from Ethics & the Environment Volume 7, Number 1RISKY BUSINESS Nuclear Workers, Ethics, and the Market-Efficiency Argument*
Kristin Shrader-Frechette
Permission to CopyYou may download, save, or print for your personal use without permission. If you wish to disseminate the electronic article, or to produce multiple copies for classroom or educational use, please request permission from:
Copyright Clearance CenterFor other permissions or reprint use contact:
Professional Relations Department
222 Rosewood Drive
Danvers MA 01923 FAX: 978-750-4470/4744
Web address: www.copyright.com
Rights and Permissions, Journals Division
Indiana University Press
601 North Morton St.
Bloomington, IN 47404FAX: 812 855-8507
E-mail: journals@indiana.edu
*Work on this article was completed, thanks to a research grant from the National Science Foundation, Ethics and Values Program, grant SE-98-10611, "Nuclear Technology and the Ethics of Worker Radiation Risks." Any errors or omissions are the responsibility of the author and not the NSF.
ABSTRACT Workers generally face higher levels of pollution and risk in their workplace than members of the public. Economists justify the double standard (for workplace versus public exposures to various pollutants) on the grounds of the compensating wage differential (CWD). The CWD, or hazard-pay premium, is the increment in wages, all things being equal, that workers in hazardous environments receive, as compared to other workers. Economists defend the CWD by asserting that workers willingly trade safety for extra money. This essay (1) examines the theory behind the CWD, (2) presents and evaluates economists' Market-Efficiency Argument for the CWD, (3) offers several reasons for questioning the CWD, and (4) applies the Market-Efficiency Argument to a real-world case, that of U.S. nuclear workers. The essay concludes that this argument fails to justify the CWD, at least in the case of U.S. nuclear workers.In May 2000, a federal judge sentenced a corporate executive to seventeen years in prison and ordered him to pay $6 million to the family of one of his workers who was permanently brain damaged from cyanide poisoning on the job. The sentence was the largest ever imposed anywhere for an environmental crime. In May 1999, a Pocatello, Idaho, jury had found the executive, Allan Elias, guilty of knowingly endangering the lives of his employees at Evergreen Resources, a company that made fertilizer from vanadium mining waste. Elias did nothing to protect his workers, said the court, despite the fact that employees repeatedly complained of sore throats, said they needed protective gear to clean company tanks, and argued that the tanks needed to be tested for toxic chemicals. Overcome by hydrogen cyanide gas while cleaning a tank, twenty-year-old Scott Dominguez was not rescued from the tank for more than an hour. Because the company did not have the proper equipment, no one was able to help the fallen worker (Sissell 2000). Although the court award and the prison sentence are large, the fact of workers dying or being seriously injured on the job, because of an unsafe occupational environment, is nothing new. Annually in the U.S., 7,000–11,000 people die prematurely from injuries sustained in the workplace and another 62,000–86,000 people die prematurely from occupationally-induced diseases. This means that a total of nearly 100,000 workers die needlessly each year in the U.S. from unsafe work environments, even though their deaths could have been prevented. These victims of environmental injustice represent a largely silent minority, not only because their number represents less than one one-thousandth of the U.S. work force but also because their deaths frequently have undetected causes for which it is difficult to hold employers responsible. Also, there are only 2,700 practicing occupational-medicine physicians in the U.S. and only a handful of Occupational Safety and Health Administration (OSHA) inspectors, enough to check a worksite for safety once every seventy-five years. And although from 1980–88 OSHA had referred thirty cases of job-related, criminal homicide to the U.S. Justice Department, only four had been prosecuted or were being prosecuted by 1989, in part because the Bush Administration cut the funding of the Justice Department section responsible for prosecuting companies for workplace deaths (Leigh 1995, 3–7, 215).
How can society ignore victims of the workplace environment when annual occupation-related deaths in the U.S. are approximately five times greater than those caused by the illegal drug trade and approximately four times greater than those caused by AIDS? One reason may be that few people are aware of the alarming occupational-fatality data, and almost no epidemiological studies track worker deaths over the long term. Another may be that most casualties of the workplace environment are poor, African-American, or Hispanic, and they have few advocates (Leigh 1995, 3–7, 215). Also, although unhealthy workplace environments annually cause three times more deaths and injuries than street crime (Gersuny 1981, xi; NIOSH 1999), even in developed nations employers often have so much power, privilege, and status that they can avoid responsibility for what happens to employees. In developing countries, apparent injustice in the workplace environment is even more evident. Worldwide, workplace risks are increasing, in part because of the World Trade Organization (WTO), established in 1995 as part of the Uruguay Round Agreements of the General Agreement on Tariffs and Trade (GATT). The WTO has defined all worker protections, including prohibitions against child labor, as "barriers to trade" that violate the WTO international regulations by which all member nations must abide. Because of the WTO and GATT, workers throughout the world are facing environmental injustices, including the inability to control the health risks that employers impose on them (Wallach and Sforza 1999, chaps. 6–7).
Still another reason society minimizes the massive number of occupation-related fatalities is that economists tend to justify risky workplaces on the grounds of the compensating wage differential (CWD). The CWD, or hazard-pay premium, is the alleged increment in wages, all things being equal, that workers in risky jobs receive. According to this theory, employees trade safety for money on the job market, and they know some of the workers will bear the health consequences of their employment in a risky occupational environment. To determine whether the CWD or hazard-pay premium succeeds in justifying alleged environmental injustices in the workplace, this essay addresses one of the main arguments used to support the CWD and then applies it to the 600,000 U.S. workers who have labored in government nuclear-weapons plants and laboratories.
WORKPLACE VERSUS PUBLIC STANDARDS
As a consequence of workplace risks some policy experts argue that there should be no double standard, one for occupational, and another for public, exposure to various gases, chemicals, particulates, radiation, noise, and other forms of environmental pollution. They believe that, unless industrial employees are protected by health and safety standards that are equal to those protecting the public, then workers will face environmental injustice. According to critics of this double standard, employees ought not to have to trade their health and well-being for higher wages. Moreover, say critics of the CWD, paying people to put themselves at risk at work is not significantly different from murder for hire. Those who agree with the double standard for worker and public exposure to environmental risk usually maintain that the compensating wage differential (CWD), the additional pay received by employees in hazardous occupations, compensates them for their increased risks. They also claim that workplace risk is overemphasized and sensationalized by "the danger establishment" (Douglas and Wildavsky 1982, 9). They say most countries, notably the U.S., have unacceptably "rigid standards" for workplace risks. For example, for those who believe that occupational safety requirements are too strict, a recurrent target of ridicule is the portable toilet standard for cowboys that the U.S. Occupational Safety and Health Administration (OSHA) employs (Viscusi 1983, 114–15, 136).
