[8]

Regulatory Reform

The problems facing OSHA have been severe, and, based on the record, it is clear that the American approach to occupational hazards should be changed. The preceding analysis suggests that regulatory reform must confront how the political and economic power of business has combined with an overly statist approach to regulation to limit severely the ability of public officials to change what employers do about working conditions. Too few resources have been transferred from other, more profitable investments to worker health and safety. Workers have not taken a large part in decision making over the conditions of work.

Accordingly, any serious regulatory-reform proposal will have simultaneously to increase the incentives for employers to invest in health and safety and increase workers’ ability to participate in determining the conditions of work. This chapter evaluates alternative regulatory-reform proposals according to these two criteria. Three alternative proposals to restructure OSHA are examined; the first two are conventional and have been discussed in earlier chapters. One approach is conservative; it leaves the allocation of risk and protection to private action in markets, supplemented by lawsuits. The second approach is a neoliberal one that seeks to “rationalize” administrative regulation by imposing cost-benefit and cost-effectiveness tests on standards and enforcement. Both approaches will be considered in greater detail here than was possible earlier.

The third approach is not conventional, at least not in discussions of regulatory reform in the United States. For lack of a better term, I call it a labor-oriented approach. This term is used to designate that cluster of features that underlies the neocorporatist and social democratic occupational safety and health programs that are common in many other advanced capitalist societies. I focus in particular on the possibilities and limits of the social democratic variant. This approach relies on a more radical, worker-oriented reordering of the relationship among the state, the firm, and the market than is usual in discussions of American policy, or likely in the foreseeable future. Nonetheless, the approach offers an important comparative perspective on the nature of the liberal approach to workplace regulation and the limits of the current policy debate in the United States.

Market-Conservative Approaches

As I noted in the discussions of deregulation and regulatory reform in Chapters 4 and 6, market-conservative approaches are based on a neoclassical economic model in which the private, self-interested actions of employers and employees in markets determine the level of protection and the choice of control technologies. Based on this model, regulatory reformers propose two changes in the current approach to occupational safety and health that could be adopted independently or used to supplement each other. The first would leave worker protection entirely to labor markets. As Smith writes, “Occupational safety would be bought and sold on the same basis as most other goods.”¹ The second change would amend the current no-fault workers’ compensation to facilitate employee negligence suits. This would, it is claimed, create additional economic incentives for employers to change their health and safety practices.²

Labor Markets and Social Welfare

The neoclassical view of the market in general, and labor markets in particular, treats occupational safety and health as a good that is bought and sold as part of the wage bargain between employers and employees. According to this model, when labor markets are allowed to operate freely, the forces of supply and demand determine the levels and kinds of risk faced by individual workers, the level of investment in occupational health and safety, and the approaches taken to control hazards.³

On the supply side, employers offer safe and healthy work when they offer jobs. Some employers, usually those for whom it is relatively easy to reduce risks, seek to attract workers by offering relatively safe jobs with relatively low wages. Other employers bid for workers by compensating for high risks with high wages. Given the number of different industries and occupations, and the variable costs of protection across jobs, in combination employers are likely to offer a wide variety of jobs combining different levels of risk and wages.

On the demand side, different workers have different preferences for income and safety. Some workers prefer safe work and will accept lower wages to get it. Other workers will take hazardous work if there are risk premiums associated with it. In an analogous fashion to what employers do, workers distribute themselves among risky jobs with high wages and lower-paying jobs that involve less hazardous work.

According to its advocates, the neoclassical approach is both efficient and democratic because it maximizes individual liberty and, simultaneously, leads to a socially optimal level of risk and protection. Based on information about hazards and wages, and workers’ own preferences for safe work versus higher incomes, some workers will choose wages over safety and others will choose safety over wages. At the same time, competition among employers for workers will lead firms to reduce risks in the least costly manner possible. Employers who choose to offer safe jobs will also be in a position to discover the least costly methods of protection because they understand the technical details of their enterprises.

In an equilibrium situation—one in which the demand for, and supply of, health and safety intersect—the worker’s total compensation (wages plus the marginal cost of safety per worker) will be equal to the value of the worker’s marginal product. At that point, the supply of safety will reflect workers’ demands for protection and the resource costs of reducing risk. In economists’ terms, this is a Pareto efficient outcome. That is, no one can be made better off by a change in the level of safety without imposing costs that are larger than the protected worker is willing to pay.⁴ In contrast, government action is likely to yield “suboptimal” levels of protection. State agencies will impose stricter standards than workers would themselves choose. Moreover, given the agencies’ legalistic approach to social problems, they will probably prefer uniform standards and engineering controls to variable standards and PPDs. As a result, employers will be prevented from adopting the least-cost methods of controlling hazards.

Those who argue for this approach acknowledge two defects in the market for worker health and safety and recommend remedies for them. First, workers do not have accurate information about health hazards because of their uncertainty about what they are exposed to and the causes of diseases. Second, the workers’ compensation system distorts incentives to firms to invest in safety because it allows employers to pool their risks. Most employers pay premiums based on industrywide injury experiences rather than their own plant records. Because insurance premiums do not accurately reflect the individual employer’s efforts, the workers’ compensation system prevents the firm that invests in safety from realizing the complete economic benefits of that investment.⁵

Presumably, state action can supplement the market-based allocation of risk and protection by correcting these defects. Specific changes have been proposed to solve these problems. Deficiencies in health information can be remedied by labeling laws and information programs that encourage and/or subsidize full disclosure of health hazards. Alternatively, the liability laws can be changed so that workers can sue employers to recover the costs of work-related disease. Employers could be allowed to defend themselves against charges of negligence by undertaking programs to monitor workplace exposure, disseminate current information to employees, and subsidize private research efforts.⁶ Disincentives to employers to invest in safety could be eliminated by experience-rating all firms. Alternatively, small deductibles, to be paid by the employer, could provide an incentive for all firms to reduce injuries.⁷

This model looks particularly attractive when it is compared to the practical problems associated with command-and-control regulation. In the market-conservative approach, standard setting and enforcement are decentralized and, therefore, more flexible and efficient. Labor markets allow employers to choose to supply a level of safety appropriate to the costs of hazard prevention in their firms. Employers can then attempt to minimize their costs by experimenting with control methods and selecting those methods that are most suitable for particular plants. Court action targets specific employers who create unusually hazardous conditions. Negotiated court settlements encourage the development of flexible solutions to particular problems.

