It should come as no surprise that conservatives like Peter J. Wallison, a friend and law school classmate, and now a senior fellow at the American Enterprise Institute, should want to rein in the administrative state. Nor is it remarkable that opponents of regulation, like Wallison, have focused their attention on overturning the Chevron doctrine, under which courts give deference to administrative agencies when they interpret statutes that they are assigned to administer.

What makes Wallison’s new book Judicial Fortitude different is that it takes direct aim at the unwillingness of the Supreme Court to overturn statutes enacted by Congress on the ground that they delegate too much power to federal agencies (or the President) and have abdicated the duty of Congress to make the hard policy choices that they were elected to do. Although Wallison and I often do not agree on the proper level of federal regulations or even when they are needed at all, we do agree that the courts have allowed the doctrine of undue delegation to lie fallow for far too long and that it is time for the Court to step in and say enough is enough. Full disclosure: I am currently lead counsel for the plaintiffs making a delegation challenge to the statute under which President Trump imposed a 25% tariff on the imports of steel and aluminum products. (Plaintiffs’ moving papers can be read here.)

Chapter 6 argues that excess delegations are a perversion of separation of powers by enabling the executive branch to both write the laws and implement them, but Wallison is less than clear as to which delegations he thinks go too far. He argues that “important” decisions should be made by Congress, not the agencies (142, 118-30), but he does not give concrete examples that are helpful in drawing the lines between proper and improper delegations. Should important be determined (solely?) by the dollar differences in costs and benefits between, for example, an amended fuel economy standard of 40 miles per gallon from 35 mpg, as compared to an increase to only 37 mpg? Must the importance be apparent at the time that Congress is debating the law, or can it arise as the agency explores the options? The reader does know that Wallison thinks that the delegation unanimously upheld in Whitman v. American Trucking Ass’n was excessive (127), but not why or which other delegations would fall on that same basis.

In our delegation case involving the steel tariffs, we focus on the failure of Congress to make any of the hard policy choices among legitimate competing interests. Instead, Congress handed the President a blank check allowing him to decide, among other policy questions: which products will be covered, what remedies he may choose, what level of tariff (or quota) is needed, and whether he need to take into account the many segments of our economy that will be seriously adversely affected by the imposition of these tariffs. Moreover, because judicial review of the President’s discretionary choices is precluded, that check on executive power is also lacking, although if it were theoretically available, there would be nothing for a court to review because Congress has included no limits on the President’s choices of remedies.

There is a tension for those like myself who generally support a robust federal role in solving our nation’s problems when we seek to revive the nondelegation doctrine. Even if we had a Congress that was not close to totally dysfunctional, so that it was capable of enacting legislation other that by pure power plays by those in the majority, it lacks the time, skills, and foresight to replace administrative agencies. If one pushed nondelegation too far, it could result in a substantial reduction in vital regulations. Wallison seems willing to allow Congress to delegate purely ministerial duties, but there is no way that even properly debated and drafted statutes can eliminate all policy choices, whether limited to important ones or not. Supporters of fulsome federal regulation are willing to accept the fact that sometimes agencies will make policy choices with which they disagree, as the price for having them implement programs that they support, whereas Wallison does not. At one point (164), he would seem to require agencies to return to Congress to fill in the blanks when they come to an uncharted fork in the road—such as in Chevron when the Environmental Protection Agency had to decide whether a large power plant was a single or multiple “stationary source” for purposes of its regulations. That approach is a recipe for agency paralysis, which is perhaps what Wallison really wants.

The strongest evidence that this may be Wallison’s goal is Chapter 5, “Was the Progressive Faith in Economic Regulation Justified?” The longest in the book, it is not limited to what most students of administrative agencies consider “economic regulation.” As Wallison recognizes at the end of the chapter, Congress has repealed laws that set rates, routes, and other terms of operation for airplanes, railroads and trucks. Rather, that chapter, and the book overall, discuss the whole gamut of federal regulations, which were enacted to deal with many problems other than the price at which a product or service is sold.

Wallison has a particular distaste for financial regulation (86-88), a field in which he has spent much of his career. To be sure, laws regulating the amount of capital that a bank must have surely have an “economic” effect on its profits. However, those laws were enacted to protect the public against runs on banks, not to control the ability of banks to compete with one another. In that sense, they resemble laws governing pollution emissions or the safety of products sold to the public, which also increase prices because Congress has concluded that the tradeoff between higher prices and reduced harms favors the latter at the expense of the former, as our democracy entitles it to do. There can be legitimate debates over whether Congress or an agency has struck the proper balance, but that is a different question from whether there should be no regulation at all or whether the courts should step in more often and stop agencies from carrying out their statutory missions.

