the

preclearance provisions when they were initially passed in 1966,158 and upheld them

again following the 1975 reauthorization.159 This, of course, was prior to the Rehnquist

Court’s string of § 5 rulings.

Seeking to ensure that the reauthorization would survive judicial review,

Congress developed one of the most extensive legislative records in its history.160 The

House and Senate Judiciary Committees held no less than 21 hearings on the VRARA,

assembling over 15,000 pages of testimony and documentary evidence from nearly one

hundred witnesses as well as other interested groups and government agencies. In

addition, the committees reviewed over a dozen outside reports on the effectiveness of,

156 Pub. L. No. 109-246. 157 42 U.S.C. § 1973c. The Act also extends the VRA’s language access provisions, which require certain jurisdictions to provide language assistance to voters with limited English proficiency. 42 U.S.C. § 1973aa- 1a. Although originally enacted in 1975, the constitutionality of the language access requirements has never yet been challenged. See James Thomas Tucker, The Battle Over “Bilingual Ballots” Shifts to the Courts: A Post-Boerne Assessment of Section 203 of the Voting Rights Act, 45 HARV. J. ON LEGIS. 507 (2008) (arguing provisions are constitutional); Daniel P. Tokaji, Intent and Its Alternatives: Defending the New Voting Rights Act, 58 ALA. L. REV. 349 (2006) (same). But see Tucker, supra, at 514 (quoting testimony by opponent that provisions would probably be challenged if reauthorized). 158 South Carolina v. Katzenbach, 383 U.S. 301 (1966). 159 City of Rome v. United States, 446 U.S. 156 (1980). 160 H.R. REP. NO. 109-478, at 5 (2006); S. REP. NO. 109-295, at 10 (2006).

36

and continuing need for the preclearance provisions in covered jurisdictions around the

country. This record included statistics regarding state registration and turnout; low

numbers of minority elected officials in covered jurisdictions, numbers of Attorney

General objections and “more information” requests filed regarding proposed voting

changes; numbers of judicial preclearance and enforcement suits brought under the Act;

constitutional suits over voting discrimination; the employment of federal election

observers; patterns of racially-polarized voting; and evidence that Section 5 deterred state

and local officials from adopting voting changes.

From this evidence, Congress concluded that while “[s]ignificant progress has

been made in eliminating first generation barriers experienced by minority voters,”

“vestiges of discrimination in voting continue to exist.”161 Congress focused especially on

the proliferation of litigation and federal government objections and oversight actions in

relation to state and local voting practices, which Congress believed demonstrated that

the Act continued to prevent discrimination that would otherwise have occurred.162

Congress also pointed to continuing patterns of racially polarized voting, which provide a

continuing political incentive for race-based manipulation of the polls.163 Additionally,

the House Report provided an extended discussion of Congress’s powers under the

Fourteenth and Fifteenth Amendments, and explained how the Act satisfied Supreme

Court precedents.164 The reauthorization of the preclearance provisions was immediately

challenged, and was upheld by a three-judge panel of the D.C. federal district court,

161 Pub. L. No. 109-246, § 2(b)(1), (2). 162 Id. at § (4)-(5), (8). 163 Id. 164 H.R. REP. NO. 109-478, at 54-60 (2006). But see Persily, supra note 131, at 182-92 (describing the controversial, post-enactment Senate Report, joined only by Republicans and raising reservations about the bill).

37

which found that the VRARA record was “far more powerful” than those found

satisfactory in Hibbs and Lane.165

At oral argument on April 28, 2009, however, five justices appeared skeptical of

the scope and supporting record of the reauthorization.166 In particular, they noted that

while Congress compiled an extensive record, it failed to reexamine the geographic

coverage of the preclearance provisions in light of the passage of time.167 Given the

tensions in the Court’s § 5 precedent and the shift in its membership, it is impossible to

predict how this case will be decided – or what shifts in § 5 doctrine may result.168

Nevertheless, the VRARA represents the kind of robust effort Congress can make to

justify its exercise of this remedial authority to protect civil rights according to the

Court’s announced criteria. Of course, Congress will likely not be able to develop such a

truly massive record each time it exercises its § 5 authority, and may not always place

such a high priority on doing so. Nor will such an overwhelming effort necessarily be