To advocates of this approach, markets in health and safety are not just theoretical possibilities; they exist in the real world. Reviewing the evidence on whether workers and employers act in ways consistent with this theory. Smith concludes that “a market for job safety probably does exist.” Moreover, “This market functions, on the whole, as it should.”⁸ Some advocates of market-based incentives argue that the existence of this market mechanism allows for the dismantling of OSHA and the restoration of the private system that dominated health and safety before 1970, although it would be supplemented by a reformed workers’ compensation program.⁹

A market-based approach will, it is claimed, maximize individual liberty because it is self-enforcing. Private action in labor markets replaces the bureaucratic supervision of employer practices. When state action is necessary, the courts replace the executive branch. Judicial decision making is preferred because it rests on the initiative of private parties. In either event, individual freedom is maximized.

In sum, private action seems to provide occupational safety and health effectively and democratically. If workers choose protection over wages, the level of investment in risk reduction should rise. Successful employee liability suits should reinforce the economic incentives to employers to devote resources to health and safety. Workers can even purchase, or bargain for, control over working conditions. Indeed, the threat of negligence suits might lead employers to share power with employees in order to limit their legal liability. Legislative reforms could sanction this arrangement.

The Limits of Market Capitalism

The market capitalist view must be rejected on two related grounds. First, it fails to acknowledge how markets actually work in capitalist economies. Second, it does not consider how market “imperfections” discourage investment in health and safety and worker participation in plant governance.

Consider, first, the problem of market competition. Even if we take the view that there are markets in occupational safety and health—a position that is itself controversial—it does not follow that these markets are competitive. Labor markets are imperfect, and since protection is bought and sold in labor markets—as an aspect of labor rather than a separately produced commodity—occupational safety and health markets are also imperfect. For example, discrimination and other employer practices divide workers into distinct, noncompeting groups. In addition, large firms in concentrated industries often enjoy considerable monopsony power. As a result, the tastes of the marginal worker (i.e., the last worker hired) will determine the amount of safety provided by the firm. But these tastes are not likely to represent the tastes of all workers; workers who readily move between jobs are likely to be younger and less experienced, less familiar with and worried about job hazards. As a result, the market value of protection will not reflect the preferences of workers as a whole.¹⁰

The chronic oversupply of labor also shapes workers’ preferences for safety and health. The shortage of jobs and the fear of economic insecurity should discourage many workers from complaining about hazards and encourage them to accept higher risks than they would otherwise. The costs of moving between jobs are likely to contribute to the undervaluation of health and safety.

Another problem arises because both forms of private action—the sale of labor in markets and court suits—assume the existing class-based distribution of income and wealth. A worker’s initial market position is apt to shape his or her preferences about the tradeoff between safety and wages, ability to choose among jobs, and ability to mount and sustain legal action. A distribution of occupational safety and health that reflects these inequalities should result. Thus, private action will leave poorer workers with less protection than more affluent workers enjoy.

Advocates of market-based incentive systems are generally indifferent to this outcome. Smith, the leading proponent of the market approach, acknowledges that workers must have “choice in the selection of jobs” for a market in occupational safety and health to function. But he fails to acknowledge existing economic constraints on those choices. To Smith, poor people prefer high wages to safe jobs; they are not led to that choice by circumstance. Speaking of the market in health and safety, he suggests:

The end result would be that people who do not value additional safety very highly relative to additional income—whether because they are poor, have high pain thresholds, or just do not care—would choose more wages over more safety. . . . Those who value additional safety highly would tend to take jobs in plants where safety is relatively cheap to ensure.¹¹

Nevertheless, Congress was quite concerned with the distributional issues raised by occupational safety and health when it wrote the OSH Act. The American welfare state has tended to promote equal protection, and this legislation extended the notion to the workplace. Congress created a universal right to safe work. In doing so, it required the agency to achieve levels of protection that are not likely to correspond to what workers would choose in markets. Administrative rules were purposefully substituted for market processes with the understanding that the mandated level of risk reduction could depart significantly from what workers selected when they traded wages for protection.

Equally important, the OSH Act orders OSHA to try to equalize risk among workers. The obligation to equalize risk can be seen in several of the act’s provisions. These have been noted previously but they are worth restating here. The act’s expressed purpose is “to assure so far as possible every working man and woman in the Nation safe and healthful working conditions” (Sec. 2b; my emphasis). To this end, the act requires that each employer “furnish to each of his employees employment and a place of employment which are free from recognized hazards that are causing or are likely to cause death or serious physical harm to his employees” (Sec. 5a[ 1 ]). To enforce these rights, the secretary of labor is required, when issuing standards that deal with toxic materials or “harmful physical agents,” to “set the standard which most adequately assures, to the extent feasible . . . that no employee will suffer material impairment of health or functional capacity even if such employee has regular exposure to the hazard dealt with by such standard for the period of his working life” (Sec. 6b[5]; my emphasis).

In short, Congress did not distinguish among hazards or attempt to regulate only particularly high-hazard industries. It sought to provide all workers with safe jobs. Smith’s recommendations, then, violate the law’s basic premises.

