Win McNamee / Getty

In January 1981, I was a few years out of law school and a career government lawyer—that is, not a political appointee. Although I had been (and still am) a lifelong Democrat, fate and the Reagan administration had conspired to hand me a dream professional assignment: I had joined the legal team as a junior member providing advice on the development of the most significant institutional innovation in federal administrative rule making of the past half century—an innovation the Trump administration has since corrupted in important ways. Senators today have a rare chance to expose how and why.

But before we get there, let’s start at the beginning. In the late fall of 1980, the general counsel to the Office of Management and Budget (OMB) had recruited me from my Justice Department post to assist in the new administration’s drafting and review of executive orders. My job would not be to formulate policy. The task was instead to review proposed presidential orders to help ensure, along with my Justice Department counterparts, that the incoming president was staying within legal bounds in his new projects.

The most important of the initiatives I helped to process was an executive order with the modest title of “Federal Regulation.” Earlier presidents—Richard Nixon, Gerald Ford, and Jimmy Carter—had each implemented different systems to intensify White House oversight of the administrative rules issued by executive agencies such as the Environmental Protection Agency, the Department of Agriculture, the Occupational Safety and Health Administration, and so on.

Each presidential-oversight effort had been more ambitious than the last. For his part, President Carter had required agencies to create a publicly available “regulatory impact analysis” for each new regulatory proposal, which was to “contain a succinct statement of the problem; a description of the major alternative ways of dealing with the problem that were considered by the agency; an analysis of the economic consequences of each of these alternatives and a detailed explanation of the reasons for choosing one alternative over the others.”

Members of the Reagan transition, however, identified what they took to be two shortcomings of the Carter system. First, even though copies of the agencies’ analyses were to be shared with OMB, there was no centralized quality control. The rigor of any agency’s analysis was pretty much left up to the agency. Moreover, although agencies were required to analyze their proposals for rule making, they were not bound in their choice of regulatory strategy by the results of their analysis. Specifically, agencies were not required to use their analyses to maximize the ratio of regulatory benefits to regulatory costs.

The lawyers and economists on the Reagan team, working with James J. Tozzi, a veteran federal economist who had helped implement the Carter-administration reforms, came up with a draft order to remedy these two supposed defects. Henceforth, agencies would be told that to the extent permitted by law, they would have to hold off on the publication of any significant new rule until OMB was satisfied with the quality of the cost-benefit analysis, providing that centralized quality control the process had lacked. Moreover, to the extent an agency’s statutes permitted, each agency would be bound to pursue its regulatory objectives in the most cost-effective way.

The Reagan administration had one obvious legal problem with its potential new strategy: Congress had never authorized any centralized federal office to oversee the mass of federal regulation. Congress hands rule-making assignments to individual agencies, not to OMB. Hence, it would be the job of career lawyers like myself, along with our supervisors, to meld the new system as much as possible with laws that Congress had already passed.

One part of doing so turned out to be straightforward. Congress in 1980 had created within OMB a new division called the Office of Information and Regulatory Affairs, or OIRA. OIRA’s statutory assignment was to police the requirements that individual agencies might impose on the public for reporting information to the government. Making OIRA the headquarters for reviewing regulatory cost-benefit analyses would therefore take advantage of an office Congress had already approved.

A second part required more care. There are a variety of technical ways in which the Reagan order was polished to avoid any conflicts between the order and existing law, but one was of foundational importance: The career lawyers agreed that presidents had constitutional authority to demand regulatory analyses from agencies, because Article II of the Constitution explicitly authorizes presidents to obtain information from government departments. But, at the same time, we also argued that presidents had no constitutional authority to draft agency regulations for them. That is, we deemed it legally crucial that each agency’s final decision-making authority would remain formally intact. The Justice Department’s memorandum approving the new requirement of cost-effectiveness was conditioned on precisely this understanding:

The requirement would not exceed the President’s powers of “supervision.” It leaves a considerable amount of decisionmaking discretion to the agency. Under the proposed order, the agency head, and not the President, would be required to calculate potential costs and benefits and to determine whether the benefits justify the costs. The agency would thus retain considerable latitude in determining whether regulatory action is justified and what form such action should take. The limited requirements of the proposed order should not be regarded as inconsistent with a legislative decision to place the basic authority to implement a statute in a particular agency.

