The Cato Review of Business & Government
Environmental User Fees
While Regulation predicted years ago that the growing budget deficit would cause agencies to find creative ways of substituting regulatory mandates for budget expenditures, we failed to predict that regulation itself would become an attractive source of revenue. In the search for "services" for which the government might charge a fee, the Office of Management and Budget and several regulatory agencies, most notably the Environmental Protection Agency (EPA) and the Food and Drug Administration, have decided that reviewing new products for safety is a service to the applicant. The Administration has proposed collecting large "user fees" from anyone seeking approval for a new drug, pesticide, or other chemical. It is requesting authority to put the revenues from these fees into special funds not subject to the appropriations process.
This proposal is ill-conceived whether evaluated on grounds of good regulatory policy or sound public finance. Moreover, it is a particularly unpromising source of revenues.
User fees are attractive in concept. When applied correctly, they enhance economic efficiency and at the same time raise revenues. Few federal taxes have both of these virtues. User fees also appear to have particular advantages in environmental management. The EPA is in the business of rationing things that are scarce-like clean air, clean water, and waste disposal sites- and user fees, or prices, are the best method we know of for rationing scarce resources.
Since the day the EPA opened for business, economists have been urging it to avoid rigid standards and to experiment with Pigovian-style pollution taxes (or their near-equivalent, marketable permits sold at auction). These user fees are scaled to extract from each polluter a charge proportional to the damage done, in an attempt to internalize the costs of pollution and to allow the market, with all its rich incentives, to find the most efficient way of improving the environment. Pollution taxes of this type could simultaneously satisfy the goals of regulatory reformers, environmental protectors, and budget balancers.
Yet pollution taxes have little hope of becoming a reality. Congress is not likely to impose them for compelling political reasons, and the EPA lacks the authority to experiment on its own. Besides, the EPA has been distracted from pursuing pollution taxes by two lesser ideas: "economic" noncompliance fees, and administrative user fees.
The first of these ideas preoccupied the EPA during the Carter Administration. Its development is credited to them-Administrator Douglas Costle and his assistant Administrator for Policy and Management, William Dryton, who first employed economic noncompliance fees at the state environmental agency in Connecticut.
The Drayton-style noncompliance fee is an enforcement penalty. Ordinarily fines for violating a standard are spelled out in a statute. For example, the law may provide for a fine of up to $10,000 per day of violation which, below that limit, can be scaled to the "seriousness" of the violation.
But often judges or juries diminish the amount of a fine in sympathy for the company or appears to be at risk of failure if the full fine is assessed. In Connecticut, Drayton and Costle found that some of the major air polluters in the state continued to violate emissions standards because the state agency could not persuade a court to impose a fine large enough to make compliance a more attractive option for the companies that continues noncompliance.
The Drayton initiative is designed to remedy this problem. Whenever a company is found to be emitting a pollutant in excess of the applicable standard, a penalty is imposed that includes two components. One component is equal to the excess profit that the company earned by avoiding compliance with the regulation; the other is a punitive component designed to time the scales in favor of compliance.
The EPA has sophisticated financial models to calculate exactly what a company's excess profit is. In effect, this component of the fine takes back from a polluter his "ill-gotten gains"; it demonstrates persuasively that crime doesn't pay, and it appeals to judges and juries as a fair penalty. The EPA has found that this type of fine is rarely reduced by a court, unless the company can show the agency's calculations to be in error. The Drayton noncompliance fee is designed by and for lawyers, and it has to be judged a success at what it set out to do. It is, all in all, a creative application of economics in the service of law enforcement.
It has little appeal for economists, however, because of one unfortunate flaw: it uses economic incentives to aggravate the worst features of command-and-control regulation, rather than to ameliorate them. It eliminates the possibility of what would be called, in contract law, "efficient breach." There are certainly occasions that arise when the cost of complying with a regulation soars to impossible heights and the public benefit from compliance drops to trivial levels. But instead of tolerating an optimum amount of noncompliance, the Drayton initiative ensures blind adherence to whatever the standard happens to be. In a system of standards that is already too rigid, it removes the last wiggle of flexibility.
