It’s Time to Combat Abuse of Agency Guidance – Again.

We often think about judicial review when considering the administrative state. Thus, for example, when the U.S. Supreme Court overruled Chevron deference, it seemed like the entire legal commentariat had something to say.

But much of what the federal government does involves no judges at all. As Professor Christopher Walker has explained, because “[s]o much of administrative law happens without courts”—indeed, “agencies regulate us in many meaningful, and sometimes frightening, ways that either evade judicial review entirely or are at least substantially insulated from such review”—we should not “fixate on courts as the core safeguard in the modern administrative state.” Americans cannot rely wholly on the judiciary because too much happens outside the courts’ review. 

For example, agencies sometimes effectively regulate by guidance—for example, policy statements or “Dear Colleague” letters. Yet agency guidance generally is not challengeable in court, even when it exerts a coercive effect on regulated parties who do not want bad press or to be crosswise with their regulators. The result is that agencies sometimes de facto can regulate outside the authority granted them by Congress or contrary to the requirements of the Administrative Procedure Act. The Administrative Conference of the United States—an agency once headed by Antonin Scalia (and, in full disclosure, I am a Senior Fellow)—thus has issued recommendations to help agencies reduce inappropriate use of guidance and related tools. But those recommendations are just that—recommendations. The Administrative Conference has no authority to tell anyone what to do.

In October 2019, President Trump issued an Executive Order 13891 entitled “Promoting the Rule of Law Through Improved Agency Guidance Documents.” This important EO explained that “sometimes” agencies use authority to issue guidance “inappropriately in attempts to regulate the public without following the rulemaking procedures of the APA.” The EO also observed that “[e]ven when accompanied by a disclaimer that it is non-binding, a guidance document issued by an agency may carry the implicit threat of enforcement action if the regulated public does not comply.”  The EO then directed the Office of Management and Budget to order agencies to make their guidance readily available, to create better processes for issuing guidance (including soliciting public comments and responding to public feedback on certain guidance documents), and to prohibit agencies from suggesting that guidance binds the public. 

An order from the President is orders of magnitude better than a recommendation from the Administrative Conference. In response to EO 13891, agencies began creating systems to comply with Trump’s instructions.

That is, until January 20, 2021—the day President Biden took office. Biden rescinded EO 13891 almost immediately through his own EO. The blanket explanation given was that EO 13891 and other Trump EOs about regulatory reform were “harmful policies and directives that threaten to frustrate the Federal Government’s ability to confront” issues like “the coronavirus disease 2019 (COVID-19) pandemic, economic recovery, racial justice, and climate change.” In other words, the Biden Administration apparently hoped to use guidance to influence behavior on hot-button issues, but without as much litigation risk. 

Agencies responded to Biden’s EO as one would expect: they obeyed. 

I had hoped that when Trump returned to office, he would promptly reinstate EO 13891. As EO 13891 put it, “Americans deserve an open and fair regulatory process that imposes new obligations on the public only when consistent with applicable law and after an agency follows appropriate procedures.”     

But the Trump Administration has not reinstated EO 13891—at least not yet. Instead, it has addressed guidance more indirectly. For example, President Trump issued an executive order requiring bank regulators to “remove the use of reputation risk … that could result in politicized or unlawful debanking …  from their guidance documents, manuals,” and related materials. He also ordered agencies to “rescind … agency actions that impose an undue burden on timber production,” specifically including “guidance documents.” And in his bold  “Unleashing Prosperity Through Deregulation” Executive Order, Trump lamented that an “average person” must understand not just “formal regulations,” but also “rules, memoranda, administrative orders, guidance documents, policy statements, and interagency agreements that are not subject to the Administrative Procedure Act.”

The Administration, however, has not reinstated EO 13891 itself.

To state the obvious, the White House has been busy on other matters, so it is possible that it simply has not gotten around to EO 13891; it took the White House three years to issue EO 13891 in the first Trump presidency. Or, given what the Biden Administration did in response to EO 13891, the White House may have opted against unilateral disarmament, or concluded that again ordering agencies to put in the time and energy to create systems to prevent abuse of guidance is not the best use of resources—especially because a future White House may again rescind EO 13891. By its nature, EO 13891 is the type of “good government” initiative that is hard to push forward even under the best of circumstances (there always is something else competing for attention), but is especially hard if the initiative isn’t going to have staying power because everything can be undone with the stroke of a pen.  

Biden’s dismissal of Trump’s effort to make the administrative state fairer is particularly frustrating because preventing abuse of guidance should be a bipartisan goal. And even if the Biden Administration believed that parts of EO 13891 went too far (I disagree), the answer should have been to amend EO 13891, not rescind it.

For example, as Professor Nick Parrillo observed, EO 13891 ordered the “executive … to adopt procedural regulations for issuing ‘significant’ guidance documents, which regulations must require the agency to take public comment on any such document before finally adopting it—and also to give ‘a public response’ to ‘major concerns raised in comments.’” Among other things, the EO defined “significant” as guidance that would “lead to an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities,” “create a serious inconsistency or otherwise interfere with an action taken or planned by another agency,” or “raise novel legal or policy issues.”

That definition seems very reasonable to me. But it is true that soliciting and responding to comments requires agency resources. Reasonable minds may disagree about how many agency resources should be used to combat guidance abuse. But if the Biden Administration believed that EO 13891 struck the wrong balance, it could have defined “significant” differently. For example, if $100 million is not enough of an impact, why not set the threshold at $200 million? Or $500 million? Or, if the Biden White House did not like the language about “a material way,” it could have deleted that language while keeping the monetary thresholds. And so on. But instead of tailoring, the Biden Administration threw it all out. 

Similar analysis applies to EO 13891’s instruction that agencies should not issue “significant guidance” without approval “by the agency head or by an agency component head appointed by the President.”  To me, it seems almost self-evident that if guidance is expected to have, say, a $100 million effect on the economy, then a presidential appointee should approve it. The idea that decisions with such enormous consequences could be made by lower officials seems out of step with representative government. But again, even if EO 13891 set the wrong threshold (from the Biden Administration’s perspective), surely some threshold is appropriate. The answer should have been modification, not deletion.

And putting all of that aside, does anyone think agencies should not offer “searchable, indexed database[s] that contain[] or link[] to all guidance documents in effect”? Even if one believes that EO 13891’s other reforms are more trouble than they are worth, surely that reform at least is a good one. 

Given what happened last time, the Trump Administration may decide that it is not worth the trouble to return to EO 13891 in any form. But if nothing else, the answer should be amendment—not deletion. For example, the White House may conclude that guidance is more appropriate for certain types of regulatory schemes and that EO 13891 cut too broadly. Or the White House could conclude that the only part of EO 13891 worth preserving is the requirement to make guidance easily accessible.

The precise contours of an EO are worth thinking about, but the principles of EO 13891 are worth fighting for. Justice Alito has warned that agencies may “exploit[]” the “uncertain boundary” between what requires rulemaking and what doesn’t. Courts, alas, are often not well-positioned to police such issues. That is why regulatory reform within the Executive Branch is important. Hopefully, the White House will give EO 13891—at least in some form—a second look. 

Aaron L. Nielson is a senior fellow at the Civitas Institute and holds the Charles I. Francis Professorship in Law at the University of Texas at Austin School of Law.