The new supervisory role carved out for OIRA attracted a variety of criticisms in the ensuing decades. Under pressure from Congress, the Clinton administration updated the original Reagan charter to make it rhetorically more balanced in its philosophy of regulation and more transparent in its operation. The OIRA-enabling order now explicitly acknowledges: “Nothing in this order shall be construed as displacing the agencies’ authority or responsibilities, as authorized by law.” But criticized or not, it turns out that both Republican and Democratic presidents from Ronald Reagan through Barack Obama have administered the OIRA program within the same general framework that Reagan’s people established. All presidents, regardless of party, value oversight of the regulations issued by the executive branch on their watch. OIRA cost-benefit review truly is the most important innovation in federal administration in the past half century, however little the public may notice it.

Unfortunately, the Trump administration has overlaid the Reagan-Clinton system with two additional sets of requirements that are unlawful and unwise, and which operate in even greater obscurity. They appear in an early Trump executive order, now being challenged in court. That order imposed two new mandates. One is called regulatory budgeting. An agency’s regulatory budget is actually an arbitrary ceiling that Trump’s OIRA places on the total economic cost that the agency is allowed to impose in a single year through all of its new regulations. The premise of the program is that compliance costs are a burden on the economy and, therefore, aggregate cost ceilings will improve the economy. For the current fiscal year, OIRA is using the cost-ceiling mandate to require agencies to actually produce negative regulatory cost. That is, the costs of new regulations must be more than offset by savings from regulations canceled.

The second is known in D.C. as “cut-go.” It is a requirement that, before issuing any new regulation, an executive agency must identify for OIRA two regulations appropriate for revocation. Both cut-go and regulatory budgeting have been proposed in GOP-sponsored “regulatory reform” bills that Congress has never enacted.

Regulatory budgeting and cut-go may sound appealing, except for three profound problems. First, cost ceilings take no account of regulatory benefits, which, of course, are supposed to be half the focus of the agency’s regulatory analysis under the Reagan-Clinton system. Second, it makes no sense for an agency intent on ramping up protection for the country’s health and safety to have to sacrifice existing protection for health and safety as part of its strategy. And finally, Trump’s order has no legal foundation in the Constitution or any federal statute.

The logic of the Reagan-Clinton system rests on the discretion that Congress gives administrative agencies to design their regulations. Under the Constitution’s separation of powers system, agencies are allowed to issue binding rules to protect the public health, safety, and welfare only when Congress empowers them to do so. Congress rarely, however, spells out the details of the regulatory activity it authorizes. Instead, the relevant statutes set forth Congress’s purposes and specify or at least imply those analytic factors that the agency is supposed to take into account when designing its rules to fulfill its assigned mission. As a result, these statutes invariably give agencies room to exercise their own best judgment in deciding how to carry out their regulatory assignments. Presidents from Reagan through Obama have wanted agencies, where allowed by law, to implement their policy-making discretion in a cost-sensitive way. But they have had to respect the legal principle that the authority to issue regulations rests where Congress almost always places it—in a specific agency.

The Trump order does not respect this essential principle. It instead requires agencies to violate the ordinary rules of administrative law. The Administrative Procedure Act effectively bars agency regulation that is “arbitrary, capricious, [or] an abuse of discretion.” As interpreted by the Supreme Court, this standard requires that agencies engaging in administrative policy making base their policies on factors that are legally and factually relevant. But there is no statute under which an agency can claim it is relevant to the merits of a new regulation whether the agency is meeting an arbitrary cost ceiling for its regulatory program as a whole or whether the cost of a new regulation is offset by the revocation of two older rules. OIRA thus appears to be forcing agencies to make decisions based on legally irrelevant considerations.

