Concept
Congruence and Proportionality
Fourteenth Amendment Section 5 enforcement power
Congruence and Proportionality
Section 5 gives Congress power “to enforce, by appropriate legislation, the provisions of this article,” and the verb “enforce” sets the limit.1 Congress may remedy and prevent violations of the Fourteenth Amendment, but it may not change what the Amendment guarantees, because a statute that alters the right no longer enforces the Amendment’s provisions at all.2 A statute does not announce which of those things it does, and the label Congress attaches to it settles nothing. A court therefore needs a method for telling a remedy from a redefinition.
The method asks how closely the statute tracks a right the courts have already defined. If the plaintiff must prove the judicially recognized violation itself, the statute adds a remedy to a wrong the Amendment already forbids, and it is easy to uphold. If the plaintiff can win without that proof, the statute reaches some conduct that courts would not call unconstitutional, and that surplus is where Congress may be moving the line. The surplus is the only thing that could make the statute something other than enforcement, so the court tests it for “congruence and proportionality” between the injury Congress seeks to prevent or remedy and the means it adopted.3 The inquiry sets the size and character of the real constitutional problem against the size and character of Congress’s response.
The test has teeth. Boerne gave Congress wide latitude in placing the line between remedial and substantive measures, but it insisted that the distinction “exists and must be observed.”4 In recent years the Court has applied the test rather strictly, most visibly when Congress has tried to abrogate state sovereign immunity.5 The subsections below take the easy case first, then state the test, then examine its two sides.
Most of the enforcement mechanisms in this unit are private civil actions by individuals against state and local governments, and suits by the federal government to compel state compliance are set aside.6 Many of the cases arise when Congress authorizes damages suits against the States themselves, a subject the later abrogation sections take up. Individuals can sometimes reach unconstitutional state action without abrogation at all: the Court has allowed an injunction suit against a state official in an individual capacity, reasoning that an official who acts unconstitutionally does not act as the state per se, because the Supremacy Clause overrides the unconstitutional state law.7
Parallel Enforcement
A remedy is parallel when it encompasses every element of the judicially recognized Fourteenth Amendment right, so that a plaintiff who wins under the statute has necessarily proved a violation of the Amendment as the courts define it.8 Classification therefore starts with the right and not with the statute. Congress must interpret a right in order to enforce it, but its interpretation is always subject to judicial review, so the court first fixes the scope of the right and then compares the statute’s required elements against it.9
Review of a parallel statute is deferential. The court asks whether Congress could rationally conclude that its chosen mechanism, such as a private action for money damages, is an “appropriate” way to redress or prevent violation of the right. Consistent with McCulloch v. Maryland, a court is quite unlikely to second-guess that judgment, and creating parallel private rights of action is a standard and constitutional exercise of the enforcement power.10 The same standard governs Section 5 means generally: so long as the means are rationally related to an objective that one of the Civil War Amendments was designed to serve, Congress has not exceeded its authority, and the special congruence and proportionality standard is the one important limit for the Fourteenth Amendment.11
Heightened review is unnecessary because a parallel statute cannot change the right. Boerne held that Congress has the power “to enforce,” not the power to determine what constitutes a constitutional violation.12 A statute that makes the plaintiff prove every element of the judicial definition leaves that definition where the courts put it. What the statute adds is a remedy, a forum, or a sanction, and those are questions about means, on which Congress receives the usual deference.
Parallel enforcement is not limited to equal protection claims. Boerne noted that the liberty protected by the Due Process Clause embraces the liberties guaranteed by the First Amendment, and it repeated that there is “no doubt of the power of Congress to enforce by appropriate criminal sanction every right guaranteed by the Due Process Clause of the Fourteenth Amendment.”13 A statute that punishes the deprivation of a right the courts already recognize follows the same logic as a damages remedy. It enforces the right as defined, and the choice between prosecution and a civil action is a judgment about means.
The equal protection ban on racial discrimination shows a parallel remedy with its elements visible. A claimant must prove state action, a purposeful racial classification, and a disproportionate racial impact. If Congress creates a remedy, such as damages, and requires the plaintiff to prove all three, the remedy easily passes as a rational method of enforcing the recognized right.14 Because state action is one of the required elements, the statute stays aimed at the conduct Section 1 reaches.
Real legislation makes the same move. Section 2 of the Voting Rights Act, as Congress amended it in 1982, bars any voting practice that “results in the denial or abridgement” of the vote on account of race, whether or not the practice has a racially discriminatory purpose.15 Because an equal protection claim of racial discrimination requires a purposeful classification, a plaintiff can prevail under a results test without proving the judicially recognized violation.
