The Constitution vests all legislative powers in Congress, yet Congress grants expansive lawmaking authority to federal agencies. The conventional model for understanding this relationship between Congress and the administrative state—as positive political theorists have long explored—is to view it in principal-agent terms. In other words, Congress delegates authority to federal agencies, and those agencies are expected to faithfully implement congressional wishes.

Despite the predominance of lawmaking by regulation and the decades-long application of principal-agent theory to the regulatory state, we know very little about whether they are faithful congressional agents. My empirical study inside agency statutory interpretation, which will be published in theStanford Law Review this summer and is available in draft form on SSRN here, explores agency rule drafters’ views of their relationship with Congress and attempts to evaluate interpreter fidelity based on a number of different metrics (textual and substantive canons, legislative history, administrative law doctrines). I frame the paper (and the survey administered at the agencies) in principal-agent terms, but in so doing the limitations of such model quickly become apparent.

To begin with, federal agencies often serve at least two principals—Congress and the President. Even putting aside the presidential principal, it’s not clear which Congress is the principal: the one that passed the legislation at issue, or the one that currently oversees the agency. Moreover, who is the principal within Congress? The Congress as a whole, whose collective intent is difficult to ascertain outside the statutory text? Or perhaps the congressional committee that oversees the particular agency and engages in an ongoing dialogue with the agency? These are just a few examples to illustrate how the principal-agent model in administrative law is too simplistic.

Aside from these theoretical concerns, there’s also the descriptive question about whether federal agencies really act as agents of Congress. A terrific new article by Brigham Daniels makes a compelling case that the relationship between Congress and federal agencies is often not one of principal-agent. (The article is titled “Agency as Principal,” and it was just published in the Georgia Law Review.) Indeed, as suggested by the article’s title, Professor Daniels exhaustively details that federal agencies sometimes assume the role of principal in their relationship with Congress (and other actors). Here’s a great summary of his argument from the introduction (some paragraph breaks inserted):

The thesis of my argument is simply this: sometimes agencies can turn the tables on Congress and the Executive. When this happens, it is wrong to think about agencies as agents, even stubborn ones. Rather, in such situations, we should think about agencies as principals manipulating the elected branches, which in turn take on the role of agents.

So, how do agencies become principals? Consider a few examples, some more surprising than others but all reinforcing this central theme. Think about the subtle, perhaps at times even passive-aggressive nudging that agencies use to refocus the elected branches and to frame issues before them. Examples of this sort of agenda-setting behavior can take the form of the problems agencies highlight in the reports they publish, the data they make public, and the responses they provide to inquiries from the elected branches. While many of these agency actions seem mundane, the political significance of such agency work is often lost on legal and political players, even as it is relied upon and cited by them to lend their arguments credence.

Next, consider the ways those in agencies attempt to lobby elected branches, ranging from endorsing pending proposals to drafting policies and legislation and then shopping these to those within the elected branches. Moreover, sometimes those in agencies criticize those in the elected branches publicly. While such a strategy comes with obvious political risks, such criticisms can have great power within the public debate. Sometimes, agency actions alter the political landscape facing the elected branches. Those in agencies might gain political leverage in many ways, including through agency decisions in rulemaking and orders, positions taken in litigation, stories leaked to the press or political players, and even the threat of resignation of those within the bureaucracy also within the public eye.

All these scenarios introduce us to a world where agencies can shed their typical role of agents to the elected branches and step into the role of political principals, if only for a time. Granted, we might find disagreement as to when those in agencies can rightly be said to be pulling the strings of the elected branches, instead of the other way around. To accept my thesis, however, all that is required is that we find agreement at some point along the way. This is enough to lead us to the conclusion that our conventional understanding of the relationship between agencies and the elected branches is lacking.

This article is a really fun read and an important contribution to the literature, as it exhaustively details examples of agencies as principals. To be sure, Professor Daniels draws primarily from environmental law and regulation, which is his primary area of expertise, but the observations and arguments extend beyond environmental law. Part II also has a great discussion of the various principal-agent relationships that could be implicated in the modern administrative state, including some very helpful figures that depict those relationships. Here’s one example:

I’m convinced that the principal-agent model in administrative law is too simplistic and often not true as a descriptive matter. But I’m not convinced that we should abandon it as a normative matter (not that Professor Daniels is necessarily suggesting that). Indeed, the examples set forth in this article should encourage the opposite—i.e., to consider ways to better constrain agency action via narrower congressional delegations and greater congressional oversight, more searching judicial review of agency action, and perhaps greater presidential control and review. If we don’t work hard to ensure that federal regulators act as faithful agents of Congress, the constitutional and democratic legitimacy of the modern administrative state is on even more perilous footing.