Regulatory Policy in the Reagan Administration

As Kip Viscusi stated in his fine paper, the primary goal of the Reagan administration in the area of regulation was to improve the efficiency of regulatory programs by hewing to economic thinking as much as possible. Thus, this area of economic policy provides a good occasion to address the questions with which Martin Feldstein opened this conference, namely, Where was economic thinking influential in policy-making. Where was it not, and, Why?

Let me begin by mentioning three reasons why regulatory policy represented a fruitful area in which to increase the role of economic reasoning in policy-making. First, the regulatory agencies have an enormous amount of discretion in interpreting the laws, despite a common belief that statutory standards are very strict (and often so uneconomic as to make almost any economist shrink in horror). In truth, the agencies are usually told in general terms to promote occupational safety, or pollution reduction, or whatever, and are then given great discretion in how they do so through particular regulations.

Second, as a sort of constitutional price for this discretion, regulatory agencies are required to be highly, and I think almost uniquely, rationalistic about what they do. They must give public notice about their intended policies and draw a coherent connection between those policies and their legislated objectives. Further, both the agencies’ decisions and the rationales that they offer for those decisions are subject to some degree of review by the courts. Clearly, rule making can be a contentious and politicized process, and the rationales that the agencies give may disguise narrower or unworthy goals, but nevertheless, there is an obligation to justify what is being done in regulatory policy that is greater than that for monetary or fiscal policy.

Third, a series of executive orders beginning in the Nixon administration has created a progressively more aggressive and formal review of regulatory policy by the Office of Management and Budget (OMB) and other White House offices. Although these orders began with rather vague and ad hoc standards—“quality of life” in the Nixon administration and “inflation impact” in the Ford administration—they have been expressed increasingly in economic or at least cost-benefit terms. That progression culminated in an executive signed by President Reagan shortly after he came into office that set forth a series of net social benefit standards that are reported below in section A of the appendix.

Many of the people supporting regulatory reform, including many colleagues and allies around this table, had academic backgrounds and were very sympathetic to the notion of using economics in regulatory policy. During the time that I administered President Reagan’s regulatory review program at OMB, roughly sixty regulations came in for review every week, and I think that everyone there tried to push the economic standard as thoroughly as possible in that review process. Indeed, I tried to some extent to codify our approach to regulation through a kind of common law of review of individual rules. By applying economics to a variety of types of regulatory action, I tried to give the agencies and the outside world a sense of how we interpreted and were applying the executive order. Eventually, a set of regulatory policy guidelines was established in another executive order; they are reported below in section B of the appendix. Someone recently described these guidelines as the most relentless application of microeconomic theory to regulatory policy ever attempted.

In the rest of these remarks, I want to discuss the extent to which we succeeded or failed in our attempt to push economic thinking. I think that the best framework for this discussion is to consider three types or classes of regulatory policy where the executive branch has broad legal discretion. In the first case, corresponding to guideline 2 in section B of the appendix, economics provides a clear prior indication of what the right policy is. The second situation is at the other extreme, where determining the correct policy is not a matter of theory but requires a lot of empirical work as well; this situation corresponds to guideline 4 in section B. The third situation is an intermediate one, corresponding to guideline 3 in section B. My conclusion will be that the attempt to apply economic thinking in the regulatory review process was most successful where the policy implications of economic analysis were most widely understood and accepted by policy officials who were not themselves economists—in other words, where economic thinking had already suffused the thinking of regulators who were not professional economists.

The first class of regulation that I want to consider is agricultural marketing orders, which might involved removing 25 percent of the almonds from the market for the coming year or denying certain people the permission to plant hops on their underutilized cropland. Many people at the white House had thought that the executive order on regulation would affect primarily the regulation of transportation, occupational health and safety, and the environment and were surprised to discover a spate of these egregious agricultural marketing orders coming in for review. In some ways, I wished that I could ignore these orders because, although they were offensive, I suspected that their economic impact was not nearly a great as the impact of the health, safety, and environmental regulations. At the same time, many people outside the administration believed that the executive order on regulation was simply a way to provide relief for industry rather than sound economic policy. So, in that situation, we simply could not ignore these marketing orders.

The analysts at the Agriculture Department were shocked when I told them that I did not even want to see their cost-benefit analyses demonstrating the marketing orders to be beneficial. I knew without looking at the facts that the rules could not be beneficial, that price controls and entry controls in thoroughly competitive markets were empirically uninteresting. I started rejecting these rules left and right, including one Michigan cherry order the rejection of which put an end to any prospects that Dan Stockman had for a future political career, for which he has been, I hope, deeply grateful to me.

