A recent settlement has postponed the Supreme Court's verdict on whether the Fair Housing Act covers practices that, while not intended to discriminate, have a disparate impact on some protected group. However, a development occurred while the case (Mount Holly v. Mt. Holly Gardens Citizens in Action, Inc.) was pending that may help to bring some much-needed clarity to an area where confusion has abounded.  

In February 2013, the Department of Housing and Urban Development issued a final rule on housing discrimination. The rule purports that the FHA covers practices with a disparate impact (and states that the FHA applies to homeowners' insurance). It also specifies that for a practice with a disparate impact to be upheld there must be no less discriminatory alternative that equally serves the covered entity's interest. But while the rule states that its purpose is to provide "greater clarity and predictability for all parties engaged in housing transactions as to how the discriminatory effects standard applies," it does not state how to measure a disparate impact or determine whether one practice has a less discriminatory effect than another. These are, however, rather important issues.

I have previously explained an anomaly in fair lending enforcement arising from the failure of federal regulators to understand certain fundamental aspects of statistics. Since at least 1994, out of concern about the disparate impact of standard lending criteria on minority mortgage loan applicants, regulators have been encouraging lenders to relax lending criteria and otherwise reduce the frequency of adverse lending outcomes.

However, most actions that reduce the frequency of an outcome will tend to increase relative (percentage) differences in rates of experiencing the outcome at the same time that they reduce relative differences in rates of experiencing the opposite outcome. For example, reducing a credit score requirement, while reducing relative differences in meeting the requirement, will increase relative differences in failing to meet it. Unaware of the latter pattern, federal regulators consistently monitor fair lending compliance on the basis of relative differences in adverse outcomes. Thus, by responding to regulator pressures to reduce the frequency of adverse outcomes, lenders increase the chance that the federal government will sue them for discrimination.

In consequence of its applying the discriminatory effects rule to insurers, HUD now faces a suit brought by insurer associations challenging the rule, at least as it applies to insurers. This suit, in the U.S. District Court for the District of Columbia, is just now resuming activity after being stayed while the Mount Holly case was before the Supreme Court. Plaintiffs recently filed a motion for summary judgment on the legal issue of whether the FHA covers disparate impact. If the court rules for the plaintiff insurer associations, that may resolve the issue, subject to appellate review.

But if summary judgment is denied, further litigation of the case will provide an opportunity for the plaintiffs to demand that HUD address exactly how disparate impact is to be measured. That would include, for example, clarifying whether reducing a credit score requirement for securing some desired outcome increases or decreases the disparate impact of the requirement.

As I explained in a recent workshop paper, rationally determining whether reducing the stringency of a requirement increases or decreases the impact of the requirement is a good deal more complicated than simply choosing to measure the impact in terms of relative differences in favorable outcomes or relative differences in adverse outcomes. Ideally, HUD and other agencies enforcing federal fair lending and other antidiscrimination laws will thoughtfully address these issues.

Whether or not that happens, however, requiring HUD to take a stand on how it measures impact may at least obviate the anomaly whereby the government encourages conduct that increases the chances a lender or other covered entity will be sued for discrimination.

James P. Scanlan is a lawyer in Washington. He specializes in the use of statistics in litigation.