Controversy over workplace risks originated at least as early as the emergence of a division of labor between manual and nonmanual work. Ancient Greek and Roman writings are filled with references to the diseases peculiar to one or another profession. Later, during the Renaissance, miners and metal workers became the first subjects of medical research into diseases of the workplace. Perhaps the first publication to address occupational hazards and their prevention was a booklet written in Germany in 1472. It told goldsmiths how to avoid poisoning by mercury and lead. In 1556, in his treatise on the mining industry, the German mineralogist, Agricola, wrote the first known review of miners' health problems. Besieging his medical colleagues and statesmen to make workplaces safer, in 1700 the Italian physician, Ramazzini, wrote Diseases of Workers (Eckholm 1977, 31–32).
Despite the historical knowledge that various diseases are associated with particular jobs, governments have done surprisingly little to avoid or to reduce many known occupational risks. As J. K. Wagoner of the U.S. National Institute for Occupational Safety and Health (NIOSH) observes, two centuries have passed since Percival Pott linked coal tars to the scrotum cancer that killed young chimney sweeps in England. Yet, "thousands of coke-oven workers in steel mills around the world continue to inhale the same deadly substances, and they are dying of lung cancer at ten times the rate of other steel workers" (Eckholm 1977, 32).
One reason for the continuing controversy over workplace hazards, and over whether to employ a double standard for public and occupational risk exposures, is that some nations do not appear to protect public health and safety more rigorously than worker health and safety. Either they have the same standards for occupational and public environmental exposures, or they treat workers the same as other citizens when they compensate them for accidents or injuries. For example, in 1972 New Zealand passed a universal, state-run scheme to compensate all victims of accidents, workers and nonworkers, the same. Another reason for controversy over the safety of the occupational environment is that U.S. standards for health in the workplace appear to permit greater risks than do those of many other nations. In terms of wage differentials for fatal-injury risk, for example, Australian workers appear to enjoy a wage increment that is nearly triple the U.S. increment for risky work (Kniesner and Leeth 1991, 75–90). And in terms of permissible levels of chemicals in the work environment, U.S. regulations are less strict than those of countries such as Germany, Sweden, and the Czech Republic. Standards in Argentina, Great Britain, Norway, and Peru are approximately the same as those in the United States. In Sweden and Germany, for example, unlike the U.S., workers have more extensive rights to be informed about hazards and to take steps to reduce exposures. Strikes there are rare, and labor productivity rates are among the highest in the world, while maximum-allowable-‚concentration values (MACs) are among the lowest in the world. The United States, however, has not adopted the approach of Sweden and Germany (Kates 1978, 46–47; Robinson 1991, 74; Dwyer 1991, 250).
THE COMPENSATING WAGE DIFFERENTIAL
Unlike the U.S., the former U.S.S.R. had a long tradition of providing for occupational justice. In 1923, the U.S.S.R. founded the first hospital devoted entirely to the study and treatment of occupational diseases. No such hospital exists in the U.S. Of course, Soviet enforcement patterns are not known and, although maximum-allowable-concentration (MAC) values may have been lower in the U.S.S.R. and in the new Soviet republics, such as Belarus and Ukraine, control there likely is far less stringent than in western countries. If so, then despite safer environmental standards in these nations, workplace risks could be higher. Regardless of whose enforcement patterns are better, however, comparisons between countries (such as the United States and Germany) raise a number of interesting philosophical questions. Among these are when a workplace environment is so dangerous that it is unjust. Do the Germans have a more or less desirable risk philosophy than their American counterparts? Why do German MAC values tend to be lower, often by a factor of 10 or more, than corresponding U.S. standards, even though Germany must confront many of the same problems that the U.S. faces (Berman 1978, 192–93; Kates 1978, 168–74)? Yet another question is whether one can ethically justify workplace MAC values, which are sometimes higher than corresponding values for public exposure, on the grounds that workplace exposure time is shorter than that for the public. If lower MAC values are possible, would they be so costly as to jeopardize the economic well-being and the technological progress which have resulted in enormous improvements in human welfare? Would they be so costly that most workers and citizens would not be willing to pay for them by raising the price of goods and services produced in risky ways?
Although many factors are likely responsible for the more lenient occupational safety standards in the U.S., as compared to those in other countries, at least one of the reasons for the disparity is a surprising lower emphasis on equity or environmental justice in the U.S. where standards typically allow much higher pollution-exposure levels for workers than for the public. In large part, this is because U.S. policymakers do not believe that equity requires occupational and public exposure levels to be the same, given that workers allegedly receive higher pay because of their higher exposures. For example, the U.S. maximum permissible dose of whole-body ionizing radiation which can be received annually by the public is 100 millirems. The maximum permissible dose, for the same time period, for industrial workers is 2,000 millirems per year, averaged over five years, with a maximum of 5,000 mrem for any given year. Thus a nuclear worker could legally receive 50 times as much radiation as a member of the public, in a given year (ICRP 1991). This double standard is even more troubling when one realizes that, before 1990, the public standard was 10 times stricter than the worker standard for ionizing radiation. After 1990, the public standard became 50 times stricter for a given year. These numbers reveal that, while the government is doing a better job of protecting the majority, members of the public, it may not be doing the same for workers, especially since there is no safe level of ionizing radiation. Indeed, since 1990 worker protection from ionizing radiation has been getting worse, not better.
