The sudden collapse of communism has produced a nearly worldwide outburst of enthusiasm for private property; free markets, and electoral democracy. In the United States, and perhaps in the West as a whole, the collapse has also led to an understandable but disquieting degree of self-congratulation and complacency. Understandable, and in a way even justified, because the events of the past year have dispelled any lingering doubts about the risks of collectivism to both liberty and prosperity; disquieting, because the failure of communism is hardly a reason for unquestioning satisfaction with all our own institutions, let alone for a belief in laissez faire.
To free-market conservatives, however, communism's failure seems to validate their own arguments for the irredeemable failure of the more limited government regulation adopted in the West. For the last decade and longer in the United States, they have been particularly critical of the new "social" regulation concerned with health, safety, and the environment, which the consumer and ecology movements of the 1960s and '70s were instrumental in creating. This criticism of regulation has come especially from the fields of law and economics and, with recent judicial appointments, enjoys increasing influence in the courts. In the critics' view, social regulation is typically inefficient, counterproductive, and indefensible in principle because it amounts to unjustified public meddling in private affairs. Deregulation, therefore, seems to them a blessing in more ways than one.
But this view is myopic. There are convincing, principled arguments for regulation that are not paternalistic. And, overall, social regulation has not been a failure. On the contrary, it has helped clean up the air and water, protected endangered species, and saved many lives.
Of course, there have been failures, some of them quite serious. The good news is that the failures come in identifiable patterns and can be avoided in the future. What we learn from the failures should point us in the direction of new regulatory alternatives.
Those alternatives often call for use of market-like incentives to minimize adverse side effects, reduce costs, and accomplish regulatory goals. But even in these cases the framework of market forces will depend on legal and political choices, and as I shall argue, these should properly reflect broader democratic aspirations than the unregulated market ordinarily registers.
Why We Regulate
No industrialized democracy can or should do without a substantial regulatory state. Although the private market is usually an engine of individual freedom and economic welfare, it sometimes breaks down, and then democratic controls are the only real option.
Some of the reasons for regulation come under the heading "market failure," as conventionally described by economists. For a variety of reasons, the market often fails to provide consumers enough information on products or services. Risks posed by carcinogens in the workplace or in food and drugs are a good example. When too little information is provided, some sort of government response is indispensable. Suitably designed, such information-enriching regulation can promote freedom of choice while protecting lives and health, often at low cost.
Sometimes private conduct imposes harms on other people, and the costs of those harms, or negative "externalities," are not taken into account by the marketplace. The familiar example is pollution. Environmental harms caused by coal-burning power plants are severe, and they fall on numerous people far from the plants. The people who are harmed (not just living adults but children and future generations as well) lack the organization to bring suit or otherwise to seek redress. A less obvious example of distant harms is the loss of endangered species, which may provide medicinal, agricultural, educational, and recreational benefits to human beings. In these cases, there is ample justification for a regulatory response, forcing those who produce harm to pay for the damage and ultimately to reduce or eliminate it.
Somewhat more subtly, government controls can sometimes help facilitate an outcome that people want but cannot obtain without governmental assistance. The citizens of a community or state may want a recycling program -- provided that everyone participates. But if there is no way to make everyone participate, they will not recycle on their own. (People often want to promote the public good, but they don't want to be suckers.) The option preferred most by most people -- communities when everyone recycles -- can be obtained only with governmental help. A governmentally mandated recycling program of the son enacted in nearly a dozen states can satisfy private wishes far better than a market that fails to make universal recycling available.
In other words, instead of overriding private choices, government regulation sometimes helps to fulfill them. The point is a broad one, relevant not only in the environmental area, but in energy conservation, automobile safety, and elsewhere as well.
These arguments involve standard cases of market failure. Even economists and others firmly committed to laissez faire should accept them in principle. But the case for regulation also draws on democratic principles, not just economic ones. In their capacity as citizens, people have political desires that are different from their preferences as consumers. Some, for example, seek stringent environmental laws even though they do not use the public parks; they support measures protecting endangered species even though they do not give to environmental groups or expect to benefit from those measures. As citizens, these people support government regulation that diverges from their behavior as consumers.
