The Troubled Adolescence of City of Boerne v. Flores
City of Boerne v. Flores,1 the case enunciating the modern standard governing Congress’s power to enforce the Fourteenth Amendment,2 turned 16 in 2013. As with many 16-year-olds, it is unsettled. In a way roughly akin to a child’s sometimes-painful development of a distinct personality, early consensus about Boerne’s “congruence and proportionality” standard3 gave way to conflicts about its proper application. Until recently, though, this disagreement was outweighed by a basic template: legislation enforcing rights the Court itself viewed as fundamental would satisfy that test, as would legislation enforcing the equality rights of groups the Court itself had held merited heightened judicial protection. Regardless of other disagreements, during Boerne’s first decade one could confidently predict a Court majority for such results.
No more. Starting in 2009 with the modern Court’s first abortive encounter with the Voting Rights Act (VRA),4 and continuing three years later with its invalidation of part of the Family and Medical Leave Act (FMLA),5 and the year after with its strike down of the VRA’s pre-clearance coverage provisions,6 enforcement legislation benefitting suspect classes and fundamental rights has been susceptible to judicial strike-down. The VRA and FMLA cases turn on different aspects of the congruence and proportionality test.7 In the VRA cases, the Court first doubted, and then outright found lacking, a factual predicate of pervasive racial discrimination in voting justifying the sovereignty burdens the VRA imposed on a select group of states. In the FMLA case, a plurality of the Court was simply unconvinced that the statute’s self-care provision effectively deterred sex discrimination, the law’s avowed aim.
The Court’s doubts about these laws’ factual and policy bases reflect a pattern that will likely recur with increasing frequency. Stubborn discrimination, or discrimination that takes subtler forms, will prod Congress to enact, respectively, stronger or more indirect remedies that will test the Court’s willingness to defer to Congress’s judgments about their need or effectiveness. At the same time, as antidiscrimination laws successfully suppress the most obvious instances of the targeted discrimination, challengers will continue arguing, as they argued with the VRA, that those laws have succeeded so completely that they have become unnecessary, and thus unconstitutional. These challenges will force the Court to decide how much latitude to accord congressional policy judgments about the need for a particular enforcement statute.
These new ambiguities have joined older ones lurking beneath the surface of the Court’s seemingly predictable applications of Boerne. Most notably, the Court has mistaken its own institutionally limited application of the Equal Protection Clause for the actual constitutional standard against which enforcement legislation should be tested for congruence and proportionality. For example, in Board of Trustees v. Garrett,8 the Court considered a challenge to enforcement legislation benefitting disabled persons, a group it had previously deemed a non-suspect class. The rough approximation of its jurisprudence noted earlier predicts that the Court would strike such legislation down—as indeed it did. However, in Garrett the issue was complicated by the ambiguous constitutional status of that earlier suspect class determination. Garrett’s response to this complexity was simply to deny it.
That denial no longer suffices. The tension between understanding suspect class decisions as true constitutional interpretations and as applications of sub-constitutional decisional heuristics will continue bubbling up. Emerging identity groups will continue to raise equality claims that, because of their political, doctrinal, or practical complexities, the Court will be tempted to vindicate on narrow grounds, without adding to the now-calcified list of official suspect classes. Those groups will also continue advocating for statutory remedies, including enforcement legislation. Judicial review of that legislation will place the Court in the same situation it found itself in in Garrett. If the Court continues to deny the contradiction Garrett papered over, its enforcement power decisions will become ever-less predictable and credible.
The Court’s 2013 decision in United States v. Windsor,9 striking down Section 3 of the Defense of Marriage Act (DOMA), exacerbates this problem. Windsor applied an unusual methodology that ignores the standard equal protection tools underlying the Court’s application of the congruence and proportionality test. It thus raises a fundamental challenge to Enforcement Clause doctrine. To the extent the Court employs Windsor’s method in future equal protection cases, this challenge will only grow.