The regulatory budget system also rests on faulty economic logic. Presumably, by keeping regulatory costs low, the plan is to free up private resources to pursue other investment opportunities. Yet as Richard J. Pierce, one of the nation’s foremost administrative-law experts, has noted, OIRA has historically reported “total benefits [from regulation] that are approximately seven times as great as total costs.” Assuming agency rules have been rigorously vetted before being issued, there is no likelihood that the resources devoted to compliance would yield a better return on investment when devoted to other ends. Blanket deregulation will not help the economy.

I wish I could report to you how OIRA determines the appropriate aggregate cost ceiling for any agency’s annual regulatory output. Its website does not publicize its methods or reasoning. I wish there were a clear record to help evaluate the actual impacts of the Trump order on agency policy making. Digging deep enough, one can find an OIRA report claiming that, for the past fiscal year, “agencies issued 176 deregulatory actions and 14 significant regulatory actions.” It claims that “57 deregulatory actions were significant,” and that “comparing significant deregulatory to significant regulatory actions yields a ratio of 4 to 1.” But the definitions of significant—borrowed from the Clinton order—are so flexible that there is no easy way of assessing what this ratio means in terms of overall public welfare. It is also impossible to discern how many of the reported regulatory decisions, if any, were driven by any agency’s regulatory budget or by the cut-go requirement—and what drove the agency’s reasoning.

I also wish I could report to you how OIRA meshes its cut-go and regulatory-budget missions with the Reagan-Clinton program it still administers. It would be good to know, for example, whether agencies are being required to forgo regulation that would satisfy rigorous cost-benefit analysis.

Most of all, I wish I could point you to a Justice Department legal opinion providing some theory as to the legality of regulatory budgeting and cut-go. Perhaps the pending litigation will ultimately unearth one, but none is now available.

In the meantime, a Senate committee has an opening this week to shed light on what is going on at OIRA under President Trump. Trump has nominated Paul J. Ray, OIRA’s acting administrator and a former law clerk to Justice Samuel Alito, to be OIRA’s next permanent head. A committee hearing today on Ray’s nomination—and the follow-up opportunity to pose written questions for the hearing record—will give senators leverage to demand specific answers to the hard questions about the legality and wisdom of the OIRA program.

Ray would succeed Trump’s first OIRA head, Neomi Rao, a former law professor who is now a federal appellate judge. In confirming Rao to the D.C. circuit, the Senate Judiciary Committee should have probed her direction of the regulatory budget and cut-go programs. As the OIRA administrator, Rao proved a Trump bulldog, making claims for the success of Trump’s deregulatory efforts that were seriously misleading.

At today’s confirmation hearing, the Homeland Security and Governmental Affairs Committee should not repeat the Judiciary Committee’s mistake. It should find out why the administration thinks regulatory budgeting without statutory authority is lawful. It should probe how OIRA decides what the total cost of any agency’s annual regulatory agenda should be. It should demand meaningful public reporting of the trade-offs that OIRA is forcing agencies to make. It should ask why OIRA’s website reveals so little of the office’s actual reasoning with regard to any agency’s permissible regulatory cost.

Sadly, many Senate Republicans may be happy to overlook Trump’s overreach and OIRA’s excesses because of their own philosophical commitment to deregulation, full stop. If the Trump administration is putting into place, even without authority, the kinds of deregulatory requirements these senators tried and failed to enact, they may be happy enough with the result to let the administration’s lawlessness pass without comment. One has to hope that, at some point, members of Congress show some protectiveness for the exclusivity of their legislative role even when they share the president’s party affiliation.

Nearly 40 years after my brief stint at OMB, I remain proud of how carefully the career lawyers worked to assist in giving elected policy makers the best shot at embedding their initiative in a legally sound presidential order. This administration’s abandonment of that care now thrives mainly because of its obscurity. The Paul Ray confirmation is at least an opportunity to correct that.

This story is part of the project “The Battle for the Constitution,” in partnership with the National Constitution Center.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.