165 Northwest Austin Mun. Util. Dist. No. 1 v. Mukasey, 573 F.Supp.2d 221, 271 (D.D.C. 2008). Because the suit asserted both a constitutional challenge and the plaintiff jurisdiction’s claim for “bailout” from the preclearance requirements, it was subject to the VRA’s provision for review by a three-judge district court panel, with direct review by the Supreme Court. 42 U.S.C. § 1973aa-2. 166 Transcript of Oral Argument, Northwest Austin Mun. Util. Dist. No. 1 v. Mukasey (No. 08-322). 167 Id. While the fact-finding behind the VRARA was massive, lawmakers declined to reexamine the structure of the preclearance provisions themselves, judging that opening this settled “can of worms” would have caused “the political coalition behind the law [to] collapse[.]” Persily, supra note 131, at 207-16. Some commentators cautioned that, although Congress’s fact-finding process was aimed at surviving judicial review, the failure to revisit the substance of the law, including its coverage and bailout formulas, may have jeopardized it. See Richard H. Pildes, Political Avoidance, Constitutional Theory and the VRA, 117 YALE L.J. POCKET PART 148 (2007). 168 See, e.g., Luis Fuentes-Rohwer, Legislative Findings, Congressional Powers, and the Future of the Voting Rights Act, 82 IND. L.J. 99, 130 (2007) (concluding that “the question for the future is ultimately a question of judicial attitudes and whether the Court can muster the will to strike down the most effective civil rights statute in history”). Numerous commentators have argued that the preclearance provisions should be upheld. See, e.g., Kristen Clarke, The Congressional Record Underlying the 2006 Voting Rights Act, 43 HARV. C.R.-C.L. L. REV. 385 (2008); Karlan, supra note 67; Posner, supra note 125; Tokaji, supra note 160.

38

required for other laws, given the unique burdens imposed on states by the VRA.169 But

the VRARA shows that Congress can, when sufficiently motivated, recognize the

challenges posed by the Court’s § 5 precedents and make a concerted effort to overcome

them.

Findings, Rights and Remedies: the End Racial Profiling Act

Another example of building support for congressional authority into legislation is

the proposed End Racial Profiling Act (ERPA), first introduced by Rep. John Conyers

and Sen. Russ Feingold in 2001 in response to concerns that law enforcement agencies

were unfairly targeting minorities for routine both routine and spontaneous stops and

searches.170 The Act would prohibit federal, state and local law enforcement agencies

from “relying, to any degree, on race, ethnicity national origin, or religion” in

investigation or enforcement decisions, except when using a description of a specific

criminal suspect.171 ERPA contains numerous findings that support its constitutionality

under § 5, including: a finding that “[s]tatistical evidence from across the country

demonstrates that racial profiling is a real and measurable phenomenon”; specific

findings to that effect from national surveys; that racial profiling increased in the wake of

September 11, 2001; and that such profiling violates the Equal Protection Clause.172

Particularly notable in light of Lane, the legislation invokes not only equal protection but

also the rights to travel and to be free from unreasonable searches and seizures as bases

for the legislation, along with the Commerce Clause (on the basis that racial profiling

169 See, e.g., Berry v. Doles, 438 U.S. 190, 200 (1978) (Powell, J., concurring in judgment) (“It must be remembered that the Voting Rights Act imposes restrictions unique in the history of our country on a limited number of selected States”). 170 End Racial Profiling Act of 2008, S. 110-2481/H.R. 110-4611 (most recent version); End Racial Profiling Act of 2001, S. 107-989/H.R. 107-2074 (original version). 171 End Racial Profiling Act of 2008, §§ 3(6), 101. 172 Id. at § 2(a).

39

discourages interstate travel).173 While ERPA has never been reported out of committee,

these statutory findings and purposes would likely be sufficient to support the Act’s

limited ban on consideration of race, and its provision for citizen suits for injunctive

relief and attorneys’ fees.174

The Waiver Approach: the Employment Non-Discrimination Act

As previously discussed, it might have been impossible, even with the best

efforts, for Congress to document a widespread pattern of present genetic discrimination.