Thus, the congruence and proportionality standard faces several difficult years. This chapter examines whether, like most adolescents, it will grow out of that troubled period into a mature adulthood or remain mired in teenage awkwardness. Part I begins this examination by exploring the early history of the standard, from its inception in Boerne to Tennessee v. Lane,10 the 2004 case that, in retrospect, marked the close of the doctrine’s first phase. Part II picks up the story. It begins by explaining how the doctrine’s first phase has set the Court on a collision course with legislation addressing emerging groups’ equality claims. It then considers the Court’s confrontations with the VRA and the FMLA’s self-care provision. It explains how these cases pose an additional challenge for the Court, one traceable to the doctrine’s imperative that the Court second-guess Congress’s policymaking. Part III considers how the Court can resolve these challenges, consistent with Boerne’s judicial supremacy foundation. Part IV concludes by speculating on the Court’s likely path.
Part I Early Growing Pains
Boerne commanded agreement across the Court’s ideological spectrum.11 As a statement of judicial supremacy in matters of constitutional interpretation, it appealed to both wings of the Court, each of which had reasons to favor such supremacy when it suited their other goals. Soon, however, the consensus began to crack.
The First Applications
In the first set of post-Boerne cases the disagreement turned on whether Congress had identified sufficient unconstitutional conduct to justify the enforcement statute. In Florida Prepaid Postsecondary Education Expense Board v. College Savings Bank,12 a five-justice majority struck down an enforcement statute abrogating state sovereign immunity from patent infringement suits. The Court concluded that Congress had amassed an insufficient record of state government patent violations depriving patent holders of their property.13 One year later, in Kimel v. Florida Board of Regents,14 the Court rejected the Age Discrimination in Employment Act (ADEA) as legislation enforcing elderly employees’ equal protection rights. The same five justices that formed the Florida Prepaid majority similarly concluded in Kimel that Congress lacked an adequate record of unconstitutional age-based employment discrimination by states.
In addition to finding the record inadequate, Kimel introduced another concept that became crucial in later enforcement power cases. Before examining Congress’s record for evidence of unconstitutional state action, the Court canvassed its own age discrimination jurisprudence. It noted that age discrimination merited the lowest level of equal protection scrutiny—as a non-suspect classification, it received only rational basis review. This lower level of judicial protection provided what Justice O’Connor called “the backdrop”15 for the Court’s evaluation of Congress’s evidence. Essentially, the Court’s deferential attitude toward age discrimination raised the hurdle Congress faced in justifying the ADEA as enforcement legislation. As Justice O’Connor explained, an enforcement statute was less likely to be congruent and proportional to the underlying constitutional violations it targeted when it targeted a constitutionally inconsequential problem. Since, according to the Court, age discrimination was so rarely unconstitutional, for Congress to demonstrate a record of unconstitutional age discrimination it had to show not just a pattern of age discrimination per se, but a pattern so irrational or infused with animus that those examples would have failed even rational basis review.
The Garrett Synthesis
These two doctrinal strands—the requirement that Congress demonstrate a pattern of unconstitutional conduct, and the Court’s use of its own equal protection doctrine as the benchmark for identifying that conduct—combined in the next case to establish the Court’s post-Boerne template. In Board of Trustees v. Garrett16 the same five-justice majority from Florida Prepaid and Kimel rejected the enforcement power foundation of the employment provisions of the Americans With Disabilities Act (ADA). At first blush, Garrett maps neatly onto Kimel. In both cases, the Court considered discrimination to which the Court accorded only rational basis scrutiny, and concluded that Congress had not demonstrated a pattern of discrimination that would fail such review.
