The Necessity of Politics in Administration, by Cristina Rodriguez
*This is the eleventh post in a symposium on Peter Shane’s “Democracy’s Chief Executive: Interpreting the Constitution and Defining the Future of the Presidency.” For other posts in the series, click here.
A critical passage in Peter Shane’s wide-ranging and bracing new book comes late in its pages, in an analysis of the debate among scholars over whether the Founding generation contemplated significant delegation by Congress to executive officials. Shane writes: “The debate among these first-rate scholars, although fascinating, helps to illustrate what is bizarre about using originalism as the primary guide to constitutional interpretation. What is at stake is the capacity of our national government in the twenty-first century to address regulatory issues that are critical to public health and social welfare.” (Shane at 170). In other words, how do we keep a government that serves the people, given the state of our world?
I want to take him up on this challenge but distill the question in a way that captures a tension at the heart of the book: How much politics is too much politics in executive branch decisionmaking? Shane calls for a democratic conception of executive governance but effectively dismantles two of the originalist theories that claim to inject democracy into administration: he mentions in passing the non-delegation doctrine alluded to above, which purports to restrain the democratically elected Congress from giving its legislative powers to unelected bureaucrats, but then really devotes impressive intellectual energy to dismantling the idea of unitary (or totalizing) presidential control. He this seems to accept some but certainly not all forms of political influence over administration.
To better understand the role of democratic politics in executive decisionmaking, we need to untangle two concepts that are woven together throughout the book and often conflated with one another in contemporary commentary: how much should political or even partisan considerations shape the execution of the law versus how much control should the President himself (and the White House) have? In thinking this through, I want to offer a conception of political control and its necessity that I doubt Shane would reject altogether, but then link this conception of politics to an account of presidential control that I believe he would contest.
Shane’s concept of democratic presidentialism calls for an inclusive, deliberative, and accountable government. In a view I have defended at length elsewhere, this vision necessarily depends on the injection of political values into executive decision-making, including in the priorities set and the policies pursued. No one truly doubts that administration implicates not just technical judgment but also values choices—congressional enactments are not self-executing, nor is administration just a matter of gap filling, Justice Gorsuch’s conception of interstitial decisionmaking notwithstanding.
On this view, administration should be infused with a combination of popular preferences and values, as well as the principles and policy agendas advocated by political parties, civil society, and other interest groups. The administrative state, in some demonstrable way, should reflect responsiveness to these sorts of public values, and government must remain capable and motivated to adapt to the wants and needs of the public. As Shane himself recognizes, periodic elections and legislation by Congress are inadequate for these purposes. In one of his more important observations—that democracy depends on “the opportunity between elections for genuine policy dialogue both within and outside government (p. 163)—Shane captures the essentiality of politics to administration.
A critical means by which these public values inform administration is through the influence of political officials on the bureaucracy (which is not to say that bureaucrats don’t have values, but that they are less likely and legitimate sources of the popular values of the moment). It matters that political officials are connected to a democratically elected regime, not because the president is a national representative—an idea Shane and others effectively have debunked—but because the regime they represent is the product of an election and the rise to power of a set of values and preferences. On this view, the fact that a new political regime has come into office should often be sufficient to justify a new set of policies or a new regulatory agenda, as long as an administration stays within its statutory bounds (admittedly a large discretionary space).
In our work together, Anya Bernstein and I underscore that what is often portrayed as conflictual—the relationship between political appointees and career civil servants—is at least as often complementary. The value of the political official is epistemic—they introduce forms of thinking critical to effective and inclusive decisionmaking. In our qualitative study of agency decisionmaking, political officials are portrayed as pro-active, policy-driven, politically sensitive, risk-tolerant, and career officials were more reactive, expertise-driven, implementation sensitive, and risk averse. Civil servants might have longer time horizons and will be more invested in the success of the agency qua agency, but political officials offer a conception of how government ought to work tied to the views of a democratic constituency.
This particular vision of political control does not describe and ought not be understood as displacing other forms of public and popular participation. Shane helpfully cites Jerry Mashaw’s reminder that agency decisionmaking may be the preeminent site of democratic deliberation, particularly given the APA-imposed requirements for notice and comment rulemaking and reasoned elaboration. (p. 145) To this formal picture, Bernstein and I would add a range of informal modes of responsiveness characteristic of much of agency decisionmaking, including solicitations of public input throughout decisionmaking processes through external information gathering and informal meetings with civil society and other regulated parties. But the larger point is to resist any presumption that political influence over agency decisionmaking is ultra vires.
I’m not certain that Shane would disagree with this political conception of administration. I think our disagreement is more likely to be found in my second set of observations—about how presidentialism and presidential control fit into this picture.
The fact that Shane invokes authoritarianism to describe a possible, even likely, result of the power the presidency has accrued since the Reagan years underscores the dangers he perceives in presidential control. But presidential control, as a descriptive matter, is also now constitutive of the administrative state—a present-day reality brought home by Shane’s own narrative of the last forty-five years of legal development. More important, as a descriptive matter, it amounts to far less that both its proponents and opponents claim. As Bernstein and I have shown, the primary way presidential values or priorities shape governance is through the loose diffusion of values and abstract policy preferences through the political appointees scattered across the administrative state. White House involvement is often peripheral or facilitative and collaborative. This common picture belies the frequent use of the term “unilateralism” by Shane and many other commentators.