Some economists have derided the distinction between political and consumption choice by suggesting that political behavior is a confused reflection of "actual" choice because voters, unlike consumers, do not face the full costs of programs they support. But this explanation is far too crude. As citizens, people may seek to implement their highest aspirations or most altruistic goals through government. Social and cultural norms often press people, as citizens, in the direction of a concern for others or for the public interest; such concerns should not be disparaged. The collective and deliberative character of democratic politics helps to provide a fairly mundane explanation for this phenomenon. People may not want to satisfy their aspirations, or to be altruistic, unless they are sure that other citizens will be doing so as well. And the very process of public debate may bring out more concern for the community's longer-run interests than is evident in personal consumption.
Politics may also help to overcome resignation. Without the possibility of common action, current practices and conditions often seem intractable to us individually. We accordingly adapt, not only our actions, but even our desires. Psychologists have long observed that most of us typically adjust what we want in life to what we can get. If something bad seems inevitable, we may feel, "That's tough, but that's the way it is."
But if we can act in concert and through government, we may come to expect and want more -- and support change. Social movements involving the environment and occupational health and safety provide conspicuous examples. Where the air and water have long been filthy, people may seem indifferent to it, accepting the consequences for their health. But when political action and government regulation open up other possibilities, popular preferences may shift. In other words, regulation may be justified as a means of increasing opportunities and thereby enabling people to express desires otherwise repressed.
All these information-enriching, externality-correcting, opportunity-increasing effects of regulation suggest that there is a sound basis for regulation in liberal democratic principles, despite the frequent objection that regulation is "paternalistic" or insufficiently respectful of private choice. In some cases -- inadequate information, harms imposed on third parties, and governmental coordination of private behavior -- regulation facilitates the satisfaction of private desires. But people in a democracy may also have collective aspirations or altruistic goals that override the outcomes of markets. To satisfy such aspirations should hardly be seen as an objectionable interference with freedom -- even in a system that prizes, as all systems should, private property, freedom of contract, and other voluntary arrangements.
What the Record Shows
How have these considerations worked out in practice? In fact, the record shows many successes. In environmental protection, automobile safety, protection of endangered species, and other areas, regulation has made things much better.
The effort to reduce air pollution has produced huge gains. Since the 1970s the United States has seen substantial decreases in both levels and emissions of all major pollutants, including sulfur dioxide, carbon monoxide, lead, and nitrogen dioxide. Lead reductions have been especially dramatic, showing no less than a 94 percent decrease in ambient concentrations between 1975 and 1988. Transportation emissions have been cut from 122.6 million metric tons in 1975 to merely 3.5 million in 1986. (Further reductions have occurred since, and more are on the way.) The average lead levels in children's blood decreased by 37 percent between 1976 and 1980. More generally, the vast majority of counties in the United States are now in compliance with national air quality goals with respect to all of the major pollutants. Though there is substantial room for improvement, air in the United States is much cleaner and healthier than it would have been without regulatory controls.
Significant gains can be found in controlling water pollution as well. The Great Lakes are much cleaner than they were in 1965, when hundreds of lake beaches had to be closed. Especially dramatic have been the extraordinary reductions in levels of DDT, PCB, and dieldrin contaminants in Great Lake fish. Nationally, mercury in lake sediments is 80 percent lower. Lead and nitrate reductions have produced significant improvements in water quality. A number of harmful nutrients have been reduced by nearly 50 percent in national rivers. As a result, many rivers, including the Potomac, the Hudson, and parts of the Mississippi, are cleaner and safer.
Environmental regulation has thereby dramatically decreased harmful substances in animals and human beings. The percentage of DDT in body fat has declined by 79 percent; PCBs in body fat have been cut by 75 percent. At the same time, a number of programs have helped protect threatened or endangered species, including the bald eagle. Some protected species have shown large recoveries.
Regulatory successes are hardly limited to the environmental arena. Automobile safety regulation in the late 1960s and early 1970s has achieved enormous gains, measured in injuries prevented and lives saved. Automobile fatalities would have been about 40 percent higher in 1983 if not for governmental controls. Between 1966 and 1974 about 28,000 lives were saved as a result of occupant safety standards. The annual benefits from regulation are extremely high; indeed, they have been estimated, in monetary terms, as exceeding $10 billion. Moreover, the ratio of benefits to costs is excellent. Some of the regulations pay for themselves in terms of health and related savings. The large number of deaths prevented is purely a bonus.