The parallel statute and the nonparallel statute share a purpose, a target, and a remedy. They separate on one question: must the plaintiff still prove every element of the judicial definition? On an exam, compare the statute’s required elements with the courts’ definition of the right. If no element has been dropped, apply rational basis and uphold the remedy. If an element is missing, the statute is nonparallel, whatever label Congress gave it.
The Nonparallel Test
A statute that reaches beyond the judicial definition of a Fourteenth Amendment right must show that its extra reach is still enforcement, and City of Boerne v. Flores supplied the way to show it. The Court conceded that the line between measures that remedy or prevent unconstitutional action and measures that change the governing law is “not easy to discern,” and it left Congress “wide latitude” in placing the line. It insisted nonetheless that “[t]here must be a congruence and proportionality between the injury to be prevented or remedied and the means adopted to that end.”17
The extra test applies only to nonparallel statutes because of what such a statute does. A parallel remedy leaves the judicial definition intact, so it cannot alter the right. A nonparallel statute reaches some conduct that courts would not hold unconstitutional, and that surplus can serve either of two ends: a reasonable precaution against violations too hard to catch one at a time, or Congress’s own view of what the Amendment should require. The two look identical on the face of the statute, so the court looks to the connection between the ban and real violations. Legislation that lacks the connection, the Court warned, “may become substantive in operation and effect.”18 A preventive ban on certain types of laws may be appropriate when there is reason to believe that many of the laws it reaches have a significant likelihood of being unconstitutional.19
The stakes explain the Court’s insistence. If Congress could define its own powers by altering the Amendment’s meaning, the Constitution would no longer be superior law unchangeable by ordinary means, and shifting legislative majorities could change it and circumvent the amendment process of Article V.20 The same worry explains the Court’s treatment of Katzenbach v. Morgan. Morgan’s second rationale, that Congress could find a literacy test denied equal protection despite the Court’s contrary holding in Lassiter, accords Congress authority to define the Amendment’s meaning,21 and Boerne said that reading of Morgan is neither a necessary interpretation nor the best one.22
In practice the test demands three things. Congress must identify a pattern of state violations of a judicially recognized constitutional right, enact a statute plainly designed to ameliorate those violations, and devise a remedy tailored to the demonstrated pattern.23 Those demands have a natural order, so the test runs most smoothly as a checklist.
The first three steps ask the congruence question: does a real constitutional injury exist for the statute to answer? The fourth asks the proportionality question: does the response fit that injury? The opinions treat the two words as one comparison between the size of the problem and the size of the response, so a thin record and a sweeping remedy tend to reinforce each other, and a court may rest its decision on either. Deference survives inside the test: “It is for Congress in the first instance to ‘determin[e] whether and what legislation is needed to secure the guarantees of the Fourteenth Amendment,’ and its conclusions are entitled to much deference.”24 The deference attaches to the facts Congress found and the means it chose. The courts keep for themselves the question whether the statute has crossed from remedy into redefinition.
Boerne shows both questions going against Congress. The Religious Freedom Restoration Act (RFRA) restored the Sherbert compelling-interest test for all cases where free exercise is substantially burdened,25 so laws valid under Smith could fall under it whether or not they were aimed at free exercise.26 The Court looked for evidence that RFRA answered neutral-looking laws passed from religious bigotry and found none. The legislative history was bereft of “modern instances of generally applicable laws passed because of religious bigotry,” and what it did contain were neutral laws with incidental burdens, so that Congress’s concern was with the burdens and not with the object or purpose of the laws.27 The Court later summarized the record as anecdotal evidence that did not reveal a “widespread pattern of religious discrimination in this country.”28 It added that, whatever the record showed, RFRA was “so out of proportion to a supposed remedial or preventive object that it cannot be understood as responsive to, or designed to prevent, unconstitutional behavior.”29 That mismatch made the statute look like a change in the right and not a remedy for violations of it.
The Patent Remedy Act failed on the same two grounds. Congress “identified no pattern of patent infringement by the States, let alone a pattern of constitutional violations,” and because few state infringements were likely to be unconstitutional, the Act’s scope was out of proportion to any remedial or preventive object.30
Congress can strengthen a showing of proportionality with limits that confine the remedy to the problem, but the Court has said that Section 5 legislation does not require termination dates, geographic restrictions, or egregious predicates.31 The proportionality subsection returns to those features.
The demand of the test has been debated. The phrase “congruent and proportional” comes from the language of equity used to limit the scope of injunctions, and it suggests a remedy that is related to the violation and not much over- or underinclusive in relation to it. Some ask whether the phrase raises scrutiny from the rationality review of McCulloch v. Maryland to a kind of intermediate scrutiny, and others argue that Section 5 legislation should be tested by McCulloch’s rational relationship standard instead.32 The Court’s recent strict applications answer the practical question,33 and the working lesson is that a nonparallel statute survives only when Congress can point to a real pattern of state violations and a remedy sized to that pattern.