The development of the EPA's policy on noncompliance fees in the late 1 970s prevented the agency from exploring more efficient types of enforcement penalty. In economic terms, the agency's mistake was to focus all of its attention on figuring out the "reservation price" of the regulated companies, rather than the reservation price of the public.
It would be far more efficient for the EPA to research the damages being done to the community by noncomplying firms, rather than the profits being earned. Properly presented, a damage-based penalty would be persuasive to judges and juries. In addition, it would allow the EPA to apply its prosecutorial discretion and its authority to collect enforcement penalties as a form of user fee for the environment.
The Reagan Administration's user-fee initiative is equally wrong-headed. The EPA proposes to charge a fee for processing applications for new product approval, equal to the agency's administrative cost of reviewing the application. This focuses the burden of the user fees where they have little prospect of improving efficiency, and indeed where they are likely to do the most damage.
Our regulatory apparatus is already far too preoccupied with new technology. A disproportionate share of regulatory resources is devoted to what Peter Huber has dubbed "gatekeeping"-giving close scrutiny to any new entrants in an industry to ensure levels of safety many times stricter than what we tolerate among existing products or firms. (See "Exorcists vs. Gatekeepers in Risk Regulation," Regulation, November/December 1983.) Generally, new pesticides, new chemicals, and new drugs present small risks compared with older technologies-either because technology has improved, or simply because their market share is so much smaller than that of existing products.
Yet the bias against new sources of risk is pervasive. New sources are very easy targets for regulators. They can be held hostage in a way that no existing source would ever tolerate. The consequences of the bias are severe, because the misplaced regulatory focus inhibits both innovation and competition. The flaw in the Reagan Administration user-fee proposal is that it attempts to collect revenues from the very same sources that already are bearing the brunt of regulatory costs-new firms and new products.
The proposal may also create perverse incentives for the agencies, turning them from gatekeepers into toll collectors. Asking companies to pay an entry fee invites an agency to fall into the same traps that plague economic regulation: becoming a captive of the regulated industry, using licensing authorities to protect the industry from competition, and in turn relying on the industry for revenues and political support.
The real opportunity for the EPA is to charge user fees for using the environment-not for using the services of the agency. If the EPA decides to limit lead in gasoline, or acid rain, or chlorofluorocarbons, then user fees are an efficient solution. They force firms to pay for the damage done to the environment or for the scarcity value of the resources that they consume. There is no need for an entry fee-new and old alike should pay in proper proportion.
The idea that should be kept in mind in thinking about user fees at the EPA is that clean air and water are the resources that are scarce and that should be charged for. Federal bureaucrats are not particularly scarce, and should not be the focus of user fees.
Last year, the Commerce Department issued regulations that restrict the freedom of American companies to take photographs from space. Under the regulations, the Departments of State and Defense can veto a private company's application for a license to operate satellites that produce high-resolution photographs. Furthermore, license holders are required to "operate the system in a manner consistent with national security and the international obligations" of the United States. The sanctions for violations are seizure of the "offending" equipment and fines of up to $10,000 per day.
The regulations implement a provision of the Land Remote-Sensing Commercialization Act of 1984. Ironically the main purpose of this law was to speed the private development of space by privatizing land remote-sensing satellites, Landsats, owned by the U.S. government. (The new owner and operator of the Landsat system is the Earth Observation Satellite Company, EOSAT, a joint venture of RCA and Hughes Aircraft Corporation.)
The greatest objections to the regulations are being raised by the news media. This is not surprising since high-resolution photographs from space promise to make a major contribution to journalism. News reports on the Chernobyl nuclear accident, the Soviet space-shuttle site at Tyuratam, and the Soviet radar facility near Krasnoyarsk are among those that have been bolstered by satellite photographs.