The main reason U.S. policymakers do not believe that equity or environmental justice demands the same standard, for occupational and public exposure to various pollutants, is that they do not believe the two types of exposures are analogous1 (Shrader-Frechette 1985, chap. 2; Starr 1969, 1232–3; Rescher 1983, 172). They claim that environmental risks accepted "voluntarily," through one's occupation, can be regulated by means of standards less strict than those applied to public risks, precisely because people are compensated (through their wages) for the higher workplace risks. On the one hand, Chauncey Starr, one of the preeminent proponents of the CWD, claims that empirical data show that the risk entailed by a particular occupation is directly proportional to the cube of the wages for that occupation. He argues that, as the risk increases, so do the wages (Starr 1976, 16; Berman 1978, 192–3; Kates 1978, 168–74; Siebert and Wei 1998, 171–81). On the other hand, opponents of the CWD say the wage-risk relationship is not so simple, especially in western countries. They claim many factors, in addition to risk, determine the wages people accept for given work. Some of these factors include the degree of education or training necessary for the job; the extent to which people are available to perform the work; the physical strength required to do the task; or the lack of other employment opportunities. Hence, although there is some sort of wage-risk relationship, such that wages often rise as job risks increase, they say this relationship may not be nearly so simple as Starr supposes. In fact, they note that different economists actually calculate different CWDs, different increments of pay per risk increment (Craig 1995, 304–20; Gale 1998, 13–26).
Starr's view, widely accepted among risk assessors, is part of the classic theory of the compensating wage differential (CWD), as Adam Smith formulated it. According to Smith, "the whole of the advantages and disadvantages of the different employments of labor" continually tend toward equality because the wages vary according to the hardship of occupation. On Smith's theory, people exposed to a risky workplace had advantages and disadvantages whose sum was equal to that for people not exposed to such risks, because those in the high-risk occupations were provided with higher rates of pay than those in low-risk jobs. According to proponents of the CWD, a double standard for worker and public risk is acceptable because those in high-risk jobs voluntarily agree to "trade" some degree of workplace safety for higher wages. In other words, the classic market solution to the problem of how to control occupational risks, and how to decide which worker risks are acceptable, is to use an "economic fix" for setting standards (Viscusi 1983, 33ff., 156–68; Dobbs 1999, 103–9; MacLean 1981, 6–9). According to Smith, employers using dangerous technologies will lack employees unless they raise wages or offer some other inducement to attract workers. These hazard-pay premiums or CWDs thus partially compensate workers for the expected economic costs of their later work-related injury or illness. Smith's theory also suggests that the necessity for firms with risky jobs to pay higher wages also gives them incentives to invest in safety and health precautions. According to the theory, they can recover these investments in the form of lower CWDs, or hazard-pay premiums. Smith's theory thus predicts that workers will be aware of many of the hazards to which they are exposed, that quit rates will be higher in hazardous jobs than in safe jobs, and that risky occupations will pay higher wages than safe occupations.
Smith's theory of the CWD falls short on several counts. For one thing, dangerous jobs typically are not filled by rational agents who are well-informed of the risks. Workers who have little formal education and who have difficulty recognizing subtle hazards often have risky jobs. This fact makes it important to note that at least two assumptions underlie Adam Smith's theory of compensating differentials. First, workers must be aware of the hazards they face. Second, they must have a number of meaningful job possibilities. Both of these assumptions often are at variance with the facts in the real world. The number of realistic job options enjoyed by different workers varies widely depending on their skills and social status. To the extent that hazardous occupations are filled with less skilled and socially disadvantaged workers, Smith's theory requires that such jobs will offer meager CWDs or hazard-pay premiums (Robinson 1991, 5, 77).
To defend their claim that the CWD compensates for problems of alleged environmental injustice raised by the double standard for occupational and public risk, economists and public policymakers often employ (what I call) "the Market Efficiency Argument." This essay examines and evaluates it in order to determine whether the argument succeeds in justifying apparent environmental injustice.
THE MARKET-EFFICIENCY ARGUMENT
The Market-Efficiency Argument for accepting the CWD is that market allocations of individuals to jobs will promote efficient matchups in many instances. If the worker bears all of the harm associated with the risk and if he is cognizant of his own particular risk, not simply the average risk for all, he will select his job optimally . . . workers are not in jobs at random and the market promotes the most efficient matchups. (Viscusi 1983, 132, 135; Dorman 1996, 28ff)
For example, says Viscusi, "African-Americans with the gene for sickle-cell anemia may incur a greater risk of harm from the low-oxygen conditions faced by a pilot, and female mail sorters have a greater frequency of back injuries when moving the standard seventy-pound mail sacks." If these minorities and women have accurate knowledge of the greater risks they face in particular circumstances, Viscusi says they will use the market mechanism in an efficient way so as to select the job for which they are the most suited (Viscusi 1983, 132–33). Or, as Dorman (1996, 28–31) puts it, occupational safety is a commodity traded on a market, and people can buy what they want of it.
As is probably evident, the market-efficiency argument requires one to assume that employees' preferences will operate so as to attain authentic worker welfare. That is, it contains the implicit assumption that market-based preferences are accurate indicators of legitimate values. This assumption, however, is not generally true. If it were, there would never be grounds for government intervention in markets, for example, to protect potential victims or to set minimum standards for workplace conditions. Likewise, if this assumption were true, then one would have to condone the sweatshop conditions of a century ago. One would have to agree that twelve-hour workdays of a bygone era were ethically desirable, because they allowed workers to choose an "efficient matchup." On the contrary, the efficiency and the optimality of worker choices, whether among anemia-prone African-Americans or backache-prone women, is in part a function of the choices available to workers. If an economy is not diversified, and if employees have no real occupational alternatives in the face of the need to feed their families, then it hardly can be said, as Viscusi and others do, that the "market . . . will promote efficient matchups." As Elizabeth Anderson puts it, the CWD reveals neither the value of life nor how the market efficiently distributes occupational safety; instead the CWD reveals only the risks people are obliged to take, in order to discharge their responsibilities (Anderson 1997, 159; Levenstein and Wooding 1997).
The market-efficiency argument for the CWD also is questionable in that the ethical conditions necessary for desirable market transactions frequently are not met in real life. Recall that economists admit that information is necessary for the market to be efficient. As Viscusi put it earlier (emphasis mine): "If the worker bears all of the hazard associated with the risk and if he is cognizant of his own particular risk, not simply the average risk for all, he will select his job optimally" with respect to his own risk potential and personal advantages and disadvantages. This means that, even on the terms of CWD proponents, the validity of the market-efficiency argument is premised on workers having adequate knowledge of their particular risk situations.