Congruence with Recognized Rights
Congruence asks about the right being enforced: which Fourteenth Amendment right the provision serves, how closely the courts protect it, and whether Congress found state violations of it. The injury Congress may prevent or remedy is a violation of the right as the courts define it,35 and Boerne measures the means against that injury.36 A statute can fail in two ways. The right may protect so little that the statute forbids mostly lawful state conduct, or the record may show no pattern of the violations the statute is supposed to prevent.
The courts’ standard of review decides how much state conduct the Constitution condemns, and that in turn decides how much room Congress has. Under rational basis review a classification survives if it is rational,37 so the Constitution condemns only irrational distinctions and tolerates many that are unwise or unfair. A statute that forbids every unjustified distinction bans a large volume of permitted conduct, and the more lawful conduct a ban reaches, the less it resembles a safeguard against violations and the more it resembles a new standard. Heightened scrutiny reverses the proportions. When the courts demand a strong justification for a classification, or protect a fundamental right, a larger share of the conduct the statute reaches is likely to be unconstitutional. That proportion is the one Boerne looked to, since preventive measures prohibiting certain types of laws may be appropriate when many of the laws affected have a significant likelihood of being unconstitutional.38
The first two steps fix the baseline and the third measures the gap between the baseline and the statute. The fourth step carries more weight as the gap widens, because where the Constitution tolerates most of what a statute bans, only a documented pattern of state violations can show that the ban works as prevention and does not simply impose Congress’s own rule.
The age cases show the last two steps going against Congress. Age is not a suspect class, so a state age classification is constitutional if it is rational, and a state may, for example, impose a mandatory retirement age on certain employees.39 In Kimel v. Florida Board of Regents the Court examined the Age Discrimination in Employment Act (ADEA) as applied to the States and found that Congress “had virtually no reason to believe that state and local governments were unconstitutionally discriminating against their employees on the basis of age.” The Court added that a thin record is not determinative of the Section 5 inquiry, but that Congress’s failure to uncover any significant pattern confirmed that “broad prophylactic legislation” was unnecessary in this field. In light of the Act’s indiscriminate scope and the lack of evidence of widespread and unconstitutional age discrimination by the States, the ADEA was not valid Section 5 legislation, and its purported abrogation of state sovereign immunity was invalid.40
Disability produced the same structure. In Board of Trustees of the University of Alabama v. Garrett, state employees sought damages from their state employers under Title I of the Americans with Disabilities Act (ADA). The Constitution requires only that a state’s treatment of disabled persons be rational, and Congress had not documented a pattern of irrational, unconstitutional state discrimination against them, so Title I created equality rights beyond those the Amendment guarantees and could not be sustained under Section 5. Kimel and Garrett involve different classifications, but both are decided at the same two points: the right receives only rational basis protection, and the record lacks a pattern of unconstitutional state conduct.
Sex discrimination changes both answers. In Nevada Department of Human Resources v. Hibbs the Court held that Congress properly used Section 5 to abrogate state sovereign immunity for damages suits by state employees under the family-care provisions of the Family and Medical Leave Act (FMLA). The majority determined that the FMLA “aims to protect the right to be free from gender-based discrimination in the workplace.” By creating an across-the-board, routine employment benefit, Congress sought to end the stigma of treating family-care leave as a drain caused by female employees and to stop employers from evading leave obligations by hiring men. And because sex-based distinctions by states receive heightened scrutiny, Congress had an easier time meeting the standard.41
The FMLA’s self-care provision fared differently, and the contrast shows that the right each provision enforces is what counts. In Coleman v. Court of Appeals of Maryland, Justice Kennedy’s opinion held that damages suits under the self-care provision could not proceed against the States. Evidence of disparate impact alone does not make out a constitutional violation, and “it is unlikely that many of the [neutral leave policies] affected by the self-care provision are unconstitutional,” so the provision’s scope was “out of proportion to its supposed remedial or preventive objectives.”42 One statute thus passed in the part that enforced the right against sex discrimination and failed in the part that enforced no judicially recognized right. The Court’s own words show how far the two sides of the test blur: a congruence failure was stated in proportionality language.
Two cautions follow for exam answers. Outcomes vary with the right at stake, so the pattern across the cases is more reliable than any single holding. The pattern is dependable: the Court’s own heightened scrutiny of the underlying classification generally gives Congress more room to show a pattern of state violations.43 The next subsection shows how the ADA can pass in one setting after failing in another.
Proportionality of the Remedy
Proportionality assumes a real constitutional evil and asks about the remedy, which must show “congruence and proportionality between the injury to be prevented or remedied and the means adopted to that end.”45 A proportionate remedy covers the places, the actors, and the practices where the violations are. A disproportionate one covers much more, and it begins to operate as Congress’s own rule for the States instead of as enforcement of the Amendment.