Private legal action is problematic on other grounds. Strict liability could replace negligence as the legal standard for damage suits. This change would liberalize the conditions under which workers could sue employers. Employers could be held responsible for system failures such as design flaws or random and uncontrollable breakdowns, as well as gross negligence. This should result in a large number of successful class-action suits over occupational diseases such as silicosis and hazards such as hearing loss. In the long run, these suits would provide incentives for employers to prevent hazards.

But costly settlements would threaten the economic viability of many industries. In the short and mid-term, they would probably bankrupt a large number of firms and a significant number of industries. Efforts to liberalize workers’ compensation for silicosis and asbestosis have already resulted in government limitations on the recovery of medical benefits in some states.¹² Liability suits against the Manville Corporation (formerly Johns Manville) for asbestos exposure led that company to seek the protection of the bankruptcy courts. It is probable that the proliferation of successful damage suits such as these would lead the federal government to cap employers’ liability and create some form of public subsidy to affected firms. The Black Lung Program has already done this for the coal industry. The program subsidizes employers, restricts the ability of workers to sue them, and therefore limits the incentive effects of court action.

Nor are markets and court action apt to encourage worker participation in decisions over working conditions. As I stated earlier, collective action is necessary if worker participation mechanisms are to function effectively. But markets tend to discourage the kinds of collective actions that are necessary if workers are to take a more active part in determining the conditions of work. Markets are, after all, famous for encouraging privatism and individualism. As individual buyers of health and safety and sellers of labor, workers become selfinterested utility maximizers. As rational individual actors, workers are most likely to respond to the market economistically, that is, by changing jobs or demanding higher wages rather than seek control over working conditions. Legal action is unlikely to correct this situation. Instead, it might encourage economism by focusing worker demands on compensation rather than prevention, and shift worker struggles from the shop floor to the courtroom.

The market-conservative approach, then, falls short on both of the relevant criteria. First, because workers are likely to undervalue health and safety, the market is unlikely to create incentives sufficient to produce a level of investment in health and safety appropriate to the hazards that workers face, or to be responsive to demands that workers would make under less constrained circumstances. Legal remedies are not likely to change this outcome substantially. Second, by discouraging collective action, this approach would make it difficult for workers to play a central role in plant-level decisions on working conditions.

Rationalizing Regulation

A second set of reforms seeks to maintain a leading role for public power but to make regulation more “rational” by introducing economic values into standard setting and enforcement. In this sense, it reflects the growing importance of neoliberal policy analysts and post-New Deal Democrats. To both groups, it is imperative that public officials take the impact of their decisions on capital investment into account when they select policy goals and that implementation strategies are chosen that minimize economic disruption caused by state action.¹³ As we saw in Chapters 4 and 6, this view became increasingly popular after the recession of 1973-1975.

Applied to workplace safety and health regulation, three specific changes have been suggested: (1) the use of economic review to set priorities among hazards; (2) the adoption of performance rather than detailed design standards; and (3) the use of injury taxes in place of penalty-based inspections.

Making Regulation Efficient

Those who propose to rationalize regulation do not deny that worker protection raises equity issues, but they do suggest that worker rights represent one particular claim that must be balanced against the general claims of society. Since these general claims are premised on economic growth and capital investment, advocates of rationalizing regulation urge the adoption of one or more of the following methods to review standard setting and enforcement: cost-benefit analysis, cost-effectiveness analysis, and the regulatory budget.

Cost-benefit analysis has been proposed as both a decisional tool and a decisional rule. As a decisional tool, it requires that agencies identify, monetize, and quantify the consequences of regulation. The Carter administration promoted cost-benefit analysis in this form. As such, it was designed to give public officials a precise measure of the effects of proposed policies and one way of comparing various regulatory alternatives. It did not require agencies to reject actions when costs exceeded benefits. It did, however, force agency officials to justify these proposals. Adopted as a decisional rule, as the Reagan administration used it, cost-benefit analysis requires that policymakers reject rules and policies whose net benefits do not exceed their net costs.¹⁴

Cost-effectiveness analysis can be applied to regulatory goals and methods. Applied to goals, it requires that agencies that are attempting to reduce several risks at once distribute the costs of regulation in such a way as to equalize the benefits gained across programs. For example, a dollar spent controlling cotton dust should be as effective in reducing the risk of byssinosis as a dollar spent in reducing the incidence of asbestosis. Applied in this fashion, cost effectiveness is similar in effect to cost-benefit analysis, as Zeckhauser and Nichols’s summary of the case for a discretionary cost-benefit test indicates. Used as a decisional tool, cost-benefit analysis would, they claim,

force OSHA to examine the consequences of its standards more closely . . . it would highlight inconsistencies in different areas; it might show, for example, that at current levels of stringency one standard costs $5 million at the margin per expected life saved, while another could be tightened at a cost of only $5,000 per expected life, thereby yielding 1,000 times the OSH gain for its cost impositions. In such a case, by loosening the first standard and tightening the second, it would be possible both to increase longevity and to free resources for other uses.¹⁵

Alternatively, cost effectiveness has been used to recommend compliance techniques. The agency has been urged to select the most cost-efficient method to achieve whatever regulatory goal it chooses. Since neither approach requires that regulations pass a cost-benefit test, this methodology has been proposed as a compromise between those who support and those who oppose economic review. According to its proponents, this method leaves room for political and ethical considerations but also acknowledges the importance of costs to firms and the economy.¹⁶

The regulatory budget takes a different approach. Central authorities impose a limit—the budget—on the total costs to the economy that a single agency can impose each year. Within this budget constraint, regulators can choose what and how to regulate. This idea surfaced at the end of the Carter administration and was actively promoted by OMB. Because the Reagan administration dropped it in favor of mandatory cost-benefit analyses, the idea remains undeveloped, and various issues remain unresolved. Such issues include who would set agency budgets, how these figures would be determined, and whether agencies could “bank” regulatory costs from year to year. Presumably Congress would set aggregate cost limits, and the White House would review agency proposals to determine whether they meet these legislatively determined budgets.¹⁷