But are people generally aware of the hazards they face? Most environmentalists and risk assessors probably would say they are not. Starr and Whipple (1980, 1115–17), as well as Fischhoff, Slovic, Lichtenstein (1980, 192, 202, 208), and other risk assessors (Starr 1976, 16; Berman 1978, 192–3; Kates 1978, 168–74; Siebert and Wei 1998, 171–81), repeatedly have pointed out that intuitive or subjective estimates of risks made by educated lay people are quite divergent from analytical, allegedly objective, assessments of risks made by experts. Lay people typically overestimate low-probability risks and underestimate higher probability ones. For example, they overestimate catastrophic chemical risks, but underestimate risks associated with automobile accidents (Dorman 1996, 42ff). Economists also realize that the public's risk estimates sometimes err. To bridge the gap between the theoretical model of rational choice and imperfect, real-life choice, economists almost always include the costs of searching for risk information as part of their equations dealing with choice under uncertainty. If these economists and risk assessors are correct, then the conditions necessary for ethical use of the Market-Efficiency Argument (full information and the ability to pay to obtain it) often may not be met in real life. But if these conditions are not satisfied, then the argument does not provide convincing grounds for supporting the CWD and for claiming that it offsets apparent injustices in the workplace environment.
OTHER ARGUMENTS AGAINST THE CWD
In addition to objections against the Market-Efficiency Argument, several other reasons also suggest that the CWD may not succeed in justifying risky workplace environments. These reasons include the facts that the differential may not exist, that acceptance of CWD risks may impose them involuntarily on others, and that the CWD acceptance may rely on inconsistent reasoning. One of the most basic reasons for doubting that the CWD provides an ethical justification for risky workplace environments is that the CWD may not exist. Some researchers have shown that, when all workers are lumped together from lowest to highest paid, then risk and salary increase proportionately, as the CWD theory predicts. However, when researchers separate the workers into two groups, with white, male, unionized, college-educated, or skilled workers in a primary group, and with nonwhite, female, nonunionized, non-college-educated, or nonskilled workers in a secondary group, the CWD theory falls apart. Primary-group workers enjoy a CWD, while those in the secondary group do not. Hence, the alleged CWD for the entire group, primary and secondary workers (Beck 1998), appears to be merely an artifact of data aggregation. In fact, the primary-group CWD actually may exacerbate unequal treatment of those in the secondary group (nonwhite, female, nonunionized, and so on) (Rescher 1983, 172; Craig 1995, 304–20; Gale 1998, 13–26), because it covers up the lack of CWD in the secondary group, once the data are aggregated.
Indeed, some economists have shown that, for nonunionized workers, there is a negative compensating wage differential; as risk increases, wages get lower. As discussion of the CWD already noted, to the degree that risky jobs are filled by less skilled or socially disadvantaged workers, even Adam Smith's theory suggests there may be no hazard-pay premium or CWD. In fact, when one compares wage rates across jobs, not adjusting for skill requirements, one observes that hazardous jobs pay twenty to thirty percent less than safe employments. The expedient way for employers to hold down wages thus is to hold down skill requirements. Social and economic inequality in society at large provides these employers with a supply of disadvantaged workers willing to accept health and safety risks in return for compensation. In fact, a pattern of hazards and low wages could not exist without a large supply of socially disadvantaged workers willing to accept both high hazards and low wages. This general association between hazards and wages across occupations suggests that unsafe jobs are generally lower-paying than safe ones. But if market competition is to generate wage premiums in hazardous jobs, this must occur in occupations where the workers themselves are aware of the risks of the job. If a job is hazardous, but the workers are unaware of this fact, there is no need for the employer to pay a CWD or wage premium in order to keep the employees on the job (Moore 1995, 57–70; Daniel and Sofer 1998; Dorman and Hagstrom 1998). And if such reasoning is correct, then at the very least, the CWD may not exist for all labor groups (Moore 1995, 57–70; Daniel and Sofer 1998; Dorman and Hagstrom 1998; Graham and Shakow 1981, 14–45; Rees and Shah 1992; Disney and Whitehouse 1990; Harmest and Wolfe 1990, S175-S197; Hirsch 1993, 279–301). If not, then the economic rationale for higher risks and apparent environmental injustice in the workplace cannot exist where the CWD does not exist. Thus, even if there is a genuine CWD for some workers—those already most privileged in society—the CWD may not provide a general ethical justification for workplace environmental injustice.
A second argument against the CWD is that, even if it is ethically acceptable for workers, it might not be so for others. That is, workers might not have the right to accept the CWD because of the attendant risks imposed on innocent people. Consider the case in which workers allegedly accept high occupational exposures to some carcinogen in exchange for a very high wage differential. The employees might be fully cognizant of the health hazards involved, and they might agree that the compensation afforded is adequate. Nevertheless, they might not have the right to take the risk, perhaps because their acceptance inevitably puts other persons, who have not agreed to accept the risk, in jeopardy. Workers who expose themselves to carcinogenic materials may be exposing their families to them via avenues such as shoes and work clothes. Because most carcinogens also are mutagens, they also may be exposing their potential children and their descendants to mutagenic hazards. Of course, one might argue that carcinogens on work clothes are minimal, or that unborn members of future generations have no rights to be protected from mutagenic risks.
While the issues of minimal risk levels and rights of future generations are too extensive to be discussed here (Shrader-Frechette 1991, chap. 3), one fact about the carcinogenic/mutagenic risk situation does seem clear. Provided they genuinely consent and are not being exploited, workers might have rights to take risks which threaten only themselves. It is less obvious that they have rights to take risks which might damage something, the gene pool, which is beyond themselves. Hence it is questionable whether any people intending to reproduce have rights to accept workplace risks which are mutagenic when those risks are higher than those to which the public is normally exposed. As Nicholas Rescher puts it: people ought only take risks for themselves, not for others: "morality enjoins conservatism" (Rescher 1983, 161). More strongly, the moral aspect of risk taking arises when the choices of individuals bear upon the interests of others (Rescher 1983, 162).