Fortunately, this is not the only way Congress can effectively respond to the Court’s § 5

precedents. Indeed, Congress likely could have ensured GINA’s full application to state

employers by requiring that states waive their immunity in exchange for maintaining

federal grants. The Supreme Court made clear two decades ago, in Chief Justice

Rehnquist’s opinion in South Dakota v. Dole, that imposing conditions on federal funding

for the states is generally permissible.175 Responding to a Supreme Court decision that

limited remedies under the Rehabilitation Act,176 Congress passed a law in 1986 that

explicitly required, as a condition of receipt of any federal funds, a waiver of states’

sovereign immunity for suits under that Act as well as under Title VI of the Civil Rights

Act, and Title IX of the Education Amendments of 1972.177 Every court of appeals has

accepted the validity of such waivers.178 While some commentators warned during the

173 Id. at § 2(b). 174 Id. at § 102. In addition to the mandatory prohibition on racial profiling, ERPA would make the implementation of certain policies to prevent such profiling a condition for receipt of certain federal grants. These conditions are not subject to private judicial enforcement, but instead to an administrative complaint process. Id. at §§ 301-303. 175 483 U.S. 203 (1987) (upholding condition on federal highway funds that states enact minimum). drinking age of 21). 176 Atascadero State Hosp. v. Scanlon, 473 U.S. 234 (1985). 177 42 U.S.C. § 2000d-7 (Civil Rights Remedies Equalization Act). 178 Nieves-Márquez v. Puerto Rico, 353 F.3d 108, 127–29 (1st Cir. 2003); Koslow v. Pennsylvania, 302 F.3d at 167–76 (3d Cir. 2003); Constantine v. Rectors and Visitors of George Mason Univ., 411 F.3d 474,

40

Rehnquist years that aggressive use of the spending power might be seen by the Court as

impermissible “circumvention” of its federalism rulings,179 the Roberts Court has given

no indication that it will invalidate such legislation wholesale .180

The current Court has, however, said that it will interpret all such legislation

narrowly. In Arlington Central School District v. Murphy, Justice Alito wrote for the

Court that any law based on Congress’s spending power will be valid only to the extent

that it provides “clear notice” to state officials of the full extent of every duty and

potential liability it creates.181 Arlington could result in a contraction of individual rights

and remedies under a variety of federal laws.182 For example, following Boerne,

Congress expressly required that states waive their immunity in religious-expression

cases a condition of federal funding.183 While courts have upheld this waiver, several

491–96 (4th Cir. 2005); Miller v. Texas Tech Univ. Health Sciences Ctr., 421 F.3d 342, 349 (5th Cir. 2005) (en banc); Nihiser v. Ohio Envtl. Prot. Agency, 269 F.3d 626 (6th Cir. 2001); Stanley v. Litscher, 213 F.3d 340, 344 (7th Cir. 2000); Jim C. v. United States, 235 F.3d 1079, 1081–82 (8th Cir. 2000) (en banc); Lovell v. Chandler, 303 F.3d 1039, 1050–52 (9th Cir. 2002); Brockman v. Wyo. Dept. of Fam. Servs., 342 F.3d 1159, 1167–68 (10th Cir. 2003); Garrett v. Univ. of Ala. at Birmingham Bd. of Trustees, 344 F.3d 1288, 1293 (11th Cir. 2003) (per curiam); Barbour v. Wash. Metro. Area Transit Auth., 374 F.3d 1161 (D.C. Cir. 2004). But see Garcia v. SUNY Health Sciences Ctr. of Brooklyn, 280 F.3d 98, 113–15 (2d Cir. 2001) (holding waiver does not extend to disputes arising prior to time when abrogation of immunity under ADA was in question). See also Rochelle Bobroff & Harper Jean Tobin, Strings Attached: The Power of the Federal Purse Waives State Sovereign Immunity for the Rehabilitation Act, 24 CLEARINGHOUSE REV. 1- 2, 16 (2008) (summarizing these cases). 179 See e.g., Lynn A. Baker & Mitchell N. Berman, Getting Off the Dole: Why the Court Should Abandon Its Spending Doctrine, and How a Too-Clever Congress Could Provoke It to Do So, 78 IND. L.J. 459 (2003). 180 See Neil S. Siegel, Dole’s Future: A Strategic Analysis, 16 SUP. CT. ECON. REV. 165, 201 (2008); Samuel R. Bagenstos, Spending Clause Litigation in the Roberts Court, 58 Duke L. J. 345, 355-84 (2008). 181 548 U.S. 291 (2006). Prior to Arlington, a clear-notice standard had applied only to a limited set of questions about “unforeseeable” forms of liability, rather than to every question of remedies or interpretation. See, e.g., id. at 304-05 (Ginsburg, J., concurring in part and in judgment); id. at 316-86 (Breyer, J., dissenting); Barnes v. Gorman, 536 U.S. 181, 188 n. 2 (2002); id. at 191 (Souter, joined by O’Connor, J., concurring); Bennett v. Kentucky Dep’t of Educ., 470 U.S. 656, 665-66 (1985). 182 See Arlington, 548 U.S. at 318 (Breyer, J., dissenting) (“to view each statutory detail of a highly complex federal/state program …simply through the lens of linguistic clarity, rather than to assess its meanings in terms of basic legislative purpose, is to risk a set of judicial interpretations that can prevent the program, overall, from achieving its basic objectives or that might well reduce a program in its details to incoherence”). See also Bagenstos, supra note 184, at 350 (arguing that the Roberts Court is likely to apply Arlington to “skew the interpretation and limit the enforceability of conditional spending statutes”). 183 42 U.S.C. § 2000cc-2(a).