What are the journalists' objections? An attorney for the Radio and Television News Directors Association, Robert J. Aamoth, for example, argues that the regulations violate the First Amendment. The position of the RTNDA and other news organizations is that taking pictures is news gathering, which is a freedom covered by the First Amendment's protection of freedom of the press.
There is a certain logic to this argument. When a reporter covers a political speech, for instance, he is gathering news. Is the newspaper's First Amendment right to print the story protected if the reporter has no right to cover the speech? If it were, one could imagine a situation in which freedom of the press was "guaranteed' '-but reporters were legally prevented from leaving their buildings or using their telephones. As the US Supreme Court recognized in Branzburg v. Hayes (408 US 665, 1972),"without some protection for seeking out the news, freedom of the press could be eviscerated."
Media organizations also point out that the Commerce Department's rules fail to spell out in advance all the requirements that Licensees must meet, and more important, all the reasons that licenses may be revoked. Officials in the State and Defense Departments now have enormous discretion to decide whether a company's particular use of photographs has violated the terms of the license.
This pragmatic objection is crucial to evaluating the prospects for private ownership of satellites capable of producing high-resolution photographs. According to Congress's Office of Technology Assessment (OTA), a media satellite would cost $215 million to $470 million to build and launch, and another $10 million to $15 million per year to operate. Even before the regulations were issued, private investors had demonstrated their reluctance to make such investments; no private company had chosen to produce and launch a media satellite. As a result of the regulations, investors are now confronted with the risk that the government may confiscate their capital for unanticipated reasons.
US military officials defend the regulations on the grounds that the press, by printing or broadcasting photographs from space, could seriously harm our national security. A photograph of a secret US troop movement, for example, could compromise important US military operations.
This is a legitimate concern. The question, though, is whether restricting American compaflies' freedom to launch satellites to take photographs from space will solve the problem. Most likely, it will not.
Two factors undermine the probable effectiveness of such restrictions. First, the right photograph taken from the ground or from an airplane can be just as harmful as one taken from space. No licenses are required to take such photographs today.
Second and more important, nothing in the regulations prevents the US media from buying high-resolution photographs taken by the satellites of other countries. Indeed, some of the most striking photographs used in news reports to date have been taken by the French government 's SPOT satellite and sold through a private French company to the American media. And ironically, only three days after the Commerce Department issued its regulations, the Soviet Union announced plans to sell photographs to the media. The Soviet photographs have a resolution as high as six meters (that is, the photographs can reveal objects measuring six-by-six meters on earth), as compared to 10 meters for SPOT and only 30 meters for Landsat. That there are competing sources of supply of satellite photographs-even in countries with competing forms of government-minimizes the harm to US security that can be done by privately owned American satellites.
Moreover, a case can be made on national security grounds against restricting satellite photography. A look at the Western media's use of satellite photographs to date is instructive. Of 16 major uses of satellite photos listed in a 1987 study by the OTA, 15 were of closed (that is, unfree) societies. Of the 15, 10 were of the Soviet Union-8 of these were of military sites and 2 were of Chernobyl-2 were of the Iran-Iraq war, 2 were of Libya's military, and 1 was of Pakistan's controversial nuclear processing facility. Just 1 was of New York City's harbor.
This is no accident. Not much happens in the West that photographers are not allowed to photograph using more conventional means. We simply do not have much to hide. Not so for the Soviet Union or other closed societies. Photographs from space narrow the information gap between closed and open societies. Allowing unrestricted use of pictures from private American satellites would reduce the gap even further.
This leveling of the information is useful because it makes the Soviet Union and other closed societies less able to hide an event such as the Chernobyl disaster. It also constrains our government to be more honest with us about what the closed societies are up to: the American Broadcasting Company's photographs of the Soviet phased-array radar facility at Krasnoyarsk- thought to violate the SALT II treaty-are a case in point. Certainly using private American satellites seems preferable to relying exclusively on Soviet and French government satellites.