One does not have to move to future generations, of course, to discover innocent victims of workers' alleged rights to expose themselves to industrial toxins in exchange for higher wages. Some occupations, for example, that of air-traffic controller, produce high psychological risks, such as stress. It is questionable whether employees have rights to accept such high-stress risks when their effects are not borne merely by them but also by their families. Likewise, it is questionable whether particular workers, for example, in asbestos factories, have rights to accept higher workplace risks if such risks also affect their families. As already suggested, it is commonplace for family members of asbestos workers to contract cancer because they have been exposed to the fibers carried home on clothing. Some wives have died of asbestos-induced cancer merely because they washed their husbands' clothing. Close contact with their fathers has also caused the children of asbestos workers to contact cancer, and recent U.S. examinations have revealed dangerous levels of lead in the blood of lead workers' children, chiefly as a consequence of inhaling lead dusts brought home on clothes (Eckholm 1977, 30).
Where do the previous considerations leave us? Admittedly, some workers might be forced to accept risky jobs to support their families. And, admittedly, some of the hazards faced by the families of those in high-risk occupations could be eliminated or reduced by simple practices such as workers' bathing and discarding their work clothes before coming home. Nevertheless, to the extent that employees' acceptance of hazards thereby places a higher health risk on those other than themselves, then to that same degree is their right to take such risks questionable.
A third argument that the CWD does not justify a workplace environment with higher risks than a public environment is that its proponents often defend their stance by making inconsistent appeals to workers' risk perceptions. When Starr and other proponents of the CWD wish to justify workers' acceptance of higher risks in return for higher wages, they take an interesting stance. They maintain that, once employees are adequately educated regarding the risks they face, regulations ought to follow employees' risk preferences. They also say that regulators have no right to tell workers they cannot follow their preferences for higher risks (Viscusi 1983, 77, 80, 83; Starr 1969, 1233–4; Starr 1976, 15–21; Dorman 1996, 28). However, when these same proponents of the CWD wish to justify government imposition of particular standards for public risk, in the face of citizens' demands for stricter regulations, they take a different stance. They maintain that risk preferences, even of highly educated laymen, are subjective, intuitive, and generally erroneous. Therefore, they say that regulators ought not merely to follow the public's demands for lower risks. Instead, they claim, regulators also ought to adhere to the risk assessments calculated by experts because these reflect "rational" preferences, and regulators ought to implement them in standards for public exposure (Starr and Whipple 1980, 1115–19).
For example, speaking of the public's "irrational" aversion to low-probability, high-consequence nuclear accidents, Starr and Whipple (1980, 1115–17) maintain that lay (nonexpert) perceptions regarding this technology are incorrect. They say public demands for greater nuclear safety are not reasonable, since they fly in the face of experts' beliefs about acceptable levels of nuclear risk (Starr and Whipple 1980, 1115–17). Psychometric surveys of attitudes about risk, however, reveal that there is no significant difference, in level of relevant technical knowledge, between those members of the public who favor greater, and those experts who favor less, safety (Fischhoff et al. 1980, 150; Shrader-Frechette 1991; Moore 1995, 57–70; Daniel and Sofer 1998; Dorman and Hagstrom 1998). And if not, then there may be no grounds for rejecting risk-averse attitudes of either workers or the public, contrary to what CWD proponents suggest.
As the previous paragraph showed, CWD advocates reject lay risk attitudes, when they demand more safety. Yet the same CWD advocates accept lay risk attitudes when they accept less safety in the workplace. Proponents of the CWD, who claim workers voluntarily accept risky jobs, often advocate uncritical acceptance of worker perceptions of low risk. They do so in order to justify less stringent occupational standards and are thus in an apparently contradictory position. They contradict themselves when they accept workers' lay risk perceptions of low risks but reject acceptance of lay risk perceptions when lay people want more safety. They cannot have it both ways. They ought not accept lay worker risk perceptions, when they suit their laissez faire economic mentality, but reject lay public risk perceptions, when they do not. Either government standards ought to be based on lay and stakeholder (worker) risk perceptions, or they ought not to be based on these risk perceptions. If risk assessors claim that relevantly educated people err in their risk perceptions and ought to be "corrected" by experts, then both lay workers and the lay public ought to be so corrected, and not just the public. And if workers' incorrect risk perceptions or inadequate information are corrected, it is less likely that they can be said genuinely to have consented to the CWD and higher occupational risks.
Proponents of the CWD are likely to respond to this inconsistency charge by saying that the cases of worker and public perceptions are not analogous. They might object that workers voluntarily accept given modes of employment, specific CWDs, and risks, whereas the public receives none of these. Because of the alleged consent and compensation involved in the worker case, the objectors could argue that workers' preferences ought to be followed, whereas the environmental risk preferences of the public need not be followed because there is no compensation and contractual ‚consent.
As this objection correctly notes, the cases of workers' perceptions and public perceptions are disanalogous with respect to consent and compensation. It does not follow, however, that these disanalogies are morally relevant in justifying inconsistent treatment of risk perceptions. Why not? If the fact that workers are compensated (for risks they choose) is reason to follow their risk preferences, as proponents of the CWD argue, then the fact that other people (members of the public) are not compensated for risks imposed on them is even greater reason to follow their preferences for lower risks. Virtually all risk assessors, including Starr and other advocates of the CWD, maintain that voluntary risks are more acceptable than risks of the same level which are involuntarily imposed (Starr 1969, 1233–4; Starr 1976, 26–30). But if so, then there is greater reason to follow public preferences for lowering risks to which citizens are involuntarily exposed than for following worker preferences. In other words, the very compensation-and-consent disanalogies between worker risk and public risk indicate that, if anything, there is more reason to follow public preferences for lower risks than to follow worker preferences for higher risks. This conclusion follows because the public is neither compensated for societal risks nor given a choice whether to accept them and because workers' acceptance of jobs often is not voluntary, owing to questionable background conditions. Yet this rationale is exactly the opposite of the view taken by most environmental risk assessors, such as Starr, who argue against following societal preferences for lower risks and in favor of following worker preferences for higher risks. And if so, then proponents of the CWD are on shaky ground when they reject public, but accept worker, risk preferences (Shrader-Frechette 1991). In short, if Starr, Whipple, and other risk assessors or economists are correct in rejecting public preferences about societal risks, then it is highly questionable for them to invoke worker preferences in order to use the CWD as a justification for riskier workplace environments.