Proposals to replace design with performance standards and to substitute an injury tax for fines levied by inspectors are intended to increase the employer’s discretion in complying with health and safety goals. Performance standards substitute health and safety targets, such as fewer lost-workday injuries or the reduction or elimination of worker exposure to a particular hazard, for detailed design specifications. The firm then chooses among competing control technologies, ranging from personal protective devices to complex engineering controls. The agency, in turn, penalizes the firm when it fails to meet its targets, instead of penalizing the firm when it fails to adopt specific machinery or work practices.¹⁸

The injury tax is an economic incentive system that also maximizes employer discretion in the choice of control methods. Its intellectual roots can be traced back to economic arguments for effluent taxes to control pollution and the concept behind experience-rating firms under the workers’ compensation system. Schultze’s 1976 Godkin Lectures on regulation stimulated interest in applying this approach to a wide range of regulatory programs.¹⁹

Taking this approach, public officials would establish protective goals by setting charges for injuries. These charges could be set at various levels depending on how much protection was sought. If the market model is taken as a standard, the charge could be set at the level necessary to raise the costs of accidents to firms to the point where they equaled the costs of prevention. Or charges could be set above or below the level of economic efficiency to achieve more or less than the economically “optimal” level of protection. In either event, the injury tax would be cost efficient because employers would be free to choose the least expensive method of reducing injuries. In the case of workplace safety and health, an injury tax would allow a firm to experiment with various remedies until it found the particular mix that was suitable to its work environment.²⁰

Rationalizers argue that these kinds of reforms are necessary to deal with the problems of resource scarcity and political accountability raised by social regulation. Given resource scarcities, agencies accept the need to choose regulatory goals carefully and minimize compliance costs. Protective statutes, however, are often ambiguous; health and safety risks are widespread and numerous; and well-organized constituencies pressure agencies for the highest possible levels of protection. Thus, the agencies are likely to set regulatory goals that cannot be justified by efficiency or equity considerations.

In contrast, economic review helps agencies set priorities among competing goals and choose the most cost-efficient means of achieving them. It focuses attention on the implications of their choices and encourages them to make choices rationally. Cost-benefit tests, cost-effectiveness analyses, and the regulatory budget make agencies conscious of scarcity and force them to choose efficient control technologies.

The review process also helps hold agencies accountable to higher authorities and limits the debilitating effects of “special-interest” legislation on the economy. Because they supervise the process. White House officials enjoy a number of levers over agency policy, including oversight of the preparation of the analyses, the final determination of agency requests to regulate when costs outweigh benefits, and the power to set agency cost budgets. Presumably these officials would be in a better position to resist interest-group pressures for higher protection and bureaucratic pressures for expanded agency authority.

Corporatism Without Labor?

Whatever its promise, without major changes in the existing approach, economic review is apt to evolve into a distinctively American form of corporatism, one that encourages close contact between business and government but discourages participation by rank-and-file workers or organized labor. As a result, it is unlikely to be used to increase democratic decision making on protection or redistribute resources toward worker health and safety.

Economic review procedures are likely to limit rather than encourage investment in worker health and safety for several reasons. First, the monetization of costs and benefits usually devalues income and wealth transfers. All forms of economic review assess the consequences of regulation by adding and subtracting market prices. But market-determined prices result from exchanges made between unequal parties, and the review process valorizes these income and wealth differences. Any attempt to impute prices to untraded commodities—to use wage differentials to discover how workers value low- and high-risk work, or to discount a worker’s future earnings to determine the present value of his or her life—does the same. Poorer workers are revealed to value their health and safety less highly than more affluent workers; old people, because they do not have future earning power, turn out to be worth less than young people.

Given marked economic inequality, the effects can be perverse. Cost-benefit analyses can recommend highly regressive projects and recommend against redistributional programs because of the relative economic positions of different income classes. One of the leading experts in this area suggests that this disregard for the distributional consequences of cost-benefit tests is likely to occur even when the analyst attempts to introduce distributional criteria into the analysis:

Even if it were conceivably possible to secure permanent agreement . . . on the set of distributional weights to be attached to the benefits and losses of different income groups, it could not . . . be counted on to prevent the introduction of projects having markedly regressive distributional effects. Projects that would meet a weighted cost-benefit criterion could be such as to make the rich richer and the poor poorer if the beneficiaries were rich and many and the losers poor and few.²¹

A review effort that sought to calculate costs and benefits on the basis of individuals’ market choices would have similar effects. As noted, several factors shape workers’ tradeoffs between safer and better-paying jobs, including their income and wealth positions. A poor, unskilled worker is more apt to take a risky job than an affluent, professionally trained worker is. Since a worker’s initial market position will shape how he or she values health and safety, standards based on this kind of benefit analysis will distribute protection according to these preexisting market inequalities.

Unfortunately, this point is often ignored; when it is acknowledged, it is usually sidestepped. In Risk by Choice, Viscusi admits that this distributional criticism is “fundamental” and “legitimate,” but offers a weak response to it.²²

Efforts to promote present risk regulations on the basis that they enhance worker rights are certainly misguided. Uniform standards do not enlarge worker choices; they deprive workers of the opportunity to select the job most appropriate to their own risk preferences. The actual “rights” issue involved is whether those in upper income groups have a right to impose their job risk preferences on the poor.²³

Instead of redistributing through regulation, Viscusi suggests accomplishing redistributional ends through direct resource transfers. Such transfers would give poor people the same opportunity that other workers have to turn down unsafe jobs.