41

have ruled that it does not extend to damages – even though this was clearly Congress’s

intention.184

While reliance on the spending power may present something of a trade-off

because of Arlington, satisfying the Court’s standards for textual clarity with confidence

will generally be easier than satisfying its standards of evidence. A bill contemporaneous

with GINA illustrates how Congress can use the spending power to effectively respond to

Kimel and Garrett, while also avoiding the potential pitfalls of Arlington. The

Employment Non-Discrimination Act (ENDA), which passed the House in the 110th

Congress but stalled in the Senate, included explicit spending-based waiver language.185

The bill, which would prohibit discrimination on the basis of sexual orientation, has been

introduced in Congress beginning in 1994.186 Waiver language was first inserted into the

bill in 2002, shortly after the Garrett decision.187 The 2002 Senate and 2007 House

reports on ENDA provide a stark contrast to the reports on GINA. These reports

specifically invoke both the Fourteenth Amendment and the Spending Clause; discuss in

detail the Supreme Court’s decisions on both abrogation and waiver of state sovereign

immunity; and outline in detail why the bill complies with both lines of decisions.188 The

184 Compare, e.g., Smith v. Allen, 502 F.3d 1255 (11th Cir. 2007), with Madison v. Virginia, 474 F.3d 118 (4th Cir. 2006); Sossamon v. Texas, --- F.3d ----, 2009 WL 382260 (5th Cir. 2009); Williams v. Beltran, 569 F.Supp.2d 1057, 1059 (C.D.Cal. 2008); El Badrawi v. Dept. of Homeland Sec., 579 F.Supp.2d 249, 259 (D.Conn. 2008); Pugh v. Goord, 571 F.Supp.2d 477, 507 (S.D.N.Y. 2008). This distinction is especially dubious because under the doctrine of Ex Parte Young, Congress could have allowed for injunctive relief against states even without a waiver. By far the most likely explanation for the provision’s inclusion, and one states could easily comprehend, is that Congress intended to expand the relief that would otherwise be available. Compare Webman v. Fed. Bureau of Prisons, 441 F.3d 1022 (D.C. Cir. 2006) (applying suit/liability distinction to federal immunity, where Young does not apply). 185 H.R. 110-3685, § 11(b). 186 The version of the bill introduced in the 111th Congress will likely also prohibit discrimination on the basis of gender identity, which civil rights groups insist must be included. See Lou Chibbaro, Jr., Hate crimes, ENDA seen as top legislative priorities, WASH. BLADE (Dec. 5, 2008). 187 Compare H.R. 106-2355 § 13 with H.R. 107-3685 § 13. 188 S. REP. NO. 107-1284, at 19-25 (2002); H.R. REP. NO. 110–406, at 27-30 (2007). As support for abrogation, the reports pointed to “half a century’s worth of severe anti-gay bias in both the state and