A novel case for restricting the press's access to satellite photographs was given by Dino Brugioni in a 1986 Washington Post article. Brugioni, a former photo interpreter for the Central Intelligence Agency, argued that aerial photography and multisensory imagery are too important as "sources of information and credibility for the analysis to be left to amateurs." In short, the media might misinterpret photographs. But members of the media regularly misinterpret all kinds of data, including some very important economic data. Does anyone seriously argue that we should restrict their access to basic information about the economy?
President Reagan introduced a new wrinkle in the debate over licenses on February 11,1988, when he announced his "Commercial Space Initiative." Point 7 of this 15-point program covers remote sensing: The administration is encouraging the development of commercial remote sensing systems. As part of this effort, the Department of Commerce, in consultation with other agencies, is examining potential opportunities for future Federal procurement of remote sensing data from the US commercial sector.
A related National Security Decision Directive the President signed on January 5 also calls for the encouragement of US commercial remote sensing system that are "superior to foreign-operated civil or commercial systems." Presumably "superior" refers to, among other things, the resolution the systems are capable of.
It is unclear what effect these recent pronouncements from the White House will have on the Commerce Department's regulations. Regardless of the fate of the regulations, the US media will broadcast and publish photographs from space. The question that remains is which country's satellites will supply them.
On March 14,1988, the Consumer Product Safety Commission(CPSC) and the five US distributors of all-terrain vehicles reached a controversial legal settlement that, among many other things, prohibited all future sales of three-wheeled ATVs. Four-wheeled ATVs are still available, but may be sold only to certain age groups, and must be accompanied by warnings and an offer of free safety training. The settlement ended a lawsuit in which the CPSC sought to order a mandatory recall and refund program for all three-wheeled ATVs already sold, as well as for any four-wheeled machines that had been sold for use by youngsters. It was controversial mainly because several interveners in the lawsuit argued forcefully (in the press and in Congress, as well as in court) that the CPSC should refuse to drop its demand for recall.
To obtain a recall, the CPSC needed to prove to the court that the ATVs were not merely defective, but "imminently hazardous." In various analyses, the commission argued that three-wheeled ATVs were more hazardous than similar vehicles possessing one more wheel (four-wheeled ATVs, one less wheel (dirt bikes), and no wheels at all (snowmobiles) On the basis of the evidence complied, the most that an unbiased observer might confidently conclude is that all of these vehicles are associated with a certain amount of risk and that, until someone markets a motorized unicycle, none of them stands our as uniquely hazardous.
The regulatory struggle over ATVs began four years ago, in March 1984, when the CPSC staff began investigating the cause of the rising number of injuries and deaths associated with ATVs. In 1985 the commission published a seven-step action plan in the Federal Register, followed by an advance notice of proposed rulemaking. At this point the CPSC appeared mainly to be interested in working out some voluntary safety measures with ATV distributors. One commissioner would not even concur with the CPSC's preliminary finding that there "may be an unreasonable risk" associated with ATVs; two commissioners (of four then sitting) testified that there was no basis in the record to justify any mandatory action. The CPSC staff continued to work with the industry on voluntary measures to reduce risk.
But by this time, ATV accidents were making the evening news, and the public was reaching its own conclusions. In May 1985, a House subcommittee held hearings and voted to recommend a ban on some types of ATVs. Though not binding on the commission, the subcommittee vote increased the public pressure for action.
The CPSC created a special internal task force to evaluate the evidence, and received its recommendations on September 30, 1986. The task force recommended a range of actions, but found no imminent hazard, and recommended no ban, recall, or refund program for three-wheeled ATVs. Nevertheless, 10 weeks later the commission decided (in a two to one vote) to ask the Justice Department to bring a suit under Section 12 of the Consumer Product Safety Act. Section 12 is an emergency provision allowing the commission to seek relief directly in federal district court, bypassing rulemaking, adjudication, and any other administrative steps that ordinarily accompany a regulatory action.
