The most official way for an agency to communicate with the public is through binding legislative rulemaking. But an agency can send out communications much faster, and at higher volume, if it communicates by guidance—that is, by general public statements, nonbinding, giving its current thinking on how it will interpret the law and use its discretion. Despite guidance’s ubiquity and usefulness, it has long been a subject of controversy because, insofar as regulated parties feel practical pressure to follow guidance, the agency has an opportunity to change the behavior of those under its jurisdiction while avoiding notice and comment and judicial review. I recently completed a study for the Administrative Conference of the United States on how federal agencies use guidance, based on interviews with 135 individuals across agencies, industries, and NGOs; here I draw upon my research from the study to discuss some of the peculiar challenges of agency communication.
Amid the controversy about guidance documents, one might get the impression that such statements are typically hatched by high agency officials bent on using them to coerce the public. But in real life, this is not the usual story, to say the least. A great deal of guidance is formulated in response to pressure from the regulated public and as a means to manage and improve forms of advice that are already being given at the lower levels of the agency in response to stakeholder demands. It’s quite natural for an attorney or other employee at a regulated business to ask an agency official, “If we do X, is that okay?” (Indeed, there are some common structures of regulation, like high-stakes licensing, in which regulated firms hang on the agency’s every word and try to find out as much from agency officials as they can.) In the words of one former EPA program office director whom I interviewed, the “instinct” of the people at the agency “is to answer the question.” But once agency officials start giving answers to individual questions, whether in person or by phone or letter or email, there are immediate problems. What if different officials give different answers to the same question? Or give answers that high-level managers think are wrong? Or give answers only to the regulated parties who are savvy enough to ask the right person? In other words, there are problems of accuracy, uniformity, and equality of access. If agency management doesn’t reach out to identify these problems, somebody else will likely raise them: letters and emails to individual firms can be FOIA’d, or “passed around” among industry players. Thus higher-level officials are naturally inclined to centralize the otherwise haphazard advice-giving process and make it more transparent. That means converting oral advice into written responses and making sure that, when a significant question arises repeatedly, high-level management decides on a uniform answer and makes it generally available to all stakeholders, savvy or not. Thus, casual, scattered, one-on-one conversations can evolve into centrally-vetted “guidance documents” on the agency website.
Once the agency decides to give uniform, transparent guidance on some question, it seems worth it for officials to think through the answer, look into relevant research, consult stakeholders (short of the burdens of notice-and-comment rulemaking), etc. The fact that the advice will be given to a large number of people justifies this effort. But herein lies another problem. Guidance, as a legal matter, is supposed to be nonbinding. That is, if a particular regulated party proposes to follow a course of conduct other than what the guidance advises, the agency is supposed to give fair consideration to that proposed alternative. If the agency doesn’t, the guidance is inflexible, meaning it’s arguably a de facto legislative rule that didn’t go through notice and comment, and is therefore unlawful. Yet in order to make a rational decision on whether to go along with the particular firm’s proposed alternative, the agency may need to go through a lot of the same deliberation, research, and consultation as for the original guidance. But if the agency staff is strapped for time and money, it may be hard to justify spending the requisite resources (which could go to other pressing matters) on formulating advice for just one company. Maybe the agency could just give the green light without much scrutinizing the company’s proposal or communicating an explanation for its approval, but then word of the casual exception gets around – one former CMS division director told me “there are no secrets” – and pretty soon the one company’s competitors are complaining about an unlevel playing field (or knocking down the agency’s door asking to get the same exception themselves), or NGOs or journalists are accusing the agency of making sweetheart deals. Given these risks, it’s no surprise that the agency may be silent, or stonewall, when asked to provide the individualized consideration that “nonbinding” guidance is supposed to entail.
An agency can mitigate this problem by taking advantage of economies of scale in deliberation and communication. As one chemical industry attorney told me, a company wishing to proceed differently from guidance will have an easier time getting a dialogue with the agency if it “bands together” with other firms wanting the same thing. Another way for the agency to exploit economies of scale is to make a broad solicitation for stakeholder input whenever it formulates guidance to begin with (to recognize alternatives for which there is widespread industry demand and bake them into the cake from the start) or periodically to consider wholesale revisions to the guidance, for which the agency can likewise make a broad solicitation for stakeholder input.
Such solicitations are often a good idea, and many agencies do them in some way or other. But soliciting input, too, presents a communication challenge. If the agency solicits stakeholder input on a guidance document, it usually needs to make available a proposed (draft) version of the guidance, so stakeholders know what to give input about. But the very publication of “draft” guidance can cause trouble. Regulated parties are likely to view the “draft” as reflecting the agency’s current thinking, and they may feel almost as much practical pressure to follow it as they would a “final” guidance document (given that guidance, even when “final,” is only supposed to be a nonbinding suggestion of the agency’s current thinking anyway). If regulated parties do indeed follow the “draft” guidance as if it were final, then officials may thereby obtain whatever shift in regulated-party behavior they think is desirable, with the result that they have little incentive to do the labor-intensive work of processing stakeholder input and deliberating on whether to incorporate that input so as to make the document final. The result is that the document remains indefinitely in “draft” form. This has occurred with some frequency at FDA and at least occasionally at EPA. Interestingly, it appears not to occur at the USDA’s National Organic Program, where officials have successfully signaled to regulated parties that they will not follow the substance of a draft guidance document in individual adjudicatory or enforcement proceedings unless and until the document becomes final. Sending that kind of signal is a subtle but important communication challenge for agencies seeking to make meaningful use of public input on guidance.
Nicholas R. Parrillo is a Professor of Law at Yale Law School.
This post is part of a symposium entitled How Agencies Communicate. You can read all the posts here.