Despite the misgivings spelled out in chapter 5 and elsewhere, Wallison does not argue that courts should invalidate these regulatory programs under a 21^st century version of substantive due process. He is emphatic that he favors judicial restraint and would not want himself to be called a “judicial activist” (24-25, 82-83), a term that I have argued is a badge of honor or one of opprobrium, depending on the circumstances. Instead, he would rein in agencies by having the courts follow the words of the statute and avoid surmising legislative intent in order to decide whether an agency’s action was authorized by Congress (145). The difficulty comes in following that advice, as illustrated by a decision that Wallison applauds.

During the Clinton presidency, the Food and Drug Administration proposed to regulate tobacco products under its existing authority to regulate drugs (nicotine) and medical devices (a means of delivering the drug nicotine), a decision I supported at the FDA and in the courts. The Supreme Court agreed, as did the industry (reluctantly) that the words of the statute permitted the FDA to regulate tobacco products, but the Court nonetheless set aside the rule. It did so because it concluded that the FDA had exceeded the intent of Congress, not expressed in the applicable statute, but based on what Congress had done and not done in a number of other laws.

If the Court had been limited to the words Congress had enacted, the rule would have been upheld, contrary to the urgings of those, including Wallison, who would preclude inquiries into congressional intent, although he agrees with the Court’s decision against the FDA (18). On the merits, I consider the question to be quite close, but not because I reject efforts to understand what Congress intended by the law, but because I found the evidence of congressional intent to be less persuasive than did the Court. Nonetheless, if, as I tell my students when we debate this decision, you agree that the FDA was trying to fit a square peg in a round hole, as the majority concluded, then a fixation on literalism should not prevent you from agreeing with the industry and the Court majority.

Wallison’s final chapter supports the efforts to eliminate the Chevron deference given by courts to decisions of administrative agencies on questions of law. His voice on this is not new, but it is worth noting, although not surprising, that adherents of the anti-Chevron position, who now have the votes in Congress to affect the change, have not done so, almost certainly because they want the Chevron advantage when decisions of the Trump administration are in the courts. It is somewhat ironic—or perhaps proof of the neutrality of at least Wallison’s version of the end Chevron position—that the Supreme Court decision in Chevron overturned a D.C. Circuit ruling that held that the agency had under-regulated, not over-regulated, based on a Chevron-free analysis.

Although I would vote against a wholesale revocation of Chevron, I agree with Wallison that the courts have sometimes gone too far in deferring to agencies, including in the dissent of Chief Justice Roberts, in City of Arlington v. FCC, which Wallison embraces. My solution, however, is to “Mend It Not End It,” which I believe the Court can do on its own, in a more nuanced way than is likely to happen in Congress. Indeed, the Court has already taken some significant steps away from Chevron in three cases that Wallison does not mention: United States v. Mead Corp., Gonzalez v. Oregon, and King v. Burwell. My sense is that the Court will continue to chip away at Chevron, in particular with respect to agency deference to interpretations of its own regulations under Auer, but that neither it nor Congress will go all the way and end it. That may not satisfy Wallison and other strong Chevron critics, but it will bring a better balance between agency deference and judicial fortitude.

There is one other area on which Wallison and I are in full agreement: Agencies improperly use informal guidance, often in the form of letters rather than rules, both to regulate conduct not included in a statute or a rule issued after full notice and comment, or to excuse activities otherwise within an existing prohibition (4-9). The Administrative Conference of the United States, on which I serve as a Senior Fellow, has a thoughtful set of recommendations that would, if followed, go a long way to satisfying both of our objections to the overuse of guidance.

Wallison and I come to the subject of federal regulation from differing starting points, and I surely do not agree that “most Americans want less regulation and more freedom” (165). However, he does not seek to end all regulation, and I do not seek to defend everything that federal agencies have done, even when I agree with the end result. That ought to mean that there should be middle grounds on which we and others can agree, which will improve the process in Congress and the agencies, even if the result still falls short of perfection.

Alan B. Morrison is the Lerner Family Associate Dean for Public Interest & Public Service Law at George Washington University Law School.