Though the overall record of the Occupational Safety and Health Administration (OSHA) is quite mixed, its regulation of asbestos prevents an estimated 396 deaths per year, and it does so at relatively low cost. (Asbestos removal from buildings is another matter.) More generally, studies of the costs and benefits of regulatory initiatives show that a number of measures have saved lives extremely cheaply Regulation of trihalomethanes by the Environmental Protection Agency (EPA) saves a life at only $315,000; the National Highway Traffic Safety Administration's (NHTSA) fuel system integrity controls, $300,000; the Consumer Product Safety Commission's mandatory smoke detector rule, between $0 and $85,000; NHTSA's roadside hazard removal rule, $0.
To be sure, there are serious problems with using cost-benefit analysis to evaluate social regulation, especially insofar as we rely on some measure of private "willingness to pay" as the basis for assessing costs and benefits. As I have noted, social regulation is often based on democratically expressed preferences rather than private consumption choices. If they differ, the use of private willingness to pay to value costs and benefits is perverse, since it measures the worth of democratic aspirations by reference to the very consumption choices that citizens may be attempting to override.
Moreover, willingness to pay is a function of ability to pay. In a world where income and wealth are unequally distributed, it is decidedly odd to say that government policy should hinge on how much individuals are able to pay for environmental quality and occupational safety or health. For these reasons, cost-benefit analysis must be approached cautiously. But as a crude effort to assess regulatory strategy, it can be a useful tool. Sometimes it will reveal that the gains of regulation are far greater than the losses; sometimes it will show the opposite. Tradeoffs are inevitable here, and it is striking to see that a comparison of costs and benefits, even so narrowly understood, often favors social regulation.
The Limits of Current Strategies
To point to regulatory successes and to the limits of cost-benefit evaluation is not to deny that regulation has frequently failed. Sometimes it has imposed enormous costs for speculative benefits, however these are measured; sometimes it has aggravated the very problem it was designed to solve; sometimes it has accomplished little or nothing. For example, the United States spent no less than $632 billion for pollution control between 1972 and 1985, and some studies suggest that alternative strategies could have achieved the same gains at less than one-quarter of the cost. The fuel economy standards for new cars have led manufacturers to produce smaller, more dangerous cars. Some of OSHA's carcinogen regulations impose enormous costs for uncertain gains. Indeed, the pattern of OSHA regulation of carcinogens is a crazy quilt, with some types of risk unregulated and others regulated at costs up to $40 million per life saved. No matter how we assess social regulation, there can be little doubt that we have far too stringent government controls in some areas, and far too little in others.
Some of OSHA's expensive regulations for design of the workplace have produced no discernible safety improvements for workers. The EPA has promulgated regulations to control only a few of the many toxic substances in the air. And by delaying the entry of beneficial drugs onto the market, the Food and Drug Administration has sometimes increased risks to life and health -- as the gay community came to recognize early in the AIDS epidemic.
Anyone concerned about improving the regulatory system has to ask: How might the failures be prevented? The first step here is to identify the sources of failure.
A number of factors are at work. The very structure of our government may make it difficult to coordinate different programs covering related aspects of the same problem, or to deal with complex tradeoffs among competing social goals. Regulatory systems, like the technologies being regulated, face obsolescence over time. And as political scientists have well documented, private groups sometimes "capture" the regulatory process.
Many failures of the regulatory process thus have political foundations. Parts of the 1977 Clean Air Act, for example, reflected the influence of representatives of Eastern states that produce dirty, high-sulfur coal. At the time, the chief spokesmen for those interests, Robert Byrd of West Virginia, was Senate majority leader. Environmentalists joined Byrd and others in an alliance to pass legislation imposing stringent technological requirements on all new pollution sources, even though Western coal is largely free of sulfur and needs little cleaning at all. As a regulatory strategy, this was not sensible. Western coal does not need to be scrubbed, or at least Eastern coal needs more scrubbing; and requirements imposed only on new plants help to perpetuate old and worse sources of pollution. But while this strategy was not good environmental policy, it was a way of protecting Eastern coal against competition from cleaner rivals in the West.
