During the Gorsuch nomination, there was a lot of talk in the press about the Chevron doctrine. Most people have never heard of this doctrine, and only a few are aware of all the nuances. As the Trump Administration’s rulemaking efforts come before the courts, we’re going to be hearing a lot more about it. As the  Court’s 2017 Term  gets underway, I thought this would be a good time to give a roadmap to the doctrine. In fact, we’ve already seen a short opinion from Justice Gorsuch questioning the application of the doctrine.

The Chevron doctrine is a rule about court review of agency actions that many scholars consider central to modern administrative law. That doctrine calls for judges to accept reasonable interpretations of a statute by an administrative agency, even if the judges might have favored different interpretation themselves.  The Supreme Court has cited two reasons to give agencies the power to interpret ambiguous statutes: (1) agencies are more democratically accountable than courts, and (2) Congress has given the agency the main responsibility for implementing the statute.

Even before Chevron, courts agreed that an agency’s statutory arguments were entitled to respectful consideration, given an agency’s expertise about the statute it administers and the practical and technical issues involved in implementing the law. (This is now known as Skidmore deference.) But Chevron upped the level of deference to agencies.  It created a two-step test:

Step One.  Is the statute’s meaning clear?  If so, that meaning controls.

Step Two.  If the statute is ambiguous, the agency’s interpretation will be upheld by the court provided it is reasonable, even if the court would have chosen an alternative interpretation.

The application of Chevron was limited by United States v. Mead Corp. The Mead Court held that Chevron deference should apply only where “Congress delegated authority to the agency generally to make rules carrying the force of law, and that the agency interpretation claiming deference was promulgated in the exercise of that authority.” In Mead, the Court decided that the procedures Congress set out for the agency did not indicate any intention to grant the agency rulemaking authority, so the agency received only Skidmore deference. Its views were considered respectfully due to the agency’s expertise, but ultimately the Court made its own independent judgment of how to interpret the statute.

After Mead, the Chevron test had three steps:

Step Zero. Does the agency have authority to issue binding legal rules? If the answer is “no,” Chevron does not apply, but the agency may still receive some lesser degree of deference because of its expertise. If the answer is “yes,” the analysis moves to Step One.

Step One. Is the statute ambiguous? If not, the Court simply decides the interpretation of the statute by itself. Otherwise, the analysis moves to Step Two.

Step Two. Is the agency’s interpretation reasonable (even if the court itself would have chosen a different interpretation)?

Step One of Chevron enables reviewing courts to preserve their traditional authority over determining statutory meaning. If the Court proceeds to Step Two, however, the “permissibility” inquiry is similar to arbitrary or capricious review: if the agency’s interpretation is reasonable, it has typically been upheld, at least in the past.

Things have gotten even more complicated since Mead. The most important case expanding Step Zero involved a crucial issue about the interpretation of the Affordable Care Act (also called Obamacare). Reading the language of one section of the law literally would have destroyed the main purpose of the statute, which was to make health insurance available to all Americans. The majority opinion by Chief Justice Roberts declined to apply Chevron and instead interpreted the statute without regard to the agency’s interpretation. The dissenters (including all of the textualists) did not even cite Chevron.

In another, less famous case, conservatives argued strongly for an expansion of Step Zero to include some cases of statutory conflict. In Scialabba v. Cuellar de Osorio, two provisions of an immigration statute seemed to conflict. In separate opinions, Chief Justice Roberts, Justice Scalia, and Justice Alito contended that Chevron deference did not apply when a statute’s provisions are directly contradictory.

Application of Step One (is the statute ambiguous?) is complicated by the fact that the Justices don’t entirely agree on how to interpret statutes, and thus they also disagree about how to decide whether a statute has a clearcut meaning. Some follow a theory called textualism, which focuses on interpreting the language of the statute; others follow purposivism, which focuses on the statute’s purpose and legislative history.

Chevron Step Two (was the agency reasonable?) has also evolved. In Michigan v. EPA, the Court rejected EPA’s interpretation at Chevron Step Two. Brushing aside the agency’s arguments that its interpretation fit with the statute as a whole, the Court considered the interpretation unreasonable because it did not seem to be sensible policy. In an opinion by Justice Scalia, the Court said, “[o]ne would not say that it is even rational, never mind ‘appropriate,’ to impose billions of dollars in economic costs in return for a few dollars in health or environmental benefits.” Justice Scalia continued that “[t]here are undoubtedly settings in which the phrase ‘appropriate and necessary’ does not encompass cost.” “But,” he said, “this is not one of them.” Although the term “appropriate” is very broad and would seem to give great discretion to EPA to decide what is appropriate, there was no indication of any deference to EPA’s view of the statute. Michigan could be a signal that the amount of deference at Chevron Step Two is being reduced.

In these three recent cases, some or all Justices have refused to apply Chevron at all (King v. Burwell), found that a difficulty of interpretation was not the sort of ambiguity that would trigger Chevron deference  (Scialabba), or given surprisingly little deference to the agency in Step Two (Michigan v. EPA). There is enough variation in the application of Chevron to make it unclear how much significance to attach to these recent cases. They may represent a blip rather than a real trend toward weakening or complicating the Chevron doctrine.  At any event, we now have something like the following scheme:

Step Minus-One.  Is this a “major question” in that it goes to the core of the statute or would radically change the state-federal balance (or perhaps, does it involve conflicting statutory provisions)?

Step Zero.  Does the agency have rule-making authority?

Step One. Is the statute ambiguous?

Step Two. Is the agency’s interpretation reasonable, both in terms of the statutory scheme and the court’s sense of plausible public policy?

As a lower court judge, Justice Gorsuch advocated overruling Chevron, and Justice Thomas has indicated that he also has doubts about the doctrine. Gorsuch was not alone among lower court judges, and there has been something of a vogue among conservative scholars to challenge Chevron. But all that was during the Obama Administration, and it remains to be seen whether these judges will feel the same qualms about deferring to deregulatory efforts by a Republican President. Correspondingly, liberals may also find themselves feeling less enthusiastic about Chevron now that the issue is deference to the Trump Administration. They’re unlikely to call for overruling Chevron but they may be more receptive to using the exceptions to Chevron deference.

There’s another possible explanation for why the Chevron doctrine is getting more complicated.  The question of how courts and agencies should relate to each other may simply be too complex and contextual to be reduced to a simple formula.  If that theory is right, the Chevron doctrine is likely to get still more complicated.

Dan Farber is the Sho Sato Professor of Law at the University of California, Berkeley, and Co-Director of the Center for Law, Energy, and the Environment. Follow him on Twitter here. This post was originally published at LegalPlanet.