In June, the Supreme Court decided the case of Kisor v. Wilkie, which presented the question of whether Auer deference should be eliminated. In another defeat for the movement to restore limits on administrative agency discretion, a majority of the Court refused to overturn Auer v. Robbins and thereby eliminate Auer deference.
Auer deference is the latitude that an agency receives when it interprets regulations it has promulgated. The meaning of such binding regulations is often unclear, but under Auer, an agency’s interpretation can only be reversed if it is unreasonable rather than merely mistaken.
Auer deference is similar to but distinct from its more famous cousin, Chevron deference. Chevron deference provides deference to an agency’s interpretation of a statute that it administers, rather than to an agency’s interpretation of a regulation it has crafted. In many ways, Auer deference is stronger than Chevron deference, because Auer deference has in the past been subject to fewer limitations. For example, whereas Chevron deference is mainly limited to interpretations issued during an agency rulemaking or formal adjudication, Auer has no such limitation and has been applied even to interpretations that are announced in legal briefs. Since agencies promulgate so many regulations, Auer deference turns out to be extremely important.
While it had seemed possible that the five right-wing Justices were ready to eliminate Auer deference, the forces of administrative discretion once again prevailed. As in the recent case of Gundy v. United States involving the nondelegation doctrine, the conservative Supreme Court Justices continue to find ways to lose cases they should win. In this instance, the main culprit was Chief Justice Roberts.
The other eight members of the Court were sharply divided on Auer. The four progressive Justices believed that Auer was correct when it was decided and should be retained. The other four right-wing justices believed that Auer was wrong when it was decided and should be overturned. But Chief Justice Roberts took a middle position. He appeared to believe that Auer was wrong when it was decided (since he did not join the progressives on this) but should not be overturned because of stare decisis. But while this seems like a middle position, it ended up providing the progressives with a victory and the right-wingers with a loss.
Roberts’s position was extremely problematic. To begin with, the majority opinion that he joined appeared contradictory. While it sought to justify not overturning Auer on stare decisis grounds, the opinion significantly changed the law by cutting back on Auer deference in a variety of ways, turning the strong deference doctrine into a weaker one. While that, no doubt, appealed to Roberts as a matter of legal policy, it also rendered the opinion incoherent. In the name of following precedent, the Court changed the law. This harkens back to Planned Parenthood v. Casey, where the Supreme Court justified its decision to not overturn Roe v. Wade based on stare decisis, but then cut back on the protections the law afforded abortion rights.
If the Court was going to cut back on Auer, it should have justified that change. It should have acknowledged what it was doing and explained why. But the Court failed to do so. Moreover, to justify narrowing Auer, but not overturning it, the Court should have explained why the rules of stare decisis forbade overturning the case, but allowed narrowing the precedent. But the Court did not even address the propriety of cutting back on Auer.
While the Court did attempt to justify not overturning Auer, its argument is seriously deficient. First, the Court ignores the law in the most recent case that significantly discussed stare decisis, Janus v. American Federation of State Employees. Janus listed the quality of a decision’s reasoning as the first factor it considered when deciding whether to overrule that decision. But neither the Court’s discussion of stare decisis nor Chief Justice Roberts’s concurrence mentioned this factor.
Second, one of the weightiest arguments in favor of stare decisis—the reliance interests of the public and the government—did not support the majority’s decision. As Justice Gorsuch notes in his dissent, the Court could have overturned Auer but applied stare decisis to the cases where the courts had already given the agencies Auer deference. In that way, disruption would be minimized while still ending the application of Auer to future cases. In fact, reliance interests actually argue against the majority decision. Since the majority cut back on Auer deference, there is now uncertainty about how the courts should treat future cases involving the interpretation of regulations where the courts had previously applied a stronger deference rule than Kisor announced.
While the stated reasons for Chief Justice Roberts voting to maintain Auer are weak, the most obvious reason for Roberts’s vote is the one that accords with the popular perception of Roberts. Under this view, Roberts does not want the “Roberts Supreme Court” to be seen as deciding too many 5-4 cases where the five right-wing justices vote in a block, especially where they are overturning earlier decisions. Thus, Roberts joins the progressive block, because somehow progressives voting in a block is not a problem. But while Roberts votes not to overturn Auer, he nonetheless gets much of his way, because Kisor significantly cuts back on Auer. So Roberts gets the best of both worlds—the Court narrows Auer, but is not seen as overturning a precedent currently dear to progressives.
Yet, there is a real cost here. Auer unjustifiably expanded administrative power, as Justice Gorsuch’s dissent ably shows. Moreover, both Auer and Chevron are cases where the courts expanded administrative power on their own authority. While Congress often delegates authority to agencies, the courts did this on their own. To refuse to correct their error now—through the use of a less than candid justification, perhaps in order to protect the Chief Justice’s vision of the Court’s reputation—is an outrage.