The government's complaint requested an injunction halting sales of three-wheeled ATVs and requiring the repurchase of those held by dealers, a recall and refund for all three-wheeled ATVs and for all adult-sized four-wheeled ATVs used by children under 16 years, and age restrictions for the various ATV models. In addition, it requested a free, hands-on training course "to teach adequately the skills and knowledge necessary to ride an ATV safely," an effective public awareness campaign, labels on new models and made available for models already sold stressing "the complexities and hazards associated with ATV riding," and a toll-free ATV-safety hotline. The defendants were the five distributors of ATVs in the United States: American Honda Company, Yamaha Motor Corporation, US Suzuki Motor Corporation, Kawasaki Motor Corporation, and Polaris Industries. The Japanese manufacturers of ATVs were also named.
The complaint clearly abandoned the CPSC'S cooperative approach of working with the distributors, and strongly argued the case for mandatory action. The essence of the complaint, which is reproduced in part in the box on the previous page, was that the risk of serious injury or death was unreasonably high, that the skill required to avoid injury was unreasonably great, and that manufacturers and distributors had failed to implement steps to reduce risks, despite the relatively low costs of doing so.
A benefit-cost estimate by the CPSC staff supported some of the information-providing remedies sought in the complaint. Indeed, these were the sort of actions for which the staff had been negotiating with the distributors. And by this time, the proposed ban on future sales of three-wheeled ATVs would simply ratify a judgment the marketplace appeared to have made already: sales had shifted from more than 80 percent three-wheeled ATVs in the early 1980s to more than 80 percent four-wheeled ATVs. The provision that terrified the distributors was the requirement that they offer refunds for the roughly two million three-wheeled ATVs already sold.
The CPSC based its case primarily on data compiled through its National Electronic Injury Surveillance System, which tabulates product-associated injuries from approximately 65 hospital emergency rooms across the United States. After analyzing the NEISS data over the period 1981-1986, the CPSC staff concluded that ATV-related injuries were increasing at an annual rate of over 60 percent, that ATVs were unusually hazardous compared to dirt bikes and snowmobiles, and that three-wheeled ATVs were especially hazardous because they were more unstable than four-wheeled ATVs.
Despite strongly conflicting views on the safety of ATVs, the government and then manufacturers signed a preliminary consent agreement in December 1987, and a final settlement shortly thereafter. The terms of the final consent decree included most of the measures that the government had requested, including a halt in the sale of three-wheeled models. But the agreement provided for no recalls, refunds, or exchanges.
The United States of America...brings this action for declaratory and injunctive relief to protect the public form the imminent and unreasonable risk of death and severe personal injury presented by motorized vehicles known as "all-terrain vehicles," or "ATVs."
Each day millions of individuals, a large number of them children under the age of 16, are unwittingly exposed to the risk that, as a result of their operation of ATVs, they will be involved in an accident in which they will either die or suffer a severe personal injury such as quadriplegia, paraplegia, a ruptured organ, or a skull or bone fracture.
Far from being safe, easy-to-ride vehicles for harmless play, as defendants have falsely and deceptively represented them to be, ATVs actually are unique and complex vehicles, requiring for their successful operation constant and precise rider manipulation which is neither instinctive nor easily matered by a person of ordinary skill. Moreover, in contrast to the operation of other motor vehicles and unbeknownst to ATV riders, there is virtually no margin for error in the operation of ATVs because of their peculiar operating characteristics, and the penalty for making the smallest miscalculation may be death or a severe injury.
The precautions necessary to abate the grave hazard posed by ATVs...are relatively inexpensive, particularly when compared to the high cost in human suffering caused by ATVs. Nevertheless, the manufacturers and distributors of ATVs have failed to implement these precautions.
The United States brings this action to gain the immediate relief necessary to protect the public form the unreasonable and imminent peril in which the defendants have placed, and continue to place, the millions of innocent, inadequately informed individuals who presently operate, or will in the future operate, ATVs.
