Posted on May 16, 2024

Congress is Preparing to Restore Quotas in College Admissions

Stewart Baker, Reason, May 15, 2024

More than two-thirds of Americans think the Supreme Court was right to hold Harvard’s race-based admissions policy unlawful. But the minority who disagree have no doubt about their own moral authority, and there’s every reason to believe that they intend to undo the Court’s decision at the earliest opportunity.

Which could be as soon as this year. In fact, undoing the Harvard admissions decision is the least of it. Republicans and Democrats in Congress have embraced a precooked “privacy” bill that will impose race and gender quotas not just on academic admissions but on practically every private and public decision that matters to ordinary Americans. The provision could be adopted without scrutiny in a matter of weeks; that’s because it is packaged as part of a bipartisan bill setting federal privacy standards—something that has been out of reach in Washington for decades. And it looks as though the bill breaks the deadlock by giving Republicans some of the federal preemption their business allies want while it gives Democrats and left-wing advocacy groups a provision that will quietly overrule the Supreme Court’s Harvard decision and impose identity-based quotas on a wide swath of American life.

This tradeoff first showed up in a 2023 bill that Democratic and Republican members of the House commerce committee approved by an overwhelming 53-2 vote. That bill, however, never won the support of Sen. Cantwell (D-WA), who chairs the Senate commerce committee. This time around, a lightly revised version of the bill has been endorsed by both Sen. Cantwell and her House counterpart, Cathy McMorris Rodgers (R-WA). The bill has a new name, the American Privacy Rights Act of 2024 (APRA), but it retains the earlier bill’s core provision, which uses a “disparate impact” test to impose race, gender, and other quotas on practically every institutional decision of importance to Americans.

“Disparate impact” has a long and controversial history in employment law; it’s controversial because it condemns as discriminatory practices that disproportionately affect racial, ethnic, gender, and other protected groups. Savvy employers soon learn that the easiest way to avoid disparate impact liability is to eliminate the disparity – that is, to hire a work force that is balanced by race and ethnicity. As the Supreme Court pointed out long ago, this is a recipe for discrimination; disparate impact liability can “leave the employer little choice . . . but to engage in a subjective quota system of employment selection.”  Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 652-53 (1989), quoting Albemarle Paper Co. v. Moody, 422 U.S. 405, 448 (1975) (Blackmun, J., concurring).

In the context of hiring and promotion, the easy slide from disparate impact to quotas has proven controversial. The Supreme Court decision that adopted disparate impact as a legal doctrine, Griggs v. Duke Power Co., 401 U.S. 432 (1971), has been persuasively criticized for ignoring Congressional intent. G. Heriot, Title VII Disparate Impact Liability Makes Almost Everything Presumptively Illegal, 14 N.Y.U. J. L. & Liberty 1 (2020). In theory, Griggs allowed employers to justify a hiring rule with a disparate impact if they could show that the rule was motivated not by animus but by business necessity. A few rules have been saved by business necessity; lifeguards have to be able to swim. But in the years since Griggs, the Supreme Court and Congress have struggled to define the business necessity defense; in practice there are few if any hiring qualifications that clearly pass muster if they have a disparate impact.

{snip}

Id. at 35-37. In short, disparate impacts are everywhere in the real world, and so is the temptation to solve the problem with quotas. The difficulty is that, as the polls about the Harvard decision reveal, most Americans don’t like the solution. They think it’s unfair. As Justice Scalia noted in 2009, the incentives for racial quotas set the stage for a “war between disparate impact and equal protection.” Ricci v. DeStefano, 557 U.S. 557, 594 (2009).

Not surprisingly, quota advocates don’t want to fight such a war in the light of day. That’s presumably why APRA obscures the mechanism by which it imposes quotas.

Here’s how it works. APRA’s quota provision, section 13 of APRA, says that any entity that “knowingly develops” an algorithm for its business must evaluate that algorithm “to reduce the risk of” harm. And it defines algorithmic “harm” to include causing a “disparate impact” on the basis of “race, color, religion, national origin, sex, or disability” (plus, weirdly, “political party registration status”). APRA Sec. 13(c)(1)(B)(vi)(IV)&(V).

{snip}

The closer one looks, however, the worse it gets. At every turn, APRA expands the sweep of quotas. For example, APRA does not confine itself to hiring and promotion. It provides that, within two years of the bill’s enactment, institutions must reduce any disparate impact the algorithm causes in access to housing, education, employment, healthcare, insurance, or credit.

{snip}

{snip} Remember that some disparate impacts in the employment context can be justified by business necessity. Not under APRA, which doesn’t recognize any such defense. So if you use a spreadsheet to rank lifeguard applicants based on their swim test, and minorities do poorly on the test, your spreadsheet must be adjusted until the scores for minorities are the same as everyone else’s.

To see how APRA would work, let’s try it on Harvard. Is the university a covered entity? Sure, it’s a nonprofit. Do its decisions affect access to an important opportunity? Yes, education.  Is it handling nonpublic personal data about applicants? For sure. Is it using a covered algorithm? Almost certainly, even if all it does is enter all the applicants’ data in a computer to make it easier to access and evaluate. Does the algorithm cause harm in the shape of disparate impact? Again, objective criteria will almost certainly result in underrepresentation of various racial, religious, gender, or disabled identity groups. To reduce the harm, Harvard will be forced to adopt admissions standards that boost black and Hispanic applicants past Asian and white students with comparable records. {snip}

{snip}