Such examples, however, are no reason for fatalism. In the debates over regulation, criticism of irrational provisions has had an impact. As Martha Derthick and Paul Quirk show in their book The Politics of Deregulation, the decision to end much of the price and entry regulation of the telecommunications, trucking, and airlines industry emerged from a battle of ideas (led originally, as many now forget, by Ralph Nader and Senator Edward M. Kennedy). No matter what one thinks of those decisions, they do show that Congress can override the power of interest groups for the sake of what many believe to be good policy This past year, with clean air legislation again before Congress, Senator Byrd lost when he proposed amendments to protect West Virginia coal interests, though it no doubt helped, of course, that the new Senate majority leader hails from Maine.
If we cannot always be certain of defeating coalitions that have an interest in regulatory failure, we can try to focus attention on the most important errors of policy Those typically he in the creation of poor incentives and a concern more with the symptoms than with the underlying causes of problems. When social regulation relies on strategies that fail to change underlying incentives, it often generates unfortunate systemic consequences.
Perhaps the most important of these ill-conceived strategies is the use of rigid, highly bureaucratized, command-and-control regulation, which dictates, at the national level, rules for hundreds, thousands, or millions of companies and individuals in an exceptionally diverse nation. Command-and-control regulation is a dominant part of American government in such areas as environmental protection and occupational safety and health.
In the environmental context, command-and-control regulation usually involves requiring industry to adopt the "best available technology" (BAT), though only for new pollution sources. BAT strategies are a pervasive -- indeed, defining -- feature of federal regulation of the air, the water, and the workplace.
BAT strategies ignore enormous differences among plants and industries and among geographical areas. In view of these differences, nationally uniform technological requirements are wildly inefficient. It makes no sense to slap the same technology on industries in diverse areas -- regardless of whether they are polluted or clean, populated or empty, or expensive or cheap to clean up. The problem goes deeper still. By requiring all new industries to adopt costly technology, and by allowing more lenient standards for existing plants and industries, BAT strategies actually penalize new products, thus discouraging investment and perpetuating old, dirty technology.
Such strategies also fail to encourage better pollution control technology than the best now available. Under the BAT approach, a company that innovates will simply have to invest more in pollution control; there is no reward for innovation. BAT strategies are also extremely expensive to enforce, imposing on the EPA and OSHA an extraordinary monitoring burden.
Perhaps worst of all: By focusing on the technology at the end of the pipe, BAT approaches are aimed at symptoms rather than underlying causes of pollution. For example, they deal with sulfur dioxide emissions, which are the major source of acid rain, by forcing coal-fired power plants to adopt costly "scrubbing" methods. A far better approach would be to encourage American consumers and industries to increase energy conservation and efficiency, and to shift to cleaner, renewable fuels.
BAT strategies are different from so-called "technology-forcing" statutes, which do require innovation. Technology forcing has often accomplished considerable good. Its major drawback is that government is unlikely to know how much technological advance is realistic or desirable. So technology forcing involves a lot of guesswork.
The BAT strategy embodies several paradoxes of the regulatory state. As noted, a government that requires the best available technology actually discourages technology development. It does so by making it expensive rather than rewarding for companies to develop better pollution control equipment.
BAT strategies also embody the paradox created by the pervasive practice of imposing especially stringent controls on new sources of pollution, while regulating existing industries leniently or not at all. (This is, of course, a strategy that existing industries favor; perversely, it turns regulation into a brake on new competition.) The paradox arises because a decision to regulate new risks in the interest of health and safety will perpetuate old risks and in that way impair health and safety. By increasing the costs of new sources and products and encouraging consumers to resort to old ones, the new risk strategy has unquestionably perpetuated the use of old, especially dirty power plants, and increased the longevity of old, especially dirty automobiles. A good way to increase automobile pollution, at least in the short run, is to impose extremely costly pollution control requirements on new cars alone. Some of our law does precisely that.