Complaint, United States of America V. American Honda Motor Company, Inc., et al.
Why did the government settle the case without a mandatory recall and refund program? The official reason is that the settlement saved the public the cost of litigation, and it obtained most of the remedies the government was seeking, without the years of delay that litigation would entail. There have been allegations that the settlement was politically motivated, just as there were allegations that the bringing of the suit in the first place was politically motivated. Yet another possibility is that the government simply could not prove its case.
The data base used by the CPSC is fraught with problems. The NEISS sample of injuries is small, biased in a variety of ways, and tainted with misreporting. In their rebuttal risk-analysis, the ATV distributors argued that the sample used to evaluate ATVs included only 288 injuries (of an alleged universe of 300,000); that it was concentrated in the Northeast, and did not adequately represent the different terrains and riding conditions of the West and South; that it evaluated injuries only in the spring and summer, excluding injuries from fall or winter terrains; and that it failed to include injuries that resulted in death or that bypassed emergency rooms. They also investigated the NEISS data base and found extensive coding errors, typically the misclassification of the type of vehicle involved or the cause of the accident.
The CPSC's raw data on injuries showed a 70 percent annual growth rate in ATV-related injuries during the sample period-certainly an alarming trend. But after correcting for identifiable coding errors, the injury growth rate dropped to 46 percent annually. After correcting for the growth in the number of vehicles in use, it dropped to 11 percent, which was not statistically different from zero. Of course, an 11 percent annual growth rate in the per-vehicle injury rate might still be worrisome if it were accurate and likely to persist. But the distributors argued it was not. They pointed out that the injury rate associated with ATVs was following the same kind of time-profile as injuries associated with snowmobiles or with dirt bikes when those products were new. Each of them displayed a high injury rate in the early years, dropping off to a lower level as the market matured.
This "experience curve" effect can be attributed to any number of factors: the experience of the manufacturers in producing safe vehicles, the experience of the riders, the emergence of riders' clubs, or the development of special trails and riding areas that are relatively safe. In any event, the time trends in ATV injuries appeared routine. Moreover, the injury rate per hour of ATV use did not appear different from the rate for other vehicles: 0.120 per thousand hours of use, compared to 0.116 for mini/trail bikes and 0.115 for snowmobiles.
The distributors acknowledged that three-wheeled ATVs were disproportionately represented in the CPSC injury sample: 82 percent of ATVs have three wheels, and these were associated with 86 percent of the injuries in the sample. The CPSC attributed this to an inherent instability in the three-wheel design. But the distributors pointed out that the higher injury rate was not significant and was observed only for relatively minor injuries; the rate of serious injury or death was the same for three- and four-wheeled ATVs. ("Imminent hazard" is defined by the statute only in terms of serious injury and death.) Moreover, the conclusion that instability was a contributing factor relied on a statistical analysis that included accidents in which the ATVs tipped or overturned-whether or not the rider's injury was caused by the tip or overturn. For example, a rider might run into a car, injuring himself and also causing the ATV to overturn. Injuries that actually resulted from a tip or overturn showed the opposite bias: they accounted for 48 percent of injuries associated with three-wheeled ATVs and 52 percent of those associated with four-wheeled ATVs. Therefore, while three-wheeled ATVs appear somewhat more hazardous, this does not seem to be related to instability. It could simply be because three-wheeled vehicles are disproportionately ridden by young riders (16 to 24 years old), or because they are disproportionately used for stunt riding and racing. Substituting four-wheeled vehicles for three-wheeled vehicles among these riders and for these uses may have very little effect on the injury rate.
None of this will bring back the three-wheeled ATV. The "imminent hazard" label has stuck, and if the settlement agreement or shifting consumer choices had not taken them off the market, undoubtedly product liability suits soon would have. The CPSC's foray against ATVs calls into question the popular notion that federal preemption of state product liability laws will make those laws substantially more rational and predictable.
Subscribe to Regulation