However, closer inspection reveals that Garrett presented the Court with a more difficult problem. First, in enacting the ADA, Congress amassed a far more extensive record than that supporting the ADEA. Second, the Court’s disability discrimination jurisprudence was more ambivalent than its age discrimination counterpart. Sixteen years before Garrett, in City of Cleburne v. Cleburne Living Center17 the Court had struck down an instance of disability discrimination. However, determining Cleburne’s relevance to the ADA’s congruence and proportionality itself posed a difficult problem, because of Cleburne’s own ambiguity. On the one hand, Cleburne held that disability discrimination merited only rational basis review, thus suggesting that Cleburne’s facts constituted the rare example of state conduct so arbitrary or infused with animus that it failed even that low level of scrutiny. On the other hand, the Court’s ostensible rational-basis review looked suspiciously strict: the Court imposed evidentiary burdens on the government defendant, and reviewed its justifications with a skepticism inconsistent with traditional rational basis scrutiny. Further complicating the picture, Cleburne’s rationale for denying the intellectually disabled suspect class status focused heavily on the practical difficulties lower courts would face if the Court granted that status, rather than on the actual likelihood that disability discrimination posed a serious risk of unconstitutionality.
Garrett’s response to these two complications revealed the depth to which the dominant five-justice coalition had committed itself to a strict application of the congruence and proportionality standard. First, the Court was exceptionally grudging in its review of Congress’s factual record. Most notably, it restricted the range of relevant evidence to instances of employment discrimination by state governments that a court would likely judge unconstitutional. While at first blush reasonable, these limitations betrayed the majority’s insistence on limiting the effective scope of the enforcement power. For example, the Court justified its refusal to consider discrimination committed by sub-units of state governments on the ground that such sub-units are not protected by Eleventh Amendment immunity. But this explanation conflates the scope of Eleventh Amendment immunity with the scope of Congress’s power to enforce the Fourteenth Amendment; indeed, it also perhaps revealed the Rehnquist Court’s real agenda of protecting states’ sovereign immunity at any cost.18 Eleventh Amendment immunity is surely a major factor in considering the practical importance of the enforcement power, since that power constitutes Congress’s most important regulatory tool allowing it to abrogate state sovereign immunity. But no logically necessary relationship connects the two.
Garrett’s other restrictive glosses on Congress’s record trigger similar doubts. For example, it insisted that the only relevant evidence consisted of discrimination so severe as to likely be judged unconstitutional if challenged in court. That insistence explicitly linked the fate of enforcement legislation to the Court’s own willingness to strike down government action as violating the Equal Protection Clause. This linkage might make sense if the Court’s own equal protection doctrine extended to the full scope of the equal protection guarantee. In that case, the Court would simply be reiterating Boerne’s insistence on distinguishing between legitimate congressional enforcement of the Fourteenth Amendment and illegitimate congressional interpretation. The problem, however, is that judicial doctrine often stops short of enforcing the full limits of the Fourteenth Amendment. Instead, the Court often—and sometimes explicitly—shrinks from applying the Amendment to its full limits, based on concerns about judicial competence and the appropriate judicial role.
Garrett itself illustrates this dynamic. As noted above, one challenge Garrett posed to Kimel’s template lay in the fact that the Court had previously sent conflicting signals about the constitutional status of disability discrimination. In Cleburne, while the Court held that the disabled receive only rational basis scrutiny, its application of that standard suggested it was actually performing more stringent review. Thus, the core constitutional status of disability discrimination, and with it, how much latitude Congress possessed to combat the problem, remained unclear.
This problem was compounded by a second facet of Cleburne. Cleburne’s refusal to accord heightened scrutiny to disability classifications stemmed largely from concerns about courts’ institutional competence. The Cleburne Court worried that courts applying heightened review to disability classifications could not competently distinguish between when differential treatment reflected animus or irrationality and when instead it reflected appropriate judgments about the concededly real differences distinguishing the disabled from mainstream society. It also expressed concern that a decision to grant suspect class status to the disabled would open the floodgates to similar demands from other groups who, according to the Court, occupied a similar social position. These concerns suggest that the Court’s decision to apply rational basis review reflected not the true meaning or full extent of the Equal Protection Clause, but rather judicial recognition that courts lacked the capability to fully enforce that provision.