Dealing with these marketing orders absorbed a lot of time and attention, and, as Bill Niskanen pointed out, eventually there was an appropriations rider that prevented OMB from reviewing the orders in the future. As I look back on this issue, however, I am surprised at how much change we were able to make, given the political opposition based on the highly concentrated nature of the benefits of these orders. We abolished the two marketing orders that contained outright entry controls; we greatly reduced the use of “set asides” in a variety of other orders; and for the two largest and most harmful marketing orders, affecting California citrus crops, we greatly truncated the growing season to which the orders applied and allowed free marketing at the beginning and the end. Further, the people in the Agricultural Marketing Service gradually assimilated an economic way of thinking about the effects of entry and price controls. Thus, although we had no allies in the beginning, by the end it did not even matter very much when Congress passed the appropriations rider. My understanding is that these agricultural marketing orders are continuing to come unraveled, and I think that eventually they will fade away.

The second class of regulation deals with environmental protection. This regulation applies to situations where there are no markets, where there are large externalities, and where regulation is clearly justified. But, in contrast with the agricultural marketing orders, there is nothing in economic theory to decide what the appropriate extent of that regulation should be. Let me focus on the example of effluent guidelines under the Clean Water Act, where the Environmental Protection Agency (EPA) sets industry-by-industry standards for the amounts of various sorts of pollutants that can be discharged. This is an area of regulation where the use of cost-benefit analysis itself was not particularly controversial—we were not arguing that economics says you cannot control water pollution, and almost everybody at EPA acknowledged that cost-benefit analysis was an appropriate way to gauge pollution control measure. Significant controversies arose in the application of cost-benefit analysis, however.

To start, it was difficult to measure the benefits of reducing water pollution. We usually had a pretty good estimate of the costs of reducing pollution, but many of the benefits were recreational or aesthetic, which are very hard to measure or assess in any quantitative way. We noticed, however, that the effluent guidelines produced enormous discrepancies across industries in the per-unit cost of reducing pollution. EPA had developed a fairly helpful metric for comparing the pollution effects of different kinds of discharge, so we could calculate the cost per pound of reduction of this average pollution equivalent. What we found were differences across industries that were often of several orders of magnitude! Clearly, although we might have disagreed about exactly what the benefits of pollution reduction were, all those rules could not be correct. Some rules must be too costly, or some must not be costly enough.

It turned out that the explanation for these massive discrepancies was that many people at EPA believed that cost-benefit analysis meant financial analysis. That is, they would impose a much tougher standard on a particular industry if it were much more profitable than other industries. Their idea of cost-benefit analysis was to turn the screws on the pulp and paper industry, the pharmaceutical industry, and other highly profitable industries and to ease way back in steel and other industries that were experiencing commercial difficulties. This astounded us because it fit so neatly into many political economy theories of rent seeking and regulation. In fact, I believe that it has provided fodder for some subsequent academic articles.

As I look back on this class of regulations, I believe that we made very substantial headway in increasing the role of economics in policy-making. First, the range of pollution reduction costs in the effluent guidelines was much narrower at the end of the Reagan administration than at the beginning. Second, the people who were working on these matters at EPA came to take a very different approach toward effluent guidelines. Finally, the administration made similar progress in other areas of regulation involving health risks. When one rule imposed a cost of $900,000 at the margin per life saved and the next rule imposed a cost of $245 million at the margin per life saved, there was a problem. At $245 million we were clearly well above what anybody in the United States would ever spend for risk avoidance in their private lives. So we were able to eliminate a lot of harmful regulations at the top while maintaining a good deal of disagreement on what the appropriate level of spending should be lower down. In particular, I think we did a good job at putting an end to a large number of extraordinarily harmful and silly regulations in the area of hazardous air pollutants.

I want to mention also the lead phase-down regulation, which was a substantial victory for economic analysis and a case where OMB and the White House were reversed rather than EPA. One of our early targets for elimination was the lead phase-down regulation, which required that lead be removed from gasoline refining at a faster rate than would be dictated by the phasing in of new cars that required no-lead gasoline. A very fine piece of analysis persuaded everyone that the health harms of leaded gasoline were far greater than we had thought, and we ended up adopting a much tighter program than the one we had inherited. At the same time, the introduction of marketable lead permits saved many hundreds of millions of dollars from the cost of that regulation.

The third class of regulation that I want to address represents an intermediate case between price and entry controls, on the one hand, and externality regulation, on the other. The case that I have in mind is product standardization, which may be a component of health and safety regulation but often concerns a normal economic good. In particular, it often pertains to a new good or an innovation in an older good that somebody wants to have adopted uniformly.

This is the area of regulation where cost-benefit analysis is the most problematic in my view. Consider an innovation that increases people’s safety. It is easy to assume a certain level of effectiveness for this innovation and show that it would be cost beneficial to make every product conform to this new standard. But such an analysis ignores the fact that the optimal rate of diffusion of a new technology is not instantaneous but involves a learning process. Some individuals will gain more from this innovation than other people, and they will purchase the innovation at the price at which it will initially come on the market. As the innovation diffuses through the market, it will be improved much more, in terms of both quality and price, than if a government rule had universalized the initial innovation.