This makes sense, in theory, as do arguments that resource transfers and the benefits of regulation could be calculated differently. Cost-benefit tests, for example, could be weighted to favor redistributional transfers by multiplying benefits to workers by some politically determined factor: the poorer the worker, the larger the number. But political realities argue against this outcome. Instead, review procedures are likely to continue along the established trajectory and take their benchmarks from the market, with its existing inequalities.

In addition, cost-benefit and cost-effectiveness tests usually result in the disaggregation of the universalist right to protection found in the OSH Act. Review procedures normally apply a neoclassical concept of efficiency to standard setting, and this is likely to produce decisions that are directly contrary to the act’s mandate to equalize risk.

As discussed earlier, hazards are unevenly distributed across jobs and industries. Different individuals face different kinds of hazards at work. Moreover, some hazards are more difficult to control than others. Consequently, it is more expensive to protect some workers than others. Given the neoclassical economist’s notion of efficiency, standards that provide equal protection to workers in diverse settings are irrational. They do not deliver the greatest net benefits, and they do not equalize the marginal costs of protection across jobs and industries. Efforts to use economic review to correct these problems—whether to set priorities or accept or reject particular standards—will result in the disaggregation of the universalist right to health and safety created by the act.

The CWPS intervention in rulemaking in the coke-oven emissions case provides an excellent example of how this has been done. When CWPS recommended that OSHA consider regulating risks in “other occupations with both higher relative risks and much larger absolute numbers” of workers at risk, it claimed to speak in the name of efficiency. Were CWPS actually arguing that OSHA should protect other workers, this recommendation might be less disturbing. But CWPS’s critique of OSHA’s proposed standard reflected the White House’s desire to control the total costs of regulation. Moreover, it implicitly rejected the act’s legislative mandate to reduce risks for all affected workers. In short, economic review resulted in recommendations that, if implemented, would have led to a lower level of investment in health and safety than that proposed by the agency or envisaged in the act.

The use of economic review to impose regulatory methods such as performance standards and injury taxes is also likely to limit investment in health and safety. Most occupational safety and health and safety professionals subscribe to the idea that there is a “hierarchy” of preferable controls beginning with control at the source, then proceeding to control of the transmission or dispersion of toxic substances, and finally to control at the point where the worker is directly exposed to hazards by personal protective equipment, work practices, and administrative controls. But there is a tradeoff between the costs of control technologies and their effectiveness, and the profit-maximizing employer will probably opt for the least costly rather than the most effective approach. Indeed, employers have resisted engineering controls because they are much more expensive than dealing with worker behavior.

Theory and experience suggest, however, that the hierarchy-of-control concept is valid. Control at the source minimizes the problem of worker error resulting from stress, fatigue, or boredom. In practice, PPDs rarely work as well as they are supposed to. Masks, for example, leak and rarely fit well; they make it difficult to breathe and impede communication among workers—in itself a hazardous condition. Workers often remove them or falsely believe that they are receiving adequate protection when they are not. Thus the employer’s discretion to adopt this option should be carefully limited to maximize worker protection.²⁴

Nonetheless, as we have seen, White House reviewers in every administration endorsed this view and argued for PPDs on cost-effectiveness grounds. They pressed OSHA to allow employers to use ear plugs to protect workers from noise, and masks to protect them from dusts and gases. Indeed, many of OSHA’s critics argue for incentive mechanisms because they free the employer to adopt PPDs and work practices in the place of engineering controls.

Economic review is also likely to devalue worker participation. No administration has attempted to assign participation a positive monetary value in its own right. In the labeling controversy, OMB expressly denied the value of worker rights to know, apart from their specific impact on increased safety and health. In fact, economic review is apt to treat worker participation as a cost; labor time devoted to health and safety will weigh against protection in cost-benefit tests.

Recommendations to maximize the firm’s discretion in the name of economic efficiency are also likely to devalue worker participation. In theory, an employer seeking to maximize the efficient production of health and safety might involve workers in an effort to take advantage of their knowledge and skills and increase their motivation to take care. But there are powerful economic incentives to employers to resist worker control. It is more probable that the rational employer will forgo the possible health and safety benefits of increased worker participation in favor of the known benefits derived from labor discipline and control.

In sum, proposals to rationalize regulation promise to increase the effectiveness of whatever resources are devoted to health and safety, but they are likely to discourage investment in worker health and safety and worker participation at work. In this way, they administratively rewrite the OSH Act and subvert the goals of an effective occupational safety and health program.

Labor-Oriented Approaches

Labor oriented approaches refer to workplace regulation in which workers and unions play a central role in policymaking through one or another variant of neocorporatism. These approaches are common in Western capitalist countries and constitute a distinct alternative to the liberal approach that dominates workplace regulation in the United States, as well as the two approaches to regulatory reform discussed above.

Not all labor-oriented approaches are radical. Like their American counterparts, many Western European labor movements have reached accords with business and accepted a subordinate place within a welfare capitalist system; production remains organized according to capitalist criteria. Even social democratic regimes presume a good deal of private control over investment and work.

Nonetheless, in many European societies, organized labor is more powerful in politics and economics than it is in the United States. As a result, workers play a greater role in policymaking—through social democratic and labor parties and union confederations at the national level—and exercise more influence over employers at the local level—through works councils and other forms of worker organization. Where labor is more powerful, occupational safety and health policymaking tends to reflect that power by providing unions and workers with institutional arrangements that facilitate participation in standard setting and enforcement. Some of these arrangements seem to increase investment in health and safety and worker control over the conditions of work.

Neocorporatism

Neocorporatist approaches to occupational safety and health combine tripartite standard setting with worker participation in enforcement. As in the United States, basic rights and responsibilities are codified in statutes, and the state has the power to inspect and fine firms for violating standards. But this approach encourages business and labor to use government as a forum to negotiate general policy and particular standards. It also deemphasizes penalty-based inspections. At the same time, neocorporatist approaches encourage worker participation at the plant level by creating statutory health and safety committees and occupational health programs. Thus, neocorporatism combines bargaining over policy and standards by the leaders of business and labor—a highly centralized form of decision making—with decentralized systems of local enforcement.