A final regulatory paradox, also embodied in BAT approaches, is that some strict government controls, or legal requirements that forbid tradeoffs, have actually produced underregulation as well as overregulation. By threatening to impose draconian requirements on the private sector, strict controls give industry the most powerful incentives to fight regulation wherever it can. Rigid controls and the prospect of long conflicts with regulated groups also give the regulatory agencies a powerful incentive to do nothing at all.
It should, therefore, come as little surprise that despite (or, better, because of) the stringency of the relevant laws, the government has regulated only seven toxic air pollutants (out of hundreds), fewer than a dozen toxic water pollutants (again, out of hundreds), and until relatively recently only ten toxic substances in the workplace (still again, out of hundreds) -- this last even though the private organization that once performed some of OSHA's functions had for many years recommended lower exposure limits for hundreds of chemicals.
To be sure, the few regulated substances are stringently controlled. Indeed, because the law forbids balancing, some regulation of carcinogens is excessive, since Americans incur high costs for speculative gains. The predictable consequence of an especially stringent standard is therefore both underregulation and overregulation.
The picture that emerges is mixed. In some areas, government regulation has produced huge improvements. In other areas, regulation has been counterproductive, ineffectual, overly costly, or nonexistent. The cause of failure has been most of all faulty structures: measures that do not take account of the incentive effects of government strategies, that address only one part of a complex problem, that disregard side effects, and that deal with symptoms rather than causes.
What Can Be Done?
There is good news. This account of the performance of the regulatory state should lead to a set of an entirely feasible reform strategies. And contrary to the pessimism of many observers (and more than a few economists), Congress and administrative agencies are capable of carrying out those strategies and, indeed, have started to do so in some areas.
At the most general level, we should be attentive to the incentive effects of regulatory statutes and the possibility of strategic or self-interested adaptation by regulated groups and administrative agencies. Regulators should often attempt to address pollution and other harms at their source. It is better, for example, to eliminate lead and other dangerous pollutants from gasoline than to impose complex techological requirements on pollutants as they come out of the tailpipe. Pollution prevention, rather than technological fixes, should be a guiding principle for environmental policy. In general, we should reduce the levels of dangerous substances before they am introduced rather than control those substances that have already been introduced.
Moreover, Congress should create incentives for private cooperation in the development of regulations. California now requires industries to disclose toxic substances to the public, until a regulation has been issued establishing that the risks are insignificant. As a result, instead of seeking to create a regulatory logjam, companies in California have an interest in accelerating the regulatory process. There is no better way of getting regulations than to create incentives for industry itself to want them. This strategy makes the burden of inertia work in favor of the public's interest in health and safety, not against it.
The account here suggests that government should generally avoid "BAT" strategies, seek to reduce old risks as well as new ones, and as a rule require some form of balancing between the disadvantages and advantages of regulation.
Ideas of this sort have direct implications for modem regulatory reform. A general first step here is to adopt a strong presumption in favor of flexible, market-oriented, incentive-based regulatory strategies. Such strategies should be focused on ends -- the number of lives saved, the amount of pollution reduced -- rather than on the means of achieving those ends. Within the framework set by law, the means should be left primarily to the market.
For example, many observers have persuasively argued that the Clean Air and Clean Water Acts should be amended to substitute an "emissions trading" program for the BAT approach. The current versions of the proposed Clean Air Act that have passed the House and Senate contain emissions trading programs to deal with acid rain. Emissions trading programs have been controversial, but they hold out enormous promise if properly designed. Such a program would not require all industries to have the same technology, as determined by officials in Washington. In its ideal form, it would make two simple changes in current law.
First, an emissions trading program would simply require polluters to pay to pollute, unlike current law, which, astonishingly, allows firms to pollute for free. If a company is to send sulfur dioxide into the air, it would pay a fee to obtain a permit to do so. The amount to be paid, like the total amount of pollution, is a matter for democratic choice.
Second, the permits should be tradeable after they have been issued. A company that is able to reduce pollution below the permitted level should be allowed to sell all or part of its permit to someone else. If a company is able to bring about substantial reductions, it would receive large economic gains, since it could sell its permit to another polluter. In one bold stroke, such a system would create market-based disincentives to pollute and market-based incentives for pollution control. Such a system would reward rather than punish technological innovation in pollution control -- and do so with the aid of private markets.