APPLYING THE ARGUMENT TO NUCLEAR WORKERS
What happens to the CWD rationale for apparent environmental injustice, if one examines an empirical case, that of nuclear workers exposed to high levels of ionizing radiation as a result of employment in U.S. Department of Energy (DOE) nuclear facilities? The 600,000 current and former workers represent an ideal case study for applying the arguments given earlier in the essay. For one thing, one would expect these nuclear workers to be treated better than most such labor groups throughout the world because they are or have been employees of the U.S. government or its contractors and subcontractors. As employees of the richest government in the world, these workers in theory ought to receive excellent treatment, in part because the DOE runs a multi-billion dollar operation in the U.S. Another reason DOE workers represent an excellent case study is that they are such a large group. As a result, in theory it should be possible to get statistically robust conclusions about the wages and the risks to which the 600,000 employees are and have been exposed. Still another reason the group is an excellent one to study is that roughly one-sixth of them is unionized and belongs to the Paper, Allied-Industrial, Chemical, and Energy Workers Union (Miller 1999, 57, 59). This percentage of DOE union members is roughly the same as the percentage of all U.S. workers who are unionized, so they may constitute a fairly representative group, at least relative to unionization.
The U.S. DOE has 3,500 nuclear facilities at 34 sites in 13 states of the U.S. Eighty percent of these facilities are defense-related, and the remainder do commercial or laboratory work. Twenty-three of the U.S. DOE facilities are national laboratories, such as Los Alamos National Laboratory (LANL), Lawrence Livermore National Laboratory (LLNL), Idaho National Engineering and Environmental Laboratory (INEEL), and Sandia National Laboratory (SNL) (GAO 1998, 4).
Does the CWD of workers at these and other DOE facilities justify the higher radiological risks (and potential environmental injustices) they have faced in their occupation? One way to answer this question is to evaluate the Market-Efficiency Argument for the CWD relative to this DOE case. Are there empirical factors in this particular case that suggest the argument is correct or incorrect? For the market-efficiency argument to succeed in justifying riskier occupational environments, its proponents admit that workers must be cognizant of their own individual risks, in order to make economically efficient choices about employment and compensation. But are DOE employees cognizant of their risks because of the operation of market mechanisms? Does the DOE use its information about safety to penalize unsafe contractors financially and reward safe contractors so that safety information is available to workers? The answer appears to be "no."
In exchange for nuclear-liability protection of up to $9.43 billion for DOE contractors and sub-contractors (Jones 1999, 55), the U.S. Congress asked the DOE to assess penalties for safety violations at DOE sites (GAO 1999, 4). Yet, when one examines these penalties it is clear neither that they are responsive to information about site-safety compliance nor that the penalties provide information to workers about how to make efficient market choices about trading pay for safety. The penalties are supposed to be effective because DOE contractors receive not only a fixed amount for running a facility, but they also can receive an annual "performance award," if they adhere to health and safety requirements. For example, for 1999, DOE contractor Lockheed-Martin received a guaranteed $3.5 billion for operating SNL, INEEL, and Oak Ridge National Laboratories. Its possible "performance award" for that year was $94 million, if it adhered to worker health and safety requirements. The possible Westinghouse "performance award" for 1999 was approximately $82 million, for Fluor Daniel $64 million, for Bechtel $57 million, and so on (GAO 1999, 28).
Despite government documentation of massive worker-safety problems at all the DOE facilities, nevertheless DOE typically awards from 89 to 100 percent of the full, annual "performance awards" to its contractors. LANL and LLNL, for example, both facilities with repeated, serious safety violations, have never had their performance awards reduced because of their poor safety record. In fact, the DOE proposed in March 1999 to exempt them, in the future, from any penalties (for safety violations) altogether (Miller 1999, 109–11; Van Ness 1999, 103; Jones 1999, 8).
LANL, in particular, experienced many fires and explosions. LANL alone received approximately 94 percent of all DOE laboratory safety penalties from 1992 through 1999. Yet DOE never shut LANL down, and it forgave all its penalties, making them what the GAO calls "phantom penalties." Similarly, in May 1996, DOE identified a host of "multiple and recurring failures to follow criticality safety procedures" at LLNL and problems with contamination of five workers. Yet one year later, DOE noted that many of the same criticality problems were still occurring, including a loss of control of plutonium. Despite these facts, the next year the GAO reported that DOE gave LLNL a health and safety rating of "good," and it awarded LLNL 96 percent of its optional performance fees. In fact, although LLNL receives $1.1 billion annually to operate the facility, its DOE contact specifies that, regardless of its safety record, DOE can never reduce its additional "performance award" by more than 4 percent (Bliley 1999, 6; Jones 1999, 8, 13, 34; Upton 1999, 98–99). Despite the DOE's giving a safety rating of "good" to LANL and LLNL, in the face of massive safety deficiencies, the GAO claims that the real reason for the incorrect rating is that, if a DOE laboratory receives less than a "good" rating, two-thirds of its full contract amount is in jeopardy (Jones 1999, 39).