Despite Cleburne’s ambiguities, the Garrett Court perceived no serious problem citing Cleburne as support for its skeptical review of the ADA. It described Cleburne as applying standard rational basis scrutiny, and brushed aside any suggestion that Cleburne’s institutional competence concerns supported greater congressional latitude to enact enforcement legislation benefitting a group whose constitutional status was ambiguous. Describing the disabled as a group discrimination against which triggered no special level of judicial scrutiny, Garrett proceeded to draw a precise analogy between the ADA and the ADEA that the Court had so easily dispatched in Kimel.
The next enforcement power cases confirmed Garrett’s lessons, even while reaching the opposite result. First, in Nevada Dept of Human Resources v. Hibbs,19 the Court upheld as valid enforcement legislation the FMLA’s family-care leave provisions, which provided employees with uncompensated leave to care for sick relatives. The FMLA suffered from some of the same problems that afflicted the ADA: a somewhat spotty record of state violations of the underlying constitutional right the statute sought to enforce, and findings that were either compiled by a Congress previous to the enacting one or found by a private body, whose conclusions were never translated into formal legislative findings.
Nevertheless, the Court upheld the law, on a 6–3 vote that included two members of the Florida Prepaid-Kimel-Garrett majorities. Writing for the Court, Chief Justice Rehnquist stressed the Court’s recognition of the importance of the sex equality right the FMLA sought to enforce. As the Chief Justice recognized, the Court had long accorded sex discrimination heightened judicial scrutiny. For him, this fact distinguished the FMLA from both the ADEA and the ADA. As he explained, because the FMLA targeted a right the Court itself had identified as constitutionally significant, Congress faced an easier task in demonstrating the existence of constitutional violations justifying the statute’s remedial measures.
The other major case decided during this phase of Boerne’s evolution reinforced the doctrinal trend visible since Kimel, but also introduced a new element. In Tennessee v. Lane,20 the Court confronted another title of the ADA, addressing public accommodations, in a case brought by two disabled persons who, because of their disabilities, were unable to access part of a state courthouse. Writing for the same Hibbs majority (except for Chief Justice Rehnquist, who dissented), Justice Stevens stressed that courts had recognized the constitutional right at issue—the right to access the judicial process—as fundamental.
Justice Stevens’s move had two consequences. First, it allowed him to apply Hibbs’s statement that enforcement legislation targeting a constitutionally significant right enjoys more deferential scrutiny of its factual foundations. Second, it allowed him to describe this application of the statute as congruent to the underlying right. Indeed, he argued, the fit was tight: the ADA required that owners of public accommodations make “reasonable modifications” to allow disabled access, a mandate that closely tracked the constitutional requirement imposed on states to make their judicial processes accessible.
However, in order to trigger this analysis Justice Stevens had to make two preliminary moves. First, he had to identify the right at issue as the due process right to access the judicial process, rather than disabled persons’ equal protection rights. The Court had previously recognized the first of these rights as fundamental; by contrast, Garrett had dismissed the latter as constitutionally insignificant. Characterizing the right as the due process right to judicial access cleared the way for Justice Stevens to cite Hibbs, not Garrett or Kimel, as the relevant precedent.
But the focus on that right was not self-evident from the statute. After all, the ADA’s public accommodations provisions address public accommodations generally, not courthouses in particular. Thus, in a second move, Justice Stevens characterized the case as considering the enforcement power foundation of the ADA’s public accommodations provisions as applied to courthouse access. This statutory parsing provoked a sharp retort from Chief Justice Rehnquist, writing for three of the four dissenters. He argued that this move pre-determined the outcome of the congruence and proportionality test, by excising any non-courthouse applications of the statute from the denominator of the proportionality fraction. Indeed, he argued, Justice Stevens’ move would always result in a finding that the challenged statute was congruent and proportional, since the statute’s sweep would be compared only to applications (here, courthouse access) directly relevant to the case before the court.
