For “anti-administrativists” like myself, what’s the best way to reform the administrative state? In a provocative post at Law & Liberty, Professor Michael Greve rejects our ilk’s dominant approach, which is to focus on judicial deference doctrines—in particular, Chevron deference to an agency’s interpretations of its enabling statute. Chevron is a “great white whale,” Greve argues, because even if it were overturned, it would merely return in another guise. On this point, he borrows Adrian Vermeule’s thesis that courts inexorably will defer to agencies, due to the latters’ relative advantage in subject matter expertise.

In lieu of the mainstream approach to reform, Greve proposes to overhaul administrative law. His iconoclastic idea is to abolish the “appellate review model” system of administrative adjudication, by which he means the process where “agency adjudication comes first, followed by highly deferential, on-the-record judicial review.” According to Greve, “our hidden judiciary,” comprised of “12,000 plus administrative law judges and administrative judges,” is fundamentally broken because “the decisions of ALJ’s and AJ’s are virtually always subject to review and reversal by agency heads.” As a result, “the most one can expect from administrative adjudication is an appearance of impartiality.” Greve wants to raze the current system, and, in its stead, establish “administrative courts that are independent of administrative agencies—say, 100 courts with 1,000 or so judges, spread across the country.” As Greve explains it, “[c]ases brought in the administrative courts would not be appellate actions for ‘review.’” Rather, they would be original actions and the standard of review would be de novo.

Of course, his proposal is a non-starter. The politics will never be there. And, even if they were, it remains a very heavy lift to create a new system of justice from whole cloth. Greve knows this—he’s nobody’s fool. His actual aims are more modest. As aptly explained by R. St. Institute’s Philip Wallach, “[Greve] hopes, plausibly, that he can command that elite’s attention and make its members realize that an institution-building project is the best way to grapple with and discipline the administrative state at this juncture.” For this, Greve deserves praise. He’s trying to start a conversation about how best to reform the administrative state. It’s a conversation worth having.

But before such a conversation is possible, it is necessary to first come to some sort of understanding regarding background assumptions. To my eyes, Greve’s assumptions are significantly off, such that they threaten to lead the conversation astray. While I agree with Greve that the status quo for administrative adjudication suffers some serious flaws, these issues are idiosyncratic rather than systemic. That is, he overstates the threat, which leads him to overstate the solution.

Greve implicitly concedes much of this point. At a panel on his proposal last week put on by the C. Borden Gray Center, Greve said that he would exempt immigration judges and public benefits regimes (like Social Security Administration) from his plan, for practicality purposes. In response to these concessions, panelist Prof. Jeffrey Lubbers pointed to data collected by the Administrative Conference of the U.S., which demonstrates that that these two types of cases account for at least 95 percent of all administrative adjudications.

Moreover, most of the remaining ~5% of administrative adjudications tend to be mundane. Has the State Department’s Foreign Service Grievance Board ever incited controversy? Does anyone fret over the Railroad Retirement Board’s Bureau of Hearings and Appeals? Or the Black Lung Review Board? It is doubtful that these relatively innocuous tribunals are implicated by Greve’s concerns.

Of the administrative adjudications that are left, many are performed by commissions that do not follow the “appellate review model” script. Indeed, these commissions were created in response to the same concerns being addressed by Greve, albeit almost a half century ago. During the 1950s and 1960s, scholars made identical arguments as does Greve now—i.e., that the concentration of rulemaking, prosecutorial, and adjudicative functions in agencies engenders impermissible bias against regulated parties. Congress responded to these concerns by creating “separate-function” adjudications, where the tribunal is located in a different agency than the rulemaking and prosecution functions. Examples of this legislative response include the Occupational Safety & Health Review Commission (est. 1970) and the Federal Mine Safety and Health Review Commission (est. 1977). In fact, “administrative courts” were the very model for these commissions.

In other regulatory contexts, Congress has bucked the appellate review model by establishing independent tribunals within agencies. Decisions by these tribunals are insulated from further review by the Secretary or commission. Examples include the Departmental Appeals Board at the Department of Health & Human Services and the Administrative Review Board within the Department of Labor.

After you strip away the benefits/​immigration programs, the mundane regimes, and the tribunals that don’t follow the appellate review model, there isn’t much left. Basically, it’s the NLRB and a handful of agencies that employ administrative adjudications to impose significant civil monetary penalties (including primarily the SEC, CFTC, FERC, CFPB, OCC, and the DOT). With respect to this subset of agencies, Greve’s criticisms hit the bullseye. They’re all guilty of one or more of the abuses that he identifies, including “the imposition of civil fines by bureaucratic edict; sudden changes of agency policy, accomplished by means of adjudication and without fair warning to the parties; [and] the opportunistic shifting of enforcement proceedings from Article III courts to agency tribunals.”

After accounting for the proper scope of the problems, it is possible to calibrate a response from a menu of incrementalist options. Although Greve seems to set forth a binary choice of reforms—either repeal Chevron or overhaul the administrative state—there are many modest doctrinal and institutional measures that would mitigate his concerns.

First and easiest, Congress could simply amend enabling statutes to resolve the issues identified by Greve. It was only over the last three decades that Congress expanded use of the “appellate review model” for civil monetary penalties, at the goading of the Administrative Conference of the United States. Lawmakers could just as easily reverse course, and return to the model whereby agencies sought relief in an original action before an Article III court.

In a similar manner, Congress could amend enabling statutes to insulate the administrative tribunal from direct review by the agency or commission. As I explained above, this is known as the “separate-function” model, and it is how Congress responded to concerns identical to Greve’s almost a half century ago.

Turning from Congress to the judiciary, the Supreme Court could clarify several inchoate doctrines and thereby check the administrative state, in a manner perhaps exceeding the effect of overturning Chevron deference. Examples include the applicability of the Seventh Amendment to administrative adjudications, and also whether certain administrative adjudications impermissibly share “essential attributes” of the judicial Power.

I can think of other opportunities for doctrinal refinement that would mitigate the potential for abuses of power by the administrative state. For example, the Supreme Court has interpreted APA § 706 to call for courts to take a “hard look” at the reasonableness of agency decision making. To my eyes, it is plainly unreasonable for an agency like the NLRB to swing back and forth between policies, over and over again, due to alternating political parties occupying the White House. At some point, reasonableness requires a court to say: Enough! At which point it becomes Congress’s responsibility to amend the law.

Finally, there’s my preferred institutional reform, one that I think would make mitigate virtually all of the harms associated with the “appellate review model.” Here, I’m talking about building expertise in Article III courts. The Executive Branch has literally millions of employees. The Congress has about 20,000 staffers. The Supreme Court, by contrast, has about 40 clerks, most of whom are brilliant law school grads with no experience in administrative policymaking. This expertise gap is even greater in lower courts, and it goes a long way towards explaining the unfortunate deference dynamic identified by Greve:

[W]ithin the framework of the appellate review model, the case for judicial deference to the executive branch is overwhelming. When judges review policy decisions, they have no comparative advantage over agency administrators. The obvious question—why make the same decision twice?—naturally draws courts back towards deference.

If the reason that Chevron could never go away is the expertise imbalance between agencies and courts, then perhaps it’s time to think about making the courts more expert. Instead of 1,000 judges, maybe the better approach is 1,000 Article III staffers.

In sum, Greve’s grand ambitions may be misguided, but he has done a valuable service by drawing attention to what ills agency adjudications. Rather than tearing it down, a better alternative for reform is to advance the ball on a number of incremental measures.