Size matters because of the ratio of lawful to unlawful conduct a remedy displaces. Where violations cluster, a broad prophylactic rule mostly catches conduct that would be unconstitutional anyway, and the occasional lawful practice swept in is the price of reaching violations that case-by-case litigation misses. Boerne’s example was the Voting Rights Act: because jurisdictions with a demonstrable history of intentional racial discrimination create the risk of purposeful discrimination, Congress could prohibit changes with a discriminatory impact in those jurisdictions.46 Spread the same rule across every government and every subject, and nearly everything it catches is lawful under the courts’ definition of the right.
Intrusion on state authority adds a second consideration. The Court called RFRA’s demand for searching scrutiny of state law “a considerable congressional intrusion into the States’ traditional prerogatives and general authority to regulate for the health and welfare of their citizens.”47 The cases weigh the precise constitutional evil, the breadth of the remedy, the depth and pervasiveness of the problem, and whatever limits the remedy to that problem. The more generalized the evil and the response, and the more intrusive the response is on state prerogatives, the less likely the Court is to find the test satisfied.48
RFRA fails nearly every one of these questions. It applied to every agency and official of the federal, state, and local governments, covered all law whether adopted before or after its enactment, and had no termination date or termination mechanism, so any law could be challenged at any time by anyone alleging a substantial burden on religious exercise.49 It also demanded a compelling interest and the least restrictive means, which the Court called “the most demanding test known to constitutional law.”50 Each feature enlarged the share of lawful state conduct the statute displaced.
The Voting Rights Act measures answer the same questions the other way. In upholding the Act, the Court observed that Congress had intentionally confined its remedies to a small number of States and political subdivisions, most of them familiar to Congress by name, and called that a permissible method of dealing with the problem (step two).51 The preclearance requirement upheld in City of Rome v. United States applied only to jurisdictions with a history of intentional racial discrimination in voting, it permitted a covered jurisdiction to avoid preclearance under certain conditions, like the provisions upheld in South Carolina v. Katzenbach, and it lapsed in seven years (steps two through four).52 These statutes still displaced state election rules, so intrusion alone did not defeat them. Their limits kept most of what they displaced tied to the discrimination that justified them.
Those limits are evidence of proportionality and are not elements of it. Boerne said that Section 5 legislation does not require termination dates, geographic restrictions, or egregious predicates, though where a measure pervasively prohibits constitutional state action in order to remedy or prevent unconstitutional state action, such limits tend to ensure that Congress’s means are proportionate to legitimate ends.53
United States v. Morrison shows a remedy that fails the first two questions before duration or intrusion needs asking. Congress had created a civil damages action against private individuals who commit gender-motivated violence, and the Court held the remedy not proportionate to any injury Section 5 reaches. It was aimed at individuals, not at the state officials whose laws and proceedings the Amendment governs, and it applied uniformly throughout the Nation although Congress’s own findings showed that discrimination against victims of such crimes did not exist in all States or even most of them. The two defects compound. A remedy aimed at private persons touches no state conduct that could itself be unconstitutional, so the ratio of lawful to unlawful conduct displaced cannot favor the statute, and nationwide application then places the same burden on States where Congress found no problem. Morrison also announced a separate state-action rule, which the next section takes up.
Proportionality is measured against a particular target. In Tennessee v. Lane the plaintiffs sued under Title II of the ADA, which bars States from denying disabled persons access to public services, and the denial they alleged was of access to state courts, a potential violation of the judicially recognized fundamental right of access to the courts. Title II’s language also reaches nonfundamental services such as recreation facilities, and measuring the remedy against the whole text would charge the statute’s weakest applications against its strongest. The Court instead held the damages remedy proportional to the specific right the plaintiffs claimed. That distinguishes Garrett, where the only judicially recognized protection was freedom from irrational discrimination, and it leaves Title II with a conditional validity: in settings where the plaintiff invokes only that weaker protection, the same remedy might exceed Section 5.
The two sides of the test can fail together, which is how proportionality connects back to congruence. In Kimel the ADEA reached state employers across the board although age classifications need only be rational, and the Court held it invalid in light of its “indiscriminate scope” and the lack of evidence of widespread and unconstitutional age discrimination by the States.54 The scope finding is a proportionality finding and the evidence finding is a congruence finding, and each reinforces the other. The more a statute forbids beyond what the right protects, the more the record must show, and a thin record leaves nothing to justify the breadth. A disproportionate remedy can also decide a case alone, because Boerne held that RFRA could not be remedial regardless of the state of the legislative record.55 A nonparallel statute must clear both sides, and failing either one invalidates it.