The Department of Education's Obama-Era Initiative on Racial Disparities in School Discipline: Wrong for Students and Teachers, Wrong on the Law—Part 3

Title VI is not a disparate impact statute, and executive agencies do not have the authority to transform it into one through rulemaking.


This is the third blog installment discussing my article (written with Alison Somin)—The Department of Education's Obama-Era Initiative on Racial Disparities in School Discipline: Wrong for Students and Teachers, Wrong on the Law.

In the first installment, I tried to give readers a taste of why the Obama Administration's still-in-effect Dear Colleague Letter, which declares disparate impact in school discipline to be illegal, is bad policy. It forces teachers to tolerate more classroom disorder; disorder makes learning more difficult.

In the second installment, I foreshadowed the legal argument I will make here by pointing out that disparate impact liability makes everything or almost everything presumptively illegal. There is hardly any action a Title VI federal funding recipient might take that doesn't have a disparate impact on some national origin, race or color group. (Remember, there are a lot of national origins out there.) Consequently, liability for disparate impact essentially means that funding recipients must be prepared to justify virtually everything they do as "necessary" (as defined by Griggs v. Duke Power (1971) and its progeny). That's asking a lot of them. Who among us can justify all our actions that way?

In this installment, I want to explain why that means the Obama Administration's initiative is unsupported by law.

The Obama Administration never claimed that Title VI itself imposes liability for disparate impact liability. It is clear that it doesn't. Nor did the Obama Administration claim that a regulation duly promulgated under Title VI prohibits disparate impact in the context of school discipline in particular. Instead, it claimed that two identical regulations—34 C.F.R. § 100.3(b)(2) and 24 C.F.R. § 104(b)(2), which were originally promulgated in 1965–extend Title VI's reach to cover all actions of federal funding recipients that have a disparate impact. I'll call these regulations "all-purpose disparate impact regulations."

The article does not argue that agencies cannot promulgate "targeted disparate impact rules" (by that I mean disparate impact regulations that apply only to some specific kind of activity, e.g. out-of-school suspensions). I'm inclined to believe they can, but that's an issue for another day, since no agency has done so. At minimum, however, if an agency were do so, it would have to (1) build a record showing that in order to deter actual discrimination in that targeted area, it is necessary to issue a regulation that covers disparate impact in that area as well as actual discrimination, and (2) allow a defense to a funding recipient that can prove it was not motivated by race, color or national origin. In effect, an agency should be permitted to promulgate such a rule only if it closely tailors the rule to the task of enforcing Title VI's actual prohibition.

An all-purpose disparate impact regulation is something else entirely. Such a regulation would increase the scope of original statute exponentially, thus making it clear that its purpose is not simply to enforce Title VI, but rather to extend it. That is beyond the scope of any agency's rule-making authority.

The proper comparison is to Congress's authority to pass legislation pursuant to Section 5 of the Fourteenth Amendment. Why? Because Title VI has already been authoritatively interpreted by the Supreme Court to be co-extensive with the Fourteenth Amendment's Equal Protection Clause in UC Regents v. Bakke (1978) (and hence it is not a disparate impact statute under Village of Arlington Heights v. Metropolitan Housing Development Corp. (1977)). Title VI simply applies the Equal Protection Clause (as it concerns race, color and national origin) to federal funding recipients. If a court were to determine that Congress doesn't have the authority to transform the Equal Protection Clause into an all-purpose disparate impact statute pursuant to its Fourteenth Amendment Section 5 power, it is difficult to see how an executive agency would have greater authority to transform Title VI pursuant to its Title VI rulemaking power.

That brings us to City of Boerne v. Flores (1997), which tells us that the test for determining whether Congress is acting within the scope of its Section 5 power is one of congruence and proportionality. The same test should apply to the Title VI regulations.

City of Boerne concerned the Religious Freedom Restoration Act ("RFRA"), which, among other things, purported to presumptively prohibit state action that places a "substantial burden" on the free exercise of religion. See also Board of Trustees of the University of Alabama v. Garrett (2001)(a slightly later case applying the City of Boerne framework in the context of the Equal Protection Clause).