Such observations do not mean that OIRA never frustrates agency actors (it often does), that there aren’t better or worse ways to run an inter-agency review process, or that the president and his senior advisors won’t try to steer what would otherwise be an agency-driven policy process to promote the president’s policy agenda or to serve the political imperatives produced by the news cycle. But Shane himself recognizes that presidential control or centralization can have systemic benefits, such as introducing coherence across regulatory regimes and ensuring consideration of a broader public purview than often shapes agencies in their particular contexts. Presidential control, thus, is arguably better understood as a variable to contend over, not a threat to democratic government. At least not yet, or not necessarily.
Shane is certainly not wrong to warn that a version of presidentialism could threaten democratic, deliberative decision-making, and he rightly regards the Trump era as an inflection point. He cites many causes for alarm with an air of reality in our present moment, including the specter of a President exploiting prosecutorial discretion to thwart investigations into allies or himself or to launch probes into opponents; screening civil servants for political loyalty; and firing career officials en masse for failure to protect the president or pursue his objectives regardless of the limits of their statutory mandates. To the extent the Supreme Court’s appointments and removal jurisprudence and the political culture of Washington (and the Republican party) are now abetting these sorts of developments, they can be roundly criticized based on the very principles of democratic accountability often invoked to justify the unitary executive.
But it is hard to pinpoint what marks the shift from a system that produces these virtues to an authoritarian regime with contempt for the rule of law altogether. Perhaps the former always contains within it the possibility of the latter. Even if the rise of the authoritarian populism of our day has little to do with White House control over agency decisionmaking, we might still understand presidential control as a ready means of trampling on the rule of law and bureaucratic expertise both.
And still, it is far from obvious that centralization must necessarily be eschewed, if doing so would be a combination of misdiagnosis and over correction. In this vein, as Adam Cox and I have shown in our work on the immigration bureaucracy, centralized priority setting, including with White House involvement, can often be central to ensuring that policy is controlled by politically accountable officials who will much better see systemic trade-offs required in a system of mass adjudication, including by bringing political values to bear in making those trade-offs. These are epistemic horizons that individual or decentralized line agents or administrators are less likely to possess and certainly will not be authorized to deploy. We describe and defend DACA with this perspective on centralization. The strongest, generalized version of this point is to say that we cannot have a truly democratic executive branch without some version of presidentialism—to shape executive decisionmaking in a democratic spirit—to move the bureaucracy in effective ways—centralization may be required.
Like Shane, Cox and I have expressed a preference for decision-making driven by agencies. We think their political officials will better understand how to integrate political considerations with statutory limits and other bureaucratic values, such as expertise and long-term systemic integrity, than a White House that, as a relative matter, will be disconnected from regulatory context and much more likely to be driven by transitory judgments about the political gain to be had from one decision or another. But this preference for agency-driven action need not be mutually exclusive with a certain amount of steering from the White House.
Now even if all of what I have posited above about presidential and political control turns out to be reasonably accurate, Shane will still be right to explore how to maintain a culture of accountability—a pretty basic requirement for a government of the people and a necessity given the last administration’s escalation of the assault on the rule of law that he ominously describes. Unfortunately for all of us, the two likely sources of accountability seem increasingly unlikely to be up to the job. Shane recognizes the dangers of a supine—or worse, polarized—Congress, where oversight consists of either abdication or grandstanding.
The other option is the courts. Perhaps because Shane’s major focus is in critiquing the unitary executive in order to offer a more pluralistic alternative, his exploration of the role of the federal courts in all of this turns out to be somewhat narrow. I would welcome his insight into how effective our present-day courts might be, especially since much of the judiciary will be disposed like the Supreme Court to expand top-down control but thwart bureaucratic creativity based on a distorted conception of accountability that Shane helps mightily to expose.
In part because I believe that judicial review of agency action has itself become both political and partisan (with the definition of “arbitrary” tracking sincere differences in political perspective), I have expressed reservation about further enlisting the courts in constraining executive governance. I also would not characterize the courts as providing the same sort of democracy-enhancing check as an active Congress; courts may promote accountability but they might just as easily thwart democratic decisionmaking by stymying the outcome of a notice and comment rulemaking process, blocking the implementation of a congressionally created regime, or keeping agencies from addressing pressing public problems because of formal procedural hoops or arid legal interpretations of what should be dynamic statutory regimes. And so even though it seems safe to say that the courts’ application of the reasoned deliberation requirements of the APA has over time improved the deliberative quality and transparency of administration, proceduralism should cease being our default accountability-promoting reform.
Shane’s book comes at a challenging time in the political life of our nation—one in which it seems plausible that thing could “go either way,”—toward a continuation of the pluralistic vision of executive governance that Shane touts or toward the authoritarian distortion he fears. Efforts like his are necessary if we are to have a chance at following the former rather than the latter course. And his book’s value stems less from his riposte to the conservative originalist account of the presidency (as convincing as it is) than from his pluralistic vision and theory of inclusive government that provides either a reason for optimism or a way back from the abyss.
Cristina Rodriguez is the Leighton Homer Surbeck Professor of Law at Yale Law School.