42

bill also avoids Arlington problems by linking its remedies provisions to Title VII of the

Civil Rights Act, and by expressly providing for attorney and expert fees.189

Explanations for GINA’s Shortcomings

As enacted, GINA provides less than robust support for abrogation of state

sovereign immunity based on the § 5 power, and no alternative provision for waiver of

that immunity. GINA’s to address the standards set by the Supreme Court is particularly

striking given that GINA is the first new employment discrimination law passed since

Kimel was decided in 2000. Why did Congress fail to clearly invoke its Fourteenth

Amendment power, produce a more extensive record and a more compelling explanation

of the need to apply GINA to the states, or include a provision requiring waiver of state

immunity? Moreover, why did Congress fail to take these steps with GINA when it did

take them with other, contemporaneous legislation? Although it may have been difficult –

perhaps even impossible – to develop a much stronger record in light of the emergent

nature of the problem of genetic discrimination, Congress could still have relied on a

funding-based waiver of immunity, as it did in ENDA.190

private employment contexts.” They noted that such discrimination was in the past “a matter of policy…in many police forces, fire departments, schools, and public agencies of our country.” And they cited numerous legal and historical sources detailing the history of anti-gay discrimination, including the history of constitutional litigation. H.R. REP. NO. 110–406, at 11-15 (2007). The 2007 report cited numerous studies and reports on the continued prevalence of sexual orientation discrimination. Id. at 15-16. It also described numerous individual cases, including ten litigated cases of discrimination by state and local agencies. Id. at 13-17. 189 H.R. 110-3685, §§ 10, 12. Ironically, the Eleventh Amendment was cited as one reason President Bush might veto ENDA – despite the inclusion of a spending-based waiver clause of the type courts have upheld. Office of Management and Budget, Statement of Administration Policy – H.R. 3685 – The Employment Non-Discrimination Act (Oct. 23, 2007). By contrast, the Eleventh Amendment was never raised in objection to GINA. See, e.g., OMB, Statement of Administration Policy – H.R. 1424 – Paul Wellstone Mental Health and Equity Act of 2007 (Mar. 5, 2008). 190 Other potential explanations are also unsatisfactory. If Congress had been simply lulled into complacency by the decisions in Hibbs, Lane and Georgia, it likely would not have gone to the trouble it did with the VRA. And if Congress had willfully rejected the Court’s rulings, it would presumably have said so, as it did when it passed the Partial-Birth Abortion Ban Act of 2003. See Pub. L. 108-105 § 2 (2003).

43

The most likely explanation is that, amidst the many considerations that went into

crafting this complex bill and building bipartisan support for it, justifying GINA’s

application to the states was simply neglected. The contrast with the VRARA and ENDA

is likely explained by the higher profile and more controversial nature of those bills, and

by the fact that Congress is not a unitary actor. Unlike GINA, ENDA and the VRA

preclearance provisions faced longstanding ideological opposition, and a judicial

challenge was nearly a foregone conclusion.191 Additionally, some members of Congress

(along with, of course, their staff and the advocacy groups that help draft legislation) are

presumably more sensitive than others to the possibility of federalism-based

constitutional challenges. For these reasons, lawmakers and advocates crafting the

VRARA, ERPA and ENDA responded to Kimel and Garrett with new language and

findings, while those working on GINA did not.

IV. The Solution: Renewed Congressional Engagement with the Court

The explanation for GINA’s vulnerabilities may well be some combination of

insufficient attention by lawmakers and real hurdles presented by the Court’s

jurisprudence. This suggests a twofold problem for Congress in passing effective

legislation that protects individual rights and applies to public and private entities alike.

On the one hand, lawmakers should give greater and more consistent attention to the

Court’s federalism jurisprudence in the development of legislative language and history.

The foregoing analysis suggests some guidelines Congress should follow:

191 See, e.g., James Thomas Tucker, The Politics of Persuasion: Passage of the Voting Rights Act Reauthorization Act of 2006, 33 J. LEGIS. 205 (2007); Sharon M. McGowan, The Fate of ENDA in the Wake of Maine: A Wake-Up Call to Moderate Republicans, 35 HARV. J. ON LEGIS. 623 (1998) . Indeed, the challenge to the preclearance provisions was filed just weeks after the VRARA was enacted. Northwest Austin Mun. Util. Dist. No. One v. Mukasey, 573 F.Supp.2d 221, 223 (D.D.C. 2008).