Consider the issue of passive restraints, which is bureaucratic argot for airbags in cars. When I was in the administration, economist Bill Nordhaus presented a very persuasive cost-benefit analysis showing that the airbag rule supported by the insurance companies would have substantial positive net benefits. This did not surprise me, but it did not show that it was desirable for the government to impose this technology on all consumers at once. I argued at the time that the most effective national airbag rule, in terms of promoting automobile safety, would be a constitutional amendment forbidding the federal government from making any rule having to do with airbags. My rationale was simply that airbags were a normal economic good with which we had very little practical experience. If the government made not rule about airbags, it seemed clear to me that they would be introduced and would diffuse through the market in about the way that car radios or sunroofs did. I never doubted that an affluent couple living in downtown Boston with three teenage sons would be foolish not to purchase a car with an airbag at the initial design and price or that a thirty-five-year-old single woman who always fastened her seatbelt, had a modest income, and lived in a rural part of Kansas would be foolish to purchase an airbag at the price at which it would first appear on the market. In fact, of one looks at the history of automobile manufacturers’ experiments with airbags, there is a good deal of evidence that they were seriously interested in the technology at an early stage. They turned away from airbags only when it became clear that, if this were a promising technology, they would be forced to install it instantly on every new car they sold.

Yet these arguments, which seemed highly persuasive to me and to the economists I was working with at OMB, seemed strange and irrelevant not only to program officials at the Transportation Department but also to political officials at the White House. They never became part of the administration’s public argument in the airbag controversy, and we eventually issued a complex rule that phased in airbags according to the pace of state legislation requiring the use of seat belts. And, as a general matter, I discovered that the executive order had much less practical effect in the intermediate case of product standardization than in the polar cases of price and entry controls and pollution controls.

Although it would be academically fashionable to attribute this difference to the large rents to be obtained by producer groups from product standardization, my casual impression is that the extent of political pressure brought to bear in this class of regulations was not much different than in the agricultural marketing orders and the EPA pollution controls. Instead, I attribute the difference to differences in the diffusion of economic thinking. The harms of government price and entry controls are widely understood and accepted and are bolstered by popular tales of farmers denied permission to farm their own land and of oranges left to rot in the fields. The notions of cost effectiveness and of the wastefulness of treating two identical pollution problems differently are also easy for noneconomists to grasp and apply. But the idea that the government should refrain from standardizing a product or a production process in a way that is abstractly “good” is more complicated—accepting it requires not only particular facts (as in the case of pollution controls) but also assumptions about the operation of private markets that laymen are often less willing to make than economists. It is an interesting puzzle that economists have been much more successful in persuading others of the evils of price fixing than of the evils of quality fixing.

Appendix

A. Regulatory Principles

Executive Order 12291 provides in section 2 that, “to the extent permitted by law,

(a) Administrative decisions shall be based on adequate information concerning the need for and consequences of proposed government action;

(b) Regulatory action shall not be undertaken unless the potential benefits to society for the regulation outweigh the potential costs to society;

(d) Among alternative approaches to any given regulatory objective, the alternative involving the lease net cost to society shall be chosen; and

(e) Agencies shall set regulatory priorities with the aim of maximizing the aggregate net benefits to society, taking into account the condition of the particular industries affected by regulations, the condition of the national economy, and other regulatory actions contemplated for the future.

B. Regulatory Policy Guidelines

Section 1 of Executive Order 12498 reaffirmed the following guidelines for rulemaking agencies originally set forth in 1983:

Regulations should be issued only on evidence that their potential benefits exceed their potential costs. Regulatory objectives, and the methods for achieving these objectives, should be chosen to maximize the net benefits to society.
Regulation of prices and production in competitive markets should be avoided. Entry into private markets should be regulated only where necessary to protect health or safety or to manage public resources efficiently.
Federal regulations should not prescribe uniform quality standards for private goods or services, except where these products are needlessly unsafe or product variations are wasteful, and voluntary private standards have failed to correct the problem.
Regulations that seek to reduce health or safety risks should be based upon scientific risk-assessment procedures, and should address risks that are real and significant rather than hypothetical or remote.
Health, safety, and environmental regulations should address ends rather than means.
Licensing and permitting decisions and reviews of new products should be made swiftly and should be based on standards that are clearly defined in advance.
Qualifications for receiving government licenses should be the minimum necessary. Where there are more qualified applicants than available licenses, the licenses should be allocated by auction or random lottery rather than by administrative procedures.
Where regulations create private rights or obligations, unrestricted exchange of these rights or obligations should be encouraged.
Federal regulations should not preempt State laws or regulations, except to guarantee rights of national citizenship or to avoid significant burdens on interstate commerce.
Regulations establishing terms or conditions of Federal grants, contracts, or financial assistance should be limited to the minimum necessary to achieving the purposes for which the funds were authorized and appropriated.”

C. Sources

The text of Executive Order 12291 is taken from Regulatory Program of the United States Government (Washington, D.C.: Executive Office of the President, Office of Management and Budget, March 1986), xiii. The “Regulatory Policy Guidelines” ere first set forth and elaborated in Reagan Administration Regulatory Achievements (Washington, D.C.: The White House, Presidential Task Force of Regulatory Relief, 11 August 1983). Executive Order 12291 was issued on 17 February 1981 and Executive Order 12498 on 4 January 1985.