The Trump administration is scrapping discrimination cases that are based on disparate impact analysis.

Many of them probably deserve to be scrapped.

This Washington Post report is called “Discrimination cases unravel as Trump scraps core civil rights tenet.” That “core tenet” is a version, or a perversion, of disparate impact theory. Under that version — the DEI version — policies and practices that don’t distribute burdens and benefits equally to members of all races are deemed inequitable, discriminatory, and untenable.

What’s actually untenable is the assumption that, absent discrimination, burdens and benefits will be distributed, quota style, in proportion to minority representation in the population. As Dan Morenoff, executive director at the American Civil Rights Project (of which I’m a board member) explains:

The people who most appreciate disparate impact appear, usually, to be deeply wed to the idea that any discrepancies are best explained by discrimination., All of us know that’s usually not true — that often different groups of people … just don’t behave the same, don’t prefer the same things, don’t pursue the same ends.

The Post emphasizes, correctly, that disparate impact theory is part of anti-discrimination law. However, the overreaching version of that theory implicit in at least some of the cases that are “unraveling’ is not.

Disparate impact theory took hold as a result of the Supreme Court’s decision in Griggs v. Duke Power. In that case, a power company required employees seeking low-level menial jobs to pass two aptitude tests and possess a high school diploma. These requirements were facially neutral, but excluded a much higher percentage of black applicants than white ones.

The Supreme Court ruled that even neutral rules and selection devices are unlawful under Title VII of the Civil Rights Act if they disproportionately exclude members of a particular race, unless the rules/devices are justified by business necessity. Furthermore, the employer has the burden of establishing business necessity. Subsequent law has softened the business necessity requirement to something closer to a business reasonableness standard.

The same approach can be applied beyond the employment context. For example, it can be applied to a school discipline practices (e.g. disciplining students for a certain number of unexcused absences), policing practices (e.g. stops or arrests for driving a specific number of miles above the speed limit), or lending practices (denying loans to people with credit scores below a certain number).

In each of these examples, it’s likely that the practice will be upheld regardless of its racial impact as long as the standard is universally applied and the cutoff number is reasonable. And this is a key point. It’s not the disparate impact alone that gives rise to the violation. It’s the inability to justify the selection device.

The DEI crowd wants to ignore this point. It wants to find violations based on disparate impact alone.

In addition, it doesn’t want to focus only on discrete selection devices. It wants to infer discrimination from racial imbalance in a workforce (for example), regardless of whether (or to what extent) the imbalance results from selection devices, as opposed to, say, differences in preferences and pursuits.

This brings me, at last, to the disparate-impact-based consent decrees and cases the Trump administration is scuttling. The Post focuses on the consent decrees So will I.

There are three basic reasons why a party under federal investigation for discrimination enters into a consent decree with the Justice Department (for example). First, it might have concluded, based on the evidence, that it will be found guilty of discrimination, absent a settlement.

Second, it might believe it’s not guilty but be unwilling or unable to litigate. This is often the case when the DOJ goes after a town, county, or maybe even a small state. These entities typically lack lawyers with the expertise in discrimination law needed to litigate against civil rights specialists in the government, and don’t have the resources to secure outside help.

Third, the entity under investigation might be run by liberals who, regardless of guilt or innocence, want to impose employment quotas or to radically change police practices to conform more closely with the BLM agenda. In these instances, liberal cities (for example) are in effect colluding with a liberal federal government in furtherance of DEI.

Only the first category of consent decrees is likely to be legitimate. Thus, it’s in the public interest for the Trump administration to review consent decrees that are premised on disparate impact theory. In these instances, the justification for the disparities has never been litigated and may not even have been considered by the party entering into the agreement with the government.

The Post highlights three consent decrees that the Trump administration has unraveled, or hopes to: (1) an agreement with an Atlanta bank accused of discouraging blacks and Latinos from applying for loans; (2) an agreement with the Rapid City, South Dakota school district that disciplines Native American students at a higher rate than whites; and (3) a series of agreements with the Minneapolis Police Department to change its policing practices.

If the Post’s account is correct, the Atlanta bank decree is based on substantial evidence of discriminatory lending practices. It seems to fall into the first category described above.

However, the Trump’s DOJ’s justification for setting aside the decree is that the bank has “demonstrated a commitment to remediation” and has disbursed the entirety of a $7.5 million loan subsidy fund for borrowers in Black and Latino neighborhoods, as required by the agreement. This seems like reasonable grounds to unwind the settlement.

The district court thought so. It approved the unwinding.

The unwinding of the Rapid City, South Dakota school discipline decree seems justified for a different reason — lack of evidence of discrimination. According to the Post:

An investigation of the Rapid City school district by the Education Department’s Office for Civil Rights found that Native students were twice as likely as White students to be referred for discipline, more than four times as likely to be suspended and more than five times as likely to be referred to law enforcement officials.

But what is the evidence that these disparities were the result of discrimination, rather than the neutral application of reasonable rules to all students? The Post cites none.

Remember, too, that schools that loosen their disciplinary practices to redress racial disparities tend to suffer as a result. More precisely, students and teachers, including black students and teachers, suffer due to the breakdown in classroom discipline.

I can’t tell from the Post’s article whether (or to what extent) the Rapid City decree actually mandated or would lead to less enforcement of reasonable disciplinary standards. But the seemingly weak (or non-existent) evidence of discriminatory discipline in Rapid City was probably reason enough to terminate the agreement.

The Minneapolis agreements look like a case of collusion between the leftist Biden DOJ and a leftist city government. For the reasons I presented here, the agreements are a recipe for ineffective policing.

Furthermore, the evidence does not seem to justify the agreements. According to the Post, blacks were stopped, searched, and arrested at disproportionately high rates in Minneapolis. But the Post also says that these disparities resulted from “race-neutral” policies. As long as these policies were justified by legitimate public safety concerns, there is no discrimination, even under disparate impact analysis.

In addition, the agreements do not seem narrowly tailored to address stops, searches, and arrests. For example, the city agreed to limit the use of force by police officers. Was there evidence that the police unjustifiably uses force against blacks to a disproportionate degree? Not that the Post mentions.

In sum, the Trump DOJ absolutely should review decrees entered into and cases brought by the DEI-BLM embracing Biden administration. I don’t have enough information to assess each of the decisions the DOJ has made based on its review. It’s possible that some of these assessments are incorrect. However, looking at the cases highlighted by the Post, it appears that the administration is making reasonable decisions that are consistent with the law.