Advocates of tripartite standard setting claim two major advantages for it. First, tripartitism is supposed to reduce conflict and assure that employers comply with standards. Because policymaking boards and commissions provide union organizations and business confederations with regular, guaranteed access to government decisionmakers, tripartitism encourages negotiations that lead to mutually acceptable agreements. Because employers are given formal representation, they are more apt to implement the agreements that are reached. And, to the extent that the officially sanctioned employer associations speak for business as a whole, are compulsory membership organizations, and provide essential services to their constituents, they are likely to be able to force recalcitrant firms to comply with negotiated agreements.²⁵

For several reasons, neocorporatism also is supposed to maximize the flexibility of policy. For example, secondary issues that arise within the general parameters established by legislation can be resolved through regular negotiations between labor and business. Because these negotiations are informal—they do not follow the judicial model of conflict resolution characteristic of the American system—participants can reach compromises that balance their interests across a wide range of related issues. This system encourages flexibility by facilitating package deals in which concessions are made on some points in return for victories on others. Finally, because these negotiations can be organized to design and implement long-term plans, organized labor has more certainty that bargains struck today will be enforced tomorrow. Unions are therefore more likely to trade off present gains for future benefits, thereby allowing employers the opportunity to come into compliance more slowly and at less cost.

The following comparison between the logic of the liberal and neocorporatist approaches illustrates the point: The American approach relies on an expansive, statutory commitment to worker health to be implemented by expert administrators. It does not rely on tripartite arrangements to decide policy or set standards. In tact, unions have resisted negotiating standards or compliance agreements because they see them as a departure from the statutory rights granted to workers in the OSH Act. Lacking institutional mechanisms to assure that concessions in the present will lead to implementation in the future, the labor movement expects that any departure from the act’s ethic of protection will lead them down a slippery slope of deeper and deeper cuts without any corresponding benefits.

Given the liberal approach, and the unions’ own political and economic weakness, organized labor is acting rationally when it adopts this strategy. Unless forced to, employers are unlikely to compensate reasonable workers with future benefits. Under these circumstances, each compromise and every concession is a management victory rather than a step toward a long-term goal of hazard reduction. Moreover, once they take strong positions, unions are led to defend them in order to maintain their reputation as powerful political actors and their influence over public officials and their own members.

In contrast, a neocorporatist approach to standard setting allows the state to commit public authority to an expansive definition of protection but, within this general frame, forces unions and employers to negotiate priorities and timetables and implement these settlements. For example, organized labor and employer groups can negotiate the pace of standard setting and the application of standards to particular firms and industries in keeping with a plan that seeks to reduce risk across all occupations in the long run. Organized labor can make other concessions with some certainty that these ultimately will be rewarded. Given an enforceable agreement to raise investment in health and safety, cost-effectiveness and cost-benefit tests can then be used to channel new investment to the most immediately productive uses rather than question the logic of equal protection. Organized labor also can negotiate review procedures that place a positive value on worker participation.

Indeed, it might prove easier to gain worker acceptance for incentive mechanisms and employer experimentation if workers believed that employer compliance could be easily monitored and failed experiments corrected. Requests for variances from specified design changes, approvals to experiment with new control technologies, and special dispensations to economically distressed enterprises could be negotiated industry by industry, or firm by firm, in return for commitments to detailed plans to increase investment when it became economically feasible. Where such investment might never be economically feasible, workers could negotiate concessions on other issues of concern to the affected workers, such as job security or participation in firm governance. Ideally, the entire social regulatory enterprise would take place within the context of economic planning designed to facilitate democratic determination of which distressed industries are worth saving, and at what cost in accidents and injuries.

Under these conditions, the idea of universal protection would be treated as a goal rather than an immediately enforceable rule. It is not unimaginable that organized labor might endorse a review program based on this premise. After all, it is neither impossible nor unreasonable to rank the various threats to worker health and safety and take up the most pressing ones first. An index based on what is known about a substance’s morbidity and the degree of exposure to it is one obvious measure. But, in the United States, health and safety reformers have rejected this concession because they think it might lead to the protection of some workers at the expense of others. It has. Under different arrangements, however, economic review might serve different interests.

Advocates of the neocorporatist approach to workplace safety and health also argue that it increases worker participation in enforcement. As a rule, these systems do include stronger statutory commitments to worker participation at the workplace, including legally required health and safety committees, safety representatives, works councils, and rights to know about hazards and refuse hazardous work. The existence of these rights and institutions, in turn, facilitates worker activity at the enterprise-level because workers know about, and recognize, hazards and have the organizational capacity to pressure employers to reduce them.

Antidemocratic Tendencies in Neocorporatism

The historical record indicates that corporatism can easily devolve into an elitist form of policymaking that discourages democratic participation by the rank and file and advantages only a few workers. Informal negotiations between organized labor and business groups can lead to a kind of administrative politics in which deals are struck that serve only the leaders or the best-organized interests. As a rule, the degree to which neocorporatism actually advantages workers depends on how much power workers really do exercise under these arrangements.