One last point about this phase of the congruence and proportionality test’s evolution requires notice. Until Lane, no justice had expressed disagreement with the congruence and proportionality test, even if the justices regularly disagreed on its application. That consensus cracked in Lane, where Justice Scalia wrote a solo dissent abandoning the test. He criticized it as “flabby” and enmeshing the Court in an unseemly review of Congress’s “homework,”21 that is, its fact-finding record. He argued that, with the exception of race, which he described as a central concern of the Fourteenth Amendment’s framers, the Enforcement Clause did not give Congress the latitude to do more than prescribe penalties for actual constitutional violations. With that exception, he therefore defected from even theoretical agreement that Congress enjoyed the power to enact prophylactic legislation proscribing constitutionally permissible conduct in order to deter constitutional violations.
Lane closed the first phase of Boerne’s evolution. It solidified the apparent rule of thumb in which the level of judicial protection of a right largely determined the result of Enforcement Clause analysis. At the same time, Lane revealed the underlying tensions in the justices’ understanding of the congruence and proportionality standard, and, indeed, about that standard’s basic correctness. After Lane the Court did not face a difficult Enforcement Clause issue for several years.22 When it did, those tensions widened, and threatened even the tentative consensus that had developed up to that point.
Part II Awkward Adolescence
After 16 years, Boerne’s doctrinal template faces significant pressure. This Part explains how particular types of enforcement legislation, either already enacted or likely to be enacted soon, challenges that template. It also describes how three recent cases have already widened the cracks in the Court’s Enforcement Clause jurisprudence. Taken together, these challenges have seriously unsettled the Court’s tentative post-Boerne equilibrium.
Emerging Groups and the Suspect Class Problem
The Court’s practice of according skeptical review to enforcement legislation benefitting non-suspect classes triggers the critique that it confuses the Court’s suspect class jurisprudence with core constitutional law. This critique explains the Court’s suspect class jurisprudence as merely a decisional aid that helps it identify equal protection violations within the confines of its limited institutional competence.23 Thus, it argues that the Court’s jurisprudence should not play its current outsized role in Enforcement Clause analysis. Rather, Congress, as the institution best situated to make the value choices equal protection demands, should enjoy significant latitude to conclude that certain types of discrimination are either irrational or based on animus, and thus appropriately targeted by enforcement legislation.24 The Court’s suspect class jurisprudence, while clearly relevant, should provide only a starting point for Enforcement Clause analysis.
The Court has refused to accept this reasoning. Instead, it has largely25 relied on the benefitted group’s suspect class status as a major determinant of enforcement legislation’s constitutionality. This juricentric approach to the congruence and proportionality standard has already caused the Court difficulty. Most prominently, the Court in Garrett provided only a brief and unconvincing response to the argument that its earlier decision in Cleburne effectively acknowledged room for Congress to enact aggressive enforcement legislation targeting disability discrimination.26
This problem will only grow as the Court confronts equality claims pressed by emerging groups, such as those demarcated by sexual orientation, genetic make-up, or obesity (all groups that either have recently won significant equality legislation or have come close). These groups will likely confront a Supreme Court that is unwilling to grant them suspect or quasi-suspect class status. The reasons for this reticence vary. The Court may believe (as it confessed in Cleburne) that suspect class status is too blunt an instrument to judge classifications that may be appropriate in some circumstances but not others. Relatedly, it may believe that some decisions are simply not amenable to the purpose and fit inquiries that flow from a given answer to the suspect class question.27 Or it may believe (again as Cleburne explained) that granting suspect class status to such groups will open the floodgates for analogous groups to demand similar protection. Finally, the answer may simply be that the Court has sworn off creating new suspect classes—not an unreasonable suspicion, given that it last granted such status over 30 years ago, even while finding equal protection violations in several instances of discrimination against an ostensibly non-suspect class.28