44

Congressional intent. Lawmakers should state explicitly, preferably in the

statutory text itself, the legislation’s basis in Congress’s commerce, spending, and/or

Fourteenth Amendment powers. They should also indicate, through a statutory statement

of purpose, the range of constitutional rights Congress is seeking to protect when

invoking the § 5 power. Because some laws, such as the ADA, potentially protect a wide

range of constitutional rights, this statement should use language that is exemplary rather

than exhaustive. Finally, Congress should state explicitly that it intends to abrogate

sovereign immunity pursuant to the § 5 power.

Careful fact-finding. Congress should make a concerted attempt to develop a

factual record that will satisfy the Court’s precedents. This means not only establishing

the general need for legislation, but also amassing substantial evidence that speaks to the

standards of relevance the Court has articulated with regard to state action and

constitutional violations. This record – developed through hearings, solicitation of reports

from relevant organizations, and reports of congressionally delegated agencies – should

be discussed in detail in committee reports, and reflect in specific statutory findings.

Analysis of precedent. Although the Court has never specifically faulted

lawmakers for failing to discuss its prior decisions, Congress can likely strengthen its

case by explaining how the record it has developed, and the remedies it has provided,

comport with the case law. When applying damages remedies to the states, this means

explaining in detail why Congress found a serious threat of constitutional infringements

by states. It also means justifying the precise scope of statutory remedies in terms of that

threat. In contrast to the scant discussion in the reports on GINA, the discussions of

45

congressional authority in the VRARA and ENDA reports read like an opening brief,

setting out a roadmap for courts to uphold the legislation.

Express waiver of immunity. A provision requiring waiver of immunity in

exchange for federal funding can provide a “belt-and-suspenders” approach, avoiding

uncertainties concerning the sufficiency of the legislative record. While reliance on the

spending power can raise “clear notice” problems under Arlington, the risk of relying on

§ 5 alone may be greater, for example, where legislation addresses a relatively new

problem, where much of the evidence identified by Congress may not meet the Court’s

standards of relevance, or where the strength of the supporting record is otherwise in

doubt. Waiver provisions should refer specifically to waiver of immunity and/or to

liability based on receipt of federal financial assistance.

Enumeration of remedies. Because of the Court’s parsimonious approach to

spending-based statutes in Arlington, reliance on this power makes it especially important

to spell out statutory remedies with particularity, including not only a right to sue but a

specific enumeration of remedies. One approach, used in both GINA and ENDA, is to

link the new statute to the remedies provisions of an existing law such as Title VII of the

Civil Rights Act.192 Alternatively, Congress may simply enumerate the range of available

remedies, e.g., equitable and legal relief, including back pay, damages, and reasonable

attorneys’ fees (including expert fees).

192 42 U.S.C.A. § 2000ff-6; H.R. 110-3685, § 10(b).

46

These guidelines should form a “federalism checklist” for all legislation (such as

civil rights laws and other measures protecting individual rights) that could be subject to

federalism-based challenges.193

At the same time, even Congress consistently “doing its homework” cannot

completely guarantee that remedial legislation will not be invalidated or eroded. The

shifting, fact-bound, and splintered nature of the Court’s decisions in this area means that

§ 5 legislation will always face some risk of invalidation. And while funding-based

waivers provide an alternative to this risk in the state immunity context, after Arlington

they may also present some (though probably less) risk of restrictive interpretation of

rights and remedies. Therefore, lawmakers should strive to defend Congressional

authority against further abridgment by the courts.194

To this end, Congress should consider nominees’ views of congressional power in

confirmation votes for all federal judges, as well as relevant executive officials.195

Members of the Senate Judiciary Committee took steps in this direction by raising the

Court’s Commerce Clause and § 5 decisions during the confirmations of Chief Justice