In regard to workplace regulation, four rather stringent conditions are necessary if neocorporatism is to lead to higher levels of investment in occupational safety and health and facilitate effective worker participation in plant governance. First, and probably most important, the labor movement must be strong enough to represent workers’ interests effectively in negotiations with employers and force elected officials to implement these negotiated agreements. Second, the labor movement must make occupational safety and health a priority; otherwise, it will not devote its resources to the issue, and worker interest will wane. Third, workers and worker representatives must be able to function independently of management. This means that they must enjoy statutory rights to have formal access to state inspectors, to refuse hazardous work, to know about hazards, and to veto or withhold consent to management decisions concerning health and safety. Finally, workers must have access to the resources necessary to police the workplace, monitor firm compliance with relevant standards, and participate in enforcement, including information, organizational capacity, time, and money.²⁶

The comparative experience suggests that these preconditions are rarely met in full because of deficiencies in legislation and less-than-aggressive trade union strategies. In Britain, for example, joint health and safety committees or, in small firms, safety representatives have been mandatory since the mid 1970s. A 1974 law requires that occupational safety and health representatives be appointed, and a 1975 statute gives the unions the sole right to choose these representatives. Subsequent amendments have given health and safety representatives rights to inspect enterprises, to be paid for health and safety training, and to investigate accidents.²⁷

These rights are limited in several ways. The committees’ powers are qualified, and employers can veto decisions that require resource expenditures or changes in personnel practices. Moreover, worker rights rest exclusively on trade union action; employees do not have rights as individuals. The unions select worker representatives to the joint health and safety committees, and these union-selected representatives exercise the rights to participate in inspections and refuse hazardous work.²⁸

Compounding the problem, the British labor movement has not emphasized health and safety. Consequently, the health and safety committees and safety representatives do not play an active role in the determination of working conditions. Worker participation has not resulted in control over plant-level decisions that affect health and safety, and workers have not sustained interest in these programs.²⁹

West Germany reveals a similar pattern of limited rights and trade union strategies. There, works councils and health and safety committees enjoy broad statutory powers to participate in general plant governance, including the right to veto health and safety decisions. The councils, the committees, and the rights are mandated by law; workers participate in occupational health and safety programs through them.³⁰

But worker rights are limited in several ways. Members of works councils are a minority on the safety committees and have few operative powers. As in Great Britain, the rights of individual workers are restricted. The right to refuse dangerous work is sharply limited, and workers cannot call in safety inspectors. Moreover, worker rights rest almost entirely on union initiative for their force, and the union movement has not made health and safety a priority issue. As a result, the rights that workers enjoy are largely unexercised. In contrast, business groups enjoy privileged access to standard-setting bodies, and management-oriented public health professionals dominate the system. Thus, despite their statutory rights, workers do not participate in most of the important decisions concerning occupational hazards, including the application of new technologies.³¹

The Social Democratic Variant

The social democratic approach is a variant of neocorporatism distinguished largely by the power that organized labor enjoys in politics and reforms designed to facilitate worker control over plant-level decisions. Thus, this variant combines more effective union participation in national standard setting with more active worker participation in enforcement. In both instances, strong legislation is reinforced by more radical labor strategies. Standard setting, for example, includes organized labor on an equal footing with business. Health and safety committees exercise real power over working conditions. Individual workers enjoy full rights to participate in inspections, to know about hazards, and to refuse hazardous work.

Advocates of this approach suggest that these arrangements provide the benefits of neocorporatism but avoid the problems that often surface with the neocorporatist approach. Most important, greater worker participation in this approach limits the antidemocratic tendencies of neocorporatism in several ways. By involving workers in plant programs, the social democratic approach raises their consciousness about health and safety and increases the likelihood that they will take an active part in these programs. Workers also are provided with organizational forums that help them coordinate their activities and lower the costs of participation. Additionally, hands-on experience in meeting plant-level health and safety goals provides workers with the information to assess central decisions about what is practical and necessary. Worker participation also encourages workers to pressure unions to represent their interests in health and safety. Unions are then more likely to pressure the state to take health and safety seriously, and to pressure employers to comply with government regulations. Finally, worker participation reduces the burden that occupational safety and health regulation places on public officials. Properly trained, organized, and represented, rank-and-file workers can observe firm practices, including compliance with standards and negotiated agreements, and pressure plant supervisors more effectively than a distant national agency can. In fact, a mobilized workforce should reduce the need for government inspectors to monitor working conditions.

Like market-conservative approaches, the social democratic approach also builds on private action in two ways. First, workers play a greater role in enforcement than with any other approach. Second, with better information and stronger rights, workers are able to make informed choices about the jobs they wish to take or, if they are interested in taking risky jobs for higher wages, to make informed decisions about that tradeoff. But this private action occurs within a different institutional setting. Most important, political mechanisms exist through which workers can democratically determine their rights and protective goals. These institutions, in turn, help to discourage privatism and individualism and encourage workers and unions to make the organization of work a political issue.

The social democratic approach should also overcome some of the problems encountered when workers rely exclusively on collective bargaining to reduce occupational hazards. Collective bargaining is a relatively ineffective way to change the conditions of work. Union contracts rarely cover all workers or provide covered workers with equally strong protections. Instead, they result in a highly variegated pattern of worker rights. This is apt to encourage divisions among workers and decrease the likelihood that the labor movement will make health and safety a priority. Even when contractual agreements specify worker rights, these rights are not apt to be enforced without worker representation in statutory health and safety committees or works councils.³²

In sum, social democratic arrangements have the potential to provide workers with greater information about, and leverage over, working conditions than other approaches. In-plant organizations can monitor the hazards faced by employees and make sure that state inspectors do their jobs. By applying independent pressure on employers, they can help state agencies assure that firms remedy unsafe conditions. Thus, if workers participate, a social democratic approach can use the state to supplement, reinforce, and build on the virtues of private action without succumbing to the limits inherent in market-based approaches.

Sweden

Only the Scandinavian experiments in social democracy actually approximate the conditions outlined above. Sweden in particular illustrates the potential impact of social democracy on the regulation of working conditions. There, the majority of workers are organized and the Social Democratic party has governed for all but six years since the Great Depression. The state is more interventionist than is normal in capitalist democracies and enjoys a wide variety of powers over capital flows and labor markets. Occupational safety and health policy is equally well developed.