Programs of this sort have often been criticized, especially by environmentalists (though the resistance is diminishing, and to its credit the Environmental Defense Fund has long been a strong advocate of emissions trading). It is true that such programs must be carefully designed and administered to ensure (among other things) that polluters who receive permits through trades do not concentrate in the same areas and thus create dangerous "hot spots." But most of the criticisms are unpersuasive or ill-informed. Contrary to charges that emissions trading will increase pollution levels, the program is simply a way of achieving whatever degree of pollution reduction we want at the lowest possible cost. If we want to reduce pollution very sharply, we will decide to give out few permits. Indeed, a large advantage of emissions trading is that the democratic process will be deciding on pollution levels -- the key question, after all -- rather than focusing on the often unintelligible and largely incidental question of what control technology is "best" or "available."
The same basic ideas support a trading system in the international arena. Under such a system, some countries would pay others to reduce emissions levels or to protect biological diversity. Emissions trading would be especially desirable to control the forms of pollution that cross national boundaries. Consider here the greenhouse effect, which is produced by energy sources especially likely to increase in developing nations. A trading system would produce more efficient reduction techniques and also be more equitable than an approach that would require all countries to use the same control technology. Thus, for example, the United States might help a developing nation to retire its debt, or to use new energy conservation technology, so long as that nation agreed to keep emissions of greenhouse gases at a low level or to protect biological diversity.
Of course, emissions trading is only one part of a sensible environmental program. For the most dangerous pollutants, and for those that are to be eliminated altogether, pollution prevention should be the preferred strategy.
To control pesticides and other toxic substances, the simplest reform strategy would be to tax and thereby to discourage their use. Taxes would also encourage people to use other methods of pest control, including methods that do not rely on chemicals at all. Some biological techniques have proved quite successful, and cheaper in the long run. The trick is for government to use economic incentives to get farmers to use them. Of course, here, as with other pollutants, the most dangerous pesticides should be banned altogether.
The problem of waste disposal could be handled similarly. The ultimate goal should be to minimize the total amount of waste, rather than simply to treat whatever waste has been produced. To promote recycling and to discourage the accumulation of solid and hazardous wastes, government could make it more expensive to dispose of substances in landfills or the oceans. A tax or charge for waste disposal would improve the current system, which relies more on waste treatment than on waste minimization and leads companies to create "paper trails" merely notifying the government of disposal practices. Waste treatment would, of course, continue to be necessary, but it would be part of a coordinated program of waste management.
Similarly, the problem of automobile pollution should not be handled only through technological requirements for new cars, which have a limited effect. Instead, or in addition, we might broadly adopt plans to reduce dependence on the automobile, including an "old car" tax, subsidies to mass transportation, and restrictions on parking in the major cities. Increased taxes on gasoline are an especially promising strategy. They would not only promote energy conservation and independence; they would also be an effective antipollution strategy, helping to reduce a range of environmental problems.
The basic principle that social regulation should employ incentives and market pressures, rather than centralized rule-making, has general implications for social regulation. In the area of occupational safety and health, for example, we should rely more on employers' actual performance in reducing deaths and injury and less on their compliance with rigid and unrealistic national standards for design of the workplace. Possible techniques here include a tax on employers for maintaining dangerous conditions, greater reliance on disclosure of risks to workers, and more active involvement of employees in monitoring workplace safety. This strategy has proved successful in other places, including Sweden; it enlists employees in the effort to reduce occupational risks and thus increases decentralization, participation, and voluntariness, while decreasing dependence on government.
All these are simply illustrations of the enormous opportunities we now have to accomplish the goals of social regulation without returning to discredited principles of laissez faire or continuing with the rigidities and perverse effects of current programs. In evaluating reforms, little is to be gained by generalities that point to the frequent problems in either the regulatory state or the market. The problems are too particular, too dependent on the specific context, to allow for global prescriptions. It is far more helpful to rely on a closer analysis of how both markets and regulation are likely to break down -- to learn, in short, from the past. In the reform of regulation, we should be entirely unembarrassed by the use of government to carry out democratic aspirations and to promote economic welfare. But we should also insist on strategies that embody the flexibility, adaptability, productive potential, and decentralization characteristic of private markets.