Roberts and, to a lesser extent, Justice Alito.196 Lawmakers should also be more active in

speaking directly to the Supreme Court through amicus briefs when congressional

authority is at issue, providing a potentially powerful supplement to the voice of the 193 Cf. Deborah Widiss, Shadow Precedents and the Separation of Powers: Statutory Interpretation of Congressional Overrides, 84 NOTRE DAME L. REV. 511, 62-63 (2009) (suggesting steps Congress should take to ensure the full effectiveness of overrides of the Court’s statutory decisions). 194 Some have called for stronger responses to constitutional decisions of which lawmakers disapprove, including limiting courts’ jurisdiction or appropriations, or impeaching judges based on their decisions. See, e.g., Mark C. Miller, When Congress Attacks the Federal Courts, 56 CASE W. RES. L. REV. 1015 (2003). In addition to the obvious fact that jurisdiction-stripping would be self-defeating in the context of preserving federal remedies, these responses are objectionable because they would directly interfere with the work of the federal courts and threaten judicial independence. 195 See Charles E. Schumer, Under Attack: Congressional Power in the Twenty-First Century, 1 HARV. L. & POL’Y REV. 3 (2007); Hillary Rodham Clinton & Goodwin Liu, Separation Anxiety: Congress, the Courts, and the Constitution, 91 GEO. L.J. 439 (2003). 196 See Lazarus, supra note 9, at 9-14, 28-29.

47

Solicitor General – as did fifty-one current and former members of Congress in Hibbs.197

Finally, Congress can appeal to the Court’s dependence on its perceived legitimacy

through the use of oversight hearings, resolutions, and other public statements to spotlight

disfavored decisions. The Senate Judiciary Committee held such a hearing in 2002 in

response to the initial run of § 5 cases.198 These congressional efforts to date have been

laudable but sporadic, and should be strengthened.

V. Conclusion

In crafting new legislation, considerations of individual remedies, judicial

enforcement, and possible legal challenges almost inevitably take a back seat to matters

of substance and politics. But following the Rehnquist Court’s “federalism” decisions,

inattention to these matters can have a significant and detrimental effect on the ultimate

effectiveness of legislation. Analysis of GINA, and comparison to the VRARA, ERPA,

and ENDA, suggest that Congress has been inconsistent in its responses to this challenge.

Congress has given the Court’s federalist jurisprudence careful attention when

developing high-profile and controversial bills, but not for lower-profile bills.

If Congress is to uphold its prerogatives and ensure the full effectiveness of

federal laws, its response must be more than sporadic. Due consideration must be given

to the federalism hurdles imposed by the Court in developing any legislation that protects

197 Brief of Amici Curiae Sens. Dodd & Kennedy, and Reps. Schroeder, Roukema & Miller, Nevada v. Hibbs, 2002 WL 31477652 (2002). See also 198 See Narrowing the Nation’s Power: The Supreme Court Sides with the States, 107th Cong. (Oct. 1, 2008) (Statement of Sen. Patrick Leahy). In 2008, the committee held three hearings focused on statutory and preemption decisions that, in the committee leaders’ view, undermined the rights of workers and consumers by limiting remedies and private rights of actions for a wide range of bad acts by businesses. See Protecting Consumers by Protecting Intellectual Property: Hearing Before the S. Judiciary Comm., 110th Cong. (2008); Short-change for Consumers and Short-shrift for Congress? The Supreme Court’s Treatment of Laws that Protect Americans’ Health, Safety, Jobs and Retirement: Hearing Before the S. Judiciary Comm., 110th Cong. (2008); Courting Big Business: The Supreme Court’s Recent Decisions on Corporate Misconduct and Laws Regulating Corporation: Hearing Before the S. Judiciary Comm., 110th Cong. (2008).

48

individual rights against state actors; this includes building a strong evidentiary record,

drafting committee reports that address the Court’s precedents, and including specific

statutory findings, statements of purpose, and private remedies. It will also include using

Congress’s oversight and nominations powers, as well as participation before the Court

itself.

The steps I have suggested may or may not require significant expenditures of

legislative resources or political capital; they will certainly require consistent attention.

Ensuring that widely-accepted laws do not become judicial casualties must become a

routine part of the work of Congress. Otherwise, Congress will continue to pass

important laws like the ADA and GINA – often after years of work to develop nuanced

solutions and build bipartisan support – only to see their impact blunted by the courts on

“federalism” grounds.