At the same time as these massive DOE performance awards are virtually guaranteed, regardless of contractor performance, GAO and Congressional oversight groups have revealed that DOE gives its contractors a mere slap on the wrist for serious safety violations. For example, for 1999, while Lockheed-Martin had an annual performance award of $94 million, its serious safety-violation penalties were minuscule by comparison. For instance, on February 27, 1997, the U.S. DOE criticized Lockheed-Martin for its safety violations at INEEL because it had no complete monitoring program and, as a result, five workers were seriously contaminated. Yet the penalty for these violations was only $25,000—approximately one one-hundredth of a percent of its annual performance award. Similarly, on September 21, 1998, DOE cited Oak Ridge National Labs for "failure on multiple occasions over a 2-year period, to identify significant intakes of radioactive material by 2 workers" and "failure to implement an internal dose evaluation program." Yet for these violations, it assessed the site contractor, MK Ferguson, no penalty whatsoever, even though the MK Ferguson annual performance award was approximately $47 million. Likewise, when SNL destroyed records that revealed unauthorized reactor operations, DOE nevertheless assessed no penalties and paid the site contractor millions of dollars in an annual performance award. And on July 16, 1996, DOE cited Westinghouse for failure to have adequate worker-safety monitoring equipment and therefore for causing the contamination of workers at the Hanford nuclear facility. Nevertheless, DOE assessed only a $37,500 penalty, while Westinghouse's annual performance award was $82 million. Its penalty was about a hundredth of one percent of its performance award. Likewise on October 7, 1996, the DOE noted that Kaiser-Hill, the contractor who operates the DOE Rocky Flats facility, had "repeated failures" to follow radiological work controls that caused worker exposures. DOE also noted that Kaiser-Hill later failed to report the exposures, then later failed to correct the problem. However, DOE fined the contractor only $37,500, a tiny fraction of Kaiser-Hill's annual $18 million performance award. Similarly, on June 5, 1997, DOE discovered that Mason and Hanger Corporation, which operates the DOE Pantex facility, had falsified worker safety records. Yet in response, DOE levied no penalty whatsoever, despite the Mason and Hanger annual performance award of $21 million. And on September 19, 1997, DOE cited Bechtel for inadequate workplace monitoring, for allowing workers to keep working after "stop work" radiation levels were exceeded, and therefore for exposing two workers to excessive radiation. But in response to these violations, again DOE levied no penalty, and instead it gave Bechtel an annual performance award of $57 million (GAO 1999, 28–36).
Such trivial (or nonexistent) DOE penalties for serious safety violations at DOE nuclear facilities are all the more amazing because they are inconsistent with the way the U.S. treats other threats to worker safety. Congressional oversight committees revealed that, while DOE repeatedly assessed no penalty or a $25,000 penalty for safety violations and contaminations over a two-year period, the Occupational Safety and Health Administration (OSHA) would have assessed $70,000 per day, for a total of $51,000,000 over two years, for each safety violation in the industrial facilities that it monitors (Dingell 1994, 8). Thus the DOE response to information about poor worker safety is extraordinarily more lenient than that of the main U.S. agency that monitors worker safety, OSHA. It is not surprising that, in the three-year period from 1996–1998, the GAO revealed that DOE mandated only $1.2 million in penalties for 3,500 U.S. nuclear facilities, while it paid some negligent contractors as much as $94 million each in an annual bonus, or performance award, even when their facilities were cited for safety violations (Jones 1999, 12). The same pattern of rewarding poor performers, and failing to make them accountable in terms of market information, was evident when the GAO revealed that the DOE Office of Enforcement identified more than 1,000 cases of nuclear safety noncompliance, but the DOE issued only 33 notices of violations and required a total of only $1.2 million in penalties for these violations (Upton 1999, 2). Safety information appears not to modify DOE's market behavior in an appropriate way, and DOE's market behavior, in turn, appears not to provide helpful bases for workers to make decisions about employment and pay. But if not, then this fact is contrary to what the market-efficiency argument for the CWD alleges. And if so, then the market-efficiency argument likely cannot be used to justify the apparent environmental injustices at risky DOE nuclear facilities.
CONCLUSION
Because of the factual situation accompanying U.S. nuclear workers' CWD, there are strong reasons to believe that the Market-Efficiency Argument does not justify the alleged injustices faced by these workers. If it turns out that there are plausible reasons, other than the CWD, for maintaining a double standard with respect to occupational and public risk, and for allowing alleged environmental injustice in the workplace, then those reasons need to be investigated. One place to begin such an ethical investigation might be to think of worker risk as analogous to patient risk. Although there is an ethical and legal requirement for informed consent on the part of patients being treated by a medical doctor, one of the limitations of the current CWD policy is that there are no comparable legal requirements for guaranteeing background conditions for informed consent in the workplace. Applying the medical-ethics analogy, one might well argue, for example, that just as people now claim that a doctor's withholding information from a patient is a violation of the medical doctor's fiduciary role and a way of undermining the patient's autonomy, an analogous point holds in the workplace. That is, an employer's withholding risk information from an employee is a violation of the employer's fiduciary role and a way of undermining the employee's autonomy. Withholding such information likewise jeopardizes the free information necessary to an efficient market. And if so, the ethical acceptability of the CWD is in question, at least in the case of U.S. nuclear workers.
Regardless of possible justifications for the current double standard for risk, one thing seems apparent. The CWD, as now implemented, does not adequately safeguard worker autonomy and well-being, for all the reasons spelled out earlier. Even the fact of nearly 100,000 workplace-induced premature fatalities suggests that the occupational environment, for high-risk jobs and for minority or poor workers, may be unjust. And if so, then society may need both to take steps to correct this situation and to reassess the CWD theory that often is assumed to justify it.
NOTE
1. The method of revealed preferences consists of examining actual risk levels faced in society, levels to which society allegedly gives implicit consent. The method of expressed preferences consists of using survey data to determine people's risk preferences.
REFERENCES
Ackerman, Frank. 2000. The Political Economy of Inequality. Washington, D.C.: Island Press.
Ackerman, Frank, David Kiron, Neva Goodwin, Jonathan Harris, Kevin Gallagher, eds. 1997. Human Well-Being and Economic Goals. Washington, D.C.: Island Press.
Anderson, Elizabeth. "Cost-Benefit Analysis, Safety, and Environmental Quality." In Ackerman et al, HWB, p. 159.
Beck, Matthias. 1998. "Dualism in the German Labor Market," American Journal of Economics and Sociology. July.
Berman, D. M. 1978. Death on the Job. London: Monthly Review Press.
Bliley, Thomas. 1999. "Statement," U.S. Congress, House of Representatives, Worker Safety at DOE Nuclear Facilities, Committee on Commerce, 106th Congress, Serial No. 106–43. Washington, D.C.: U.S. Government Printing Office.
Broome, John. 1999. Ethics Out of Economics. Cambridge; New York; and Melbourne: Cambridge University Press.
Craig, Lee A. 1995. "The Political Economy of Public Private Compensation Differentials," Journal of Economic History 55 (2): 304–20.
Daniel, Christophe, and Catherine Sofer. 1998. "Bargaining, Compensating Wage Differentials, and Dualism of the Labor Market," Journal of Labor Economics. July.