The similarities between City of Boerne and the Title VI regulations are many:

  • Just as the Supreme Court has authoritatively held that Fourteenth Amendment free exercise violations occur only when the state acts purposely to constrict free exercise, see Employment Division v. Smith (1990), the Supreme Court has authoritatively decided that the Equal Protection Clause (and hence Title VI) is violated only when the state (or a recipient of federal funds) intentionally discriminates. See Village of Arlington Heights, Bakke, and Alexander v. Sandoval (2001).
  • RFRA, the statute at issue in City of Boerne, applied whether or not the state intended to impose a substantial burden. Similarly, the supposedly all-purpose disparate impact regulations promulgated pursuant to Title VI, apply regardless of intent. In neither case was there any record showing that such a broad approach was necessary to effectuate the underlying Constitutional prohibition (in the case of City of Boerne) or the underlying statutory prohibition (in the case of Title VI).
  • Under RFRA, to overcome that presumption against state action that substantially burdens the free exercise of religion, a state would have had to demonstrate a "compelling purpose." This is not the same as demonstrating that the state did not have the purpose to interfere with the free exercise of someone's religion. It's a much tougher standard. Similarly, to overcome the presumption against actions with a disparate impact, under the Obama Administration's interpretation of the Title VI regulations, a recipient of federal funds would have to demonstrate necessity. This, again, is not the same as demonstrating lack of discriminatory intent. It's hard avoid the impression that in both cases, the idea is to go further than the underlying law.

In City of Boerne, the Supreme Court held that Section 5 authorizes Congress only to enforce the substantive guarantees of the Fourteenth Amendment. It does not authorize Congress to extend those guarantees, much less extend them as broadly as RFRA did. Since this provision in RFRA was much too broad to be a "congruent" and "proportional" response to actual Fourteenth Amendment violations, it was held to be beyond the scope of Congress' Section 5 powers.

By the same logic, the Title VI regulations are not a congruent and proportional response to actual Title VI violations. They are extraordinarily overbroad.

(By the way, we also disagree with the Obama Administration that the relevant regulations are in fact all-purpose disparate impact regulations (although the Obama Administration was not the first to interpret them that way). But all of that is a separate argument. Our main argument (or at least my favorite argument) is that if 34 C.F.R. § 100.3(b)(2) and 24 C.F.R. § 104(b)(2) were meant to be an all-purpose prohibitions on disparate impact, they were beyond the scope of the executive branch's rulemaking authority under Title VI.)

NEXT: Mindgeek, or the Biggest Digital Streaming Platform You've Never Heard Of

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Title VI, like Title IX, section 504, and countless other federal laws are an exercise of spending clause power. The leap from “congruence and proportionality” as a constraint on congressional lawmaking power under section five to a constraint on lawmaking power under the spending clause power requires more than an ipse dixit. Otherwise, the argument necessarily invalidates virtually all spending clause legislation that binds state and local governments, certainly including section 504 of the Rehabilitation Act and its rules, which we already know greatly exceed the permissible exercise of section five powers. I can’t find any support for that in any SCOTUS spending clause decision; to the contrary those cases presume spending clause power is sufficient because states can always turn down the money. In short, City of Boerne and all the other cases on section five seem irrelevant to spending clause power through legislation and otherwise proper rulemaking. That’s not to say that the rule is a proper exercise of rulemaking power, but it is to suggest that the congruence and proportionality cases do not provide an answer.

  2. I generally think the business of interpreting statutes maximally for the sole purpose of striking them down should be disfavored.

    Disparate impact is a judicially created doctrine, and the courts can interpret their own doctrine in a workable as distinct from non-workable fashion. It’s true that pretty much everything affects everything else, just as Justice Scalia once noted that to a curbside philosopher, everything is related to everything else. But just as the Court did with ERISA, disparate impact can require the impact to be substantial, precisely to avoid the absurd result of de minimis impacts constantly raising court cases over everything.

    The fact that by Professor Heriot’s interpretation the concept of disparate impact is absurd and leads to absurd results is an excellent reason not to use Professor Heriot’s interpretation. It is a very poor reason to strike down or refuse to apply the doctrine.

