Agencies and political control: taming an administrative wild west

Established in a long line of Supreme Court cases, Congress doesn’t maintain any sort of direct control or review over agency action. It’s limited to indirect action, such as antagonistic threats of budget cuts, “riders” that limit agency discretion in how it spends its appropriations, removing substantive statutory authority from the agency, getting rid of the agency altogether, and making life painful for agency officials by making them spend their time justifying themselves through testimony at hearings.

The President has direct constitutional control through his appointment and removal powers under the Appointments and Take Care clauses. U.S. Const. Art. II, § 2, Cl. 2 and § 3, Cl. 1. But agency control hasn’t really been that easy. Truman complained about how hard it was to get things done, and Nixon and Carter unsuccessfully sought to exert some centralized control. Reagan issued two executive orders to create an annual centralized review process of expected significant regulation, which was criticized as biased against regulation and done behind closed doors where industry cronies surely sought to influence regulation at the cost of environmental, health and safety protections for the common man.

Clinton, favoring this centralized scheme but responding to the latter charge, rescinded the previous executive orders and initiated Executive Order 12866, which in effect did not really change the Reagan practice, except that it added some transparency through publication in the federal register and actually made centralized review more powerful over agency action.

The order requires agencies to annually submit to the Office of Information and Regulatory Affairs (a subdivision of the Office of Management and Budget) all rules it expects to implement that it deems “significant” for review. Additionally, OIRA could make its own determination as to what’s significant. The agency would then be required to submit a text of the proposed regulation, as well as a report that outlines its justification, cost/benefit analysis of alternatives, etc. OIRA would then forward that information to all agencies heads, which could inform OIRA in writing of any conflict the proposed regulation might have with that agency’s policies, or actions taken or planned.

Additionally, agencies must periodically review existing regulations and modify or eliminate them to make them more effective or better aligned with the president’s regulatory vision.

OIRA must then waive review, or provide the results of review in accordance with certain time constraints (10, 45 or 90 days depending on the stage of the proposed regulation), and provide a full explanation on any parts of the regulation that are returned for further consideration (rejected by OIRA). The agency head, if he disagrees, can reply in writing outlining why he disagrees. In the case of such a dispute, the president decides which view prevails.

This is one of three direct checks and balances on agency rule promulgation. The other two are the Administrative Procedure Act, which requires agencies to provide notice and accept comments from impacted parties (and in such a case, if they don’t heed the comments, explain why they are sticking to their guns), and judicial review, which is unnecessarily deferential to the agency.

Clinton’s order has remained in tact through Bush 43 and thus far under Obama.

The order is a good thing — it’s not optimal, and very few regulations actually go through review despite how pervasive the administrative state remains. But it’s better than nothing. Some scholars give two main reasons as support for the order.

First, it fosters effective government because the president has a uniquely holistic and national perspective that facilitates coordination, coherence and rational priority setting, while agency heads tend to have tunnel vision. Second, it creates democratic accountability. We elect presidents, not bureaucrats and administrators. And shifting agency control away from Congress and toward the president is a good thing; there’s a lot of evidence committee oversight is dominated by special interests and preference outliers (committee membership is self-selected, such as Grassley on the agricultural committee influencing the USDA’s rules to benefit Iowa farmers). Further, the president is usually held politically accountable for agency action regardless of his level of control, so he ought to have the control anyway. See Elana Kagan, Presidential Administration, 114 Harv. L. Rev. 2245 (2001); Christopher C. DeMuth & Douglas H. Ginsburg, White House Review of Agency Rulemaking, 99 Harv. L. Rev. 1075 (1982); Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 Colum. L. Rev. 1 (1994).

Of course, statists criticize presidential control because it potentially means fewer and more modest regulations. First, they argue presidential review is essentially a bunch of non-experts second-guessing the policy decisions of experts. Second, both Clinton’s and Reagan’s orders have an anti-regulatory bias; important environmental, health and safety regulations are blocked by the cost/benefit analysis, and the OIRA review structure promotes the status quo through delays caused by review that incentivize agencies to play it safe by adopting more modest regulation. See Alan B Morrison, OMB Interference with Agency Rulemaking: The Wrong Way to Write a Regulation, 99 Harv. L. Rev. 1059 (1986).

The response is, of course, that an anti-regulatory bias is a good thing. Statists like agencies because they are unencumbered by the republican form of government. That form of government was chosen by men who recognized that lawmaking is an evil that should only be exercised by necessity, and thus, a substantial burden should be imposed on those who wish to change the status quo. In addition, I’d ask, why shouldn’t a cost/benefit analysis be imposed? Regulations affect the real world, and opportunity costs always exist. A regulation that seeks to increase safety of workers almost necessarily increases costs to the employer, which in turn spreads those costs to consumers. But regulators need not consider those costs in their little bubble; they can easily propagandize the results of their efforts without exposing the unintended circumstances. Just as important is the effect it has upon the distribution of costs and benefits (or risk and utility) — regulation has an anti-corporate, anti-wealth, anti-informed and anti-natural order bias. It redistributes costs not just to society overall, but often takes those costs off of society or the masses and puts them on a few. The costs of such redistribution can be detrimental to a host of liberties, and in view of a pervasive administrative state with many arms that view their missions without regard to what the other arms are doing, the collective force may demand too much of too few.

It’s important to note that centralized presidential control over agencies isn’t a great solution, and that there’s a strong case to be made against agencies as they exist today in general — not the least of which is extra-constitutionality. But presidential control is better than nothing. At best, a pro-active, anti-regulation president will hamper bureaucratic efforts to regulate all our liberties away. At worst, a pro-regulation president will simply do what he would have done without such an order. We still have two other — albeit ineffective — ways of combating the administrative state: judicial review and congressional antagonism. If nothing else, there’s a good case to be made for my presidential candidacy.