Dingell, John. 1999. "Statement," U.S. Congress, House of Representatives, Worker Safety at DOE Nuclear Facilities, Committee on Commerce, 106th Congress, Serial No. 106–43. Washington, D.C.: U.S. Government Printing Office.Disney, Richard, and Edward Whitehouse. 1990. Do Wage Differentials Compensate Occupational Pension Entitlements? University of Kent, UK: Institute for Fiscal Studies.
Dobbs, Ian M. 1999. "Compensating Wage Differentials," Economics Letters 63 (1):103–109.
Dorman, Peter. 1996. Markets and Mortality. New York: Cambridge University Press.
Dorman, Peter, and Paul Hagstrom. 1998. "Wage Compensation for Dangerous Work Revisited," Industrial and Labor Relations Review. October.
Douglas, M., and A. Wildavsky. 1982. Risk and Culture. Berkeley: University of California Press.
Dwyer, Tom. 1991. Life and Death at Work. New York: Plenum.
Eckholm, E. 1977. "Unhealthy Jobs," Environment 19 (6): 31–32.
Fischhoff, B., P. Slovic, and S. Lichtenstein. 1980. Societal Risk Assessment. Ed. R. Schwing and W. Albers. New York: Plenum.
Fischhoff, B., P. Slovic, S. Lichtenstein, S. Read, and B. Combs. 1978. "How Safe Is Safe Enough?" Policy Sciences 9 (2): 140–50.
Gale, H. Frederick. 1998. "Labor Productivity and Wages," Review of Regional Studies 28 (1): 13–26.
Gersuny, Carl. 1981. Work Hazards and Industrial Conflict. Hanover, Mass.: University Press of New England.
Graham, J., and D. Shakow. 1981. "Risk and Reward," Environment 23 (8): 14–45.
Harmest, Daniel, and John Wolfe. 1990. "Compensating Wage Differentials," Journal of Labor Economics 8 (1): S175-S197.
Hirsch, Barry T. 1993. "Trucking Deregulation and Labor Earnings: Is the Union Premium a Compensating Differential?" Journal of Labor Economics 11 (2): 279–301.
International Commission on Radiological Protection (ICRP). 1991. 1990 Recommendations of the ICRP—ICRP Publication 60. Oxford: Pergamon Press.
Jones, Gary. 1999. "Testimony," U.S. Congress, House of Representatives, Worker Safety at DOE Nuclear Facilities, Committee on Commerce, 106th Congress, Serial No. 106–43. Washington, D.C.: U.S. Government Printing Office.
Kates, R. W. 1978. Risk Assessment of Environmental Hazards. New York: John Wiley and Sons.
Kniesner, Thomas, and John Leeth. 1991. "Compensating Wage Differentials for Fatal Injury Risk," Journal of Risk and Uncertainty 4 (1): 75–90.
Lave, L.B., ed. 1982. Quantitative Risk Assessment in Regulation. Washington, D.C.: Brookings Institution.
Leigh, J. Paul. 1995. Causes of Death in the Workplace. London: Quorum Books.
Levenstein, Charles, and John Wooding, eds. 1997. Work, Health, and Environment. New York: Guilford Press.
MacLean, Douglas. 1981. "Risk and Consent: A Survey of Issues for Centralized Decision Making," College Park: Center for Philosophy and Public Policy. Working paper.
Miller, Richard. 1999. "Statement." U.S. Congress, House of Representatives, Worker Safety at DOE Nuclear Facilities, Committee on Commerce, 106th> Congress, Serial No. 106–43. Washington, D.C.: U.S. Government Printing Office.
Moore, Michael J. 1995. "Unions, Employment Risks, and Market Provision of Employment Risk Differentials," Journal of Risk and Uncertainty 10 (1): 57–70.
National Institute for Occupational Safety and Health (NIOSH). 1999. Identifying High-Risk Small Business. Washington, D.C.: NIOSH.
Rees, Hedley, and Anup Shah. 1992. The Economic Effects of Collective Bargaining. Bristol, UK: University of Bristol.
Rescher, N. 1983. Risk: A Philosophical Introduction. Washington, D.C.: University Press of America.
Robinson, James. 1991. Toil and Toxics. Berkeley: University of California Press.
Shrader-Frechette, Kristin. 1991. Risk and Rationality. Berkeley: University of California Press.
———. 1985. Risk Analysis and Scientific Method. Boston: Kluwer.
———. 1991. Environmental Ethics. Pacific Grove, Calif.: Boxwood.
Siebert, W. S., and Xiangdong Wei. 1998. "Wage Compensation for Job Risks." Asian Economic Journal 12 (2): 171–81.
Sissell, Kara. 2000. "Judge Sentences Executive to 17 Years." Chemical Week May 10.
Starr, C. 1969. "Social Benefit Versus Technological Risk." Science 165 (3899): 1232–3.
———. 1976. "General Philosophy of Risk-Benefit Analysis." Energy and the Environment. Ed. H. Ashley, R. Rudman, and C. Whipple. New York: Pergamon.
Starr, C., and C. Whipple. 1980. "Risks of Risk Decisions." Science 208 (4448): 1115–17.
Upton, Frederick. 1999. "Statement." U.S. Congress, House of Representatives, Worker Safety at DOE Nuclear Facilities, Committee on Commerce, 106th Congress, Serial No. 106–43. Washington, D.C.: U.S. Government Printing Office.
———. 1999. "Statement." U.S. Congress, House of Representatives, Worker Safety at DOE Nuclear Facilities, Committee on Commerce, 106th Congress, Serial No. 106–43. Washington, D.C.: U.S. Government Printing Office.
U.S. General Accounting Office (GAO). 1998. DOE: Clear Strategy on External Regulation Needed for Worker and Nuclear Facility Safety. Washington, D.C.: U.S. Government Printing Office.
Van Ness, Robert. 1999. "Statement." U.S. Congress, House of Representatives, Worker Safety at DOE Nuclear Facilities, Committee on Commerce, 106th Congress, Serial No. 106–43. Washington, D.C.: U.S. Government Printing Office.
Viscusi, W. K. 1983. Risk by Choice. Cambridge: Harvard University Press.
Wallach, Lori, and Michelle Sforza. 1999. Whose Trade Organization? Washington, D.C.: Public Citizen.
IU Press Journals
Home PageMore about Ethics & the Environment
Library
RecommendationTable of Contents
Advance
InformationCopyright
Clearance