    1. @ReaderY
      IIUC you prefer the Parkland/MSD method of ignoring criminal actions at schools by minority students?
      Why do you hate children?
      Why do you hate white people?
      Why do you hate teachers?

      1. …Dunkirk…dark days of the war…backs to the wall…Alvar Liddell…Berlin air lift…moral upheaval of Profumo case…young hippies roaming the streets raping, looting, and killing. Yours etc. Brigadier Arthur Gormanstrop (Mrs.)

        ?Monty Python, Police Raid Sketch

    2. The problem with that approach, ReaderY, is that there is no other approach that contains any limiting parameter. Simply inserting the word “substantial” adds no useful guidance and puts no clear limits on prosecutorial discretion.

      1. As Justice Holmes noted, there is no distinction between day and night, only shades of gray, yet despite the complete absence of any limiting principle, the concept remains useful. Numerous legal concepts, from “excessive” speed to “gross” or “reckless” negligence, involve constructing artificial cutoffs on continuums that have no inherent limiting principle, yet are given a reasonable meaning in a common-law fashion through legal decisions.

        People routinely accept this for policies they agree with, and find problems with it only for policies they object to.

    3. “Disparate impact” isn’t judicially created. The Necessary and Proper clause, for example, arguably mandates legislation be just and conform to structural principles such as the equal sovereignty of the States. In addition, Congress’s power to tax and spend is/was textually limited to the “common Defence” and “general Welfare of the . . . States.” Many argued that the Tariff of Abominations was unconstitutional precisely because of its disparate impact (benefits and burdens), and lack of necessity (surplus with no debt meant “tax” was not levied for any constitutional purpose).

      Even postbellum amendments such as the 14th imply that judges must consider disparate impact situations.

      Now, judges should be careful not to over-extend “disparate impact” considerations, and it’s probably best to cabin it to intent of the law where its facially neutral and where the plaintiff or defendant, as the case may be, can prove it up.

  3. Finally, if it is true that there are school districts that routinely mete our much harsher punishments to black children than they do to white children for the objectively same offenses, and the impact is clearly substantial rather than de minimis, I think addressing the issue would easily fall within Congress’ 14th Amendment enforcement powers.

  4. imagine (if isn’t hard to do) a police officer who simply feels more scared around black people than white people. He doesn’t intend to be. He just is. And when he feels frightened, he tends to shoot.

    So we could easily get a police officer who shoots black people more, perhaps much more, without forming a specific conscious intent to do so. We could get a whole police force full of them.

    Professor Heriot seems to be interpreting the Equal Protection Clause rather narrowly if it doesn’t and can’t encompass behavior like this. Most people who commit crimes don’t announce intent – intent is discerned from the circumstances. If one has to announce ones intent to commit a crime for it to be covered, coverage is limited indeed.

    The difficulty with Professor Heriot’s argument is she is using the maximally unintentional position – relationship is a pure statistical anomaly – to describe all behavior short of announcing ones intent. She constructs her definitions so as to simply define visceral discrimination (discrimination of a sort that is sensed but not consciously articulated) away.

    Under her definition, as long as one doesn’t announce ones intentions so as to make it come under narrow “intentional” discrimination, one can basically do anything one wants and get away with it.

    The 14th Anendment is not so powerless, nor so absurd.

    1. Does disparate impact mean a racially disparate impact on students whose cases are, apart from race, identical?

      Say a black student and a white student both decide to gang up on another student and beat him up. Both perps start their crime at the same time, inflict the same number of blows, and so on, they both have the same kind of prior record – yet the white student gets a slap on the wrist and the black student is expelled.

      That sounds fairly disparate. It sounds like enough to bring the school to account, even without a smoking-gun racist memo.

      Or does disparate impact mean misleading comparisons among cases without pausing to inquire whether they’re comparable?

      Because it’s the latter version that we hear about at Parkland, etc.

  5. Michael Masinter said “I can’t find any support for that in any SCOTUS spending clause decision; to the contrary those cases presume spending clause power is sufficient because states can always turn down the money. ”

    In that case, there should be no problem withholding grant money from sanctuary cities, right?

    1. Actually, we can withhold money. But it is the job of congress, not the executive to set those conditions.

Please to post comments