Participation in workplace decisions in Sweden is justified in political and philosophical terms. The 1977 Work Environment Act declares its aim to be “for work to be arranged in such a way that the employee himself can influence his work situation.” The Swedish Labor Ministry has interpreted “the underlying idea” of the act to be that “active participation by employees can establish a form of working life characterized by partnership and shared responsibility, security and meaningful jobs and job satisfaction.”³³

Standard setting is tripartite in form. The Worker Protection Board (Arbetarskyddsverket, or ASV), the Swedish counterpart to OSHA develops rules in consultation with tripartite committees. The committees include representatives of the Landsorganisationen (LO) and the Tjänstemannens Centralorganisationen (TCO), the Swedish central blue-collar and white-collar labor organizations, and the Svenska Arbetsgivareföreningen (SAF), the employer organization responsible for negotiating economywide collective bargaining agreements. Committee meetings are informal and coordinated by ASV. The management and labor groups serve as interlocutors between ASV and individual firms and union locals. They circulate standards proposals to their members and communicate members’ proposed revisions to ASV. The agency’s powers are substantial: it does not have to keep elaborate written records of its meetings or provide for formal hearings and comments; it enjoys final authority to set standards; legal challenges to standards are not permitted.³⁴

Although Swedish employers have input into general occupational safety and health policy and ASV decisions, the strength of the Swedish labor movement and the state’s leverage over the process of capitalist investment have encouraged public officials to adopt strong health and safety programs, including extensive worker rights to participate in plant-level decisions. The 1976 Joint Regulation of Working Life Act and the 1977 Work Environment Act give workers the right to participate generally in firm governance and specifically in the determination of working conditions. Workers are guaranteed a variety of statutory rights, including the right to refuse hazardous work, to stop work in imminent-danger situations, to know about hazards, to participate in plant activities, and to supervise the operation of in-plant health services. Employers are required to fund factory-based medical clinics for occupational safety and health. Health and safety committees supervise these in-plant programs, and shop stewards are authorized to monitor and investigate workplace hazards. They also enjoy the right to shut down plant operations temporarily in the face of imminent hazards.³⁵

Collective bargaining between SAF and the two central labor organizations has reinforced these statutory rights. In 1979 employers and employees agreed to increase the autonomous powers of the health and safety committees to veto the appointment of health and safety experts, to organize the medical service, and to give worker representatives a permanent majority on the health and safety committee.

Reflecting organized labor’s political power, ASV standards are strict—most of them are stronger than the equivalent OSHA standards. In 1978, ASV exposure levels were lower than OSHA exposure levels in over half of the 145 cases in which both agencies had standards. In contrast, OSHA levels were stricter than ASV standards in less than 3% of the cases.³⁶

The situation in Sweden also indicates that when organized labor negotiates from a position of strength, standards are likely to be decentralized and more easily revised. Labor unions frequently agree to compliance schedules that take the particular economic conditions of firms into account. In turn, SAF attempts to secure the compliance of individual firms with government standards. The ASV standards also are more frequently updated than OSHA standards and more apt to be based on current research. Finally, employers are much less likely to challenge ASV rules.³⁷

By no means has Swedish social democracy eliminated the tendency for capitalist production to undersupply health and safety. But more extensive state control over investment and labor markets and more politically conscious and well-organized labor movements create powerful countervailing forces that have led to increased investment in health and safety—as ASV’s stricter standards suggest—and increased worker participation in decisions about working conditions.

The Contradictions of Social Democracy

As the analysis above indicates, the success of labor-oriented approaches varies with the strength of the labor movement and the willingness of public officials to act against the interests of private investors and firms. Workers must be well organized and take occupational safety and health seriously. Public officials must be held accountable on this issue. Herein lies the dilemma for those who advocate a social democratic approach. Since social democratic regimes govern capitalist democracies, private property in production, profit-seeking investment, and labor markets continue to operate. In combination, they encourage firms to disregard hazards and employees to focus their efforts on short-term material gain. Although the political organizations of workers and reformist administrations can counteract these tendencies, the normal operation of the system undermines reform. Thus the state and labor must struggle continually to maintain the ground that they have won.

If the union movement shifts its attention to other issues, the standards set through tripartite negotiations are apt to weaken. The European cases discussed above suggest how labor strategies affect occupational safety and health policy. Both Great Britain and Germany adopted worker participation programs in the 1970s, but in both countries, worker rights far outstrip the reality. Neither the British nor the West German labor movement has responded to the rights that they enjoy. The British program gives unions the rights to select safety representatives and take part in public programs, but the Trades Union Congress (TUC) has not taken advantage of these opportunities. Few unions employ full-time health experts to supplement existing safety representatives. The TUC does not conduct an independent research effort, as LO does in Sweden.³⁸ Similarly, the West German program depends directly on union participation, and the unions have proved indifferent; the works councils have not focused on the issue. Worker participation is limited and, as in other areas of codetermination, employers continue to make most of the important decisions about working conditions.³⁹

Social democratic regimes also suffer from a kind of political-economic cycle that undermines reform. Brought to power by a mobilized working class, left-wing parties are able to build stronger states and adopt more redistributive welfare and economic policies. But, once in power, social democratic parties become responsible for managing a capitalist economy. The need to reconcile social policy with capital investment and international competition emerges as it does in more liberal regimes. Social democratic parties are held responsible for economic crisis, and if economic problems become severe, they may moderate their programs.⁴⁰

Nonetheless, social democratic regimes stand a better chance of prevailing in the face of these tendencies than do liberal regimes because of the existence of more radical worker organizations and greater state control over investment. To the extent that workers remain mobilized and committed to increasing public control over investment, the impact of economic decline on social policy can be contained. Most important, occupational safety and health need not suffer to the extent that it does in liberal-pluralist societies.

The Liberal Approach in Comparative Perspective

The preceding analysis brings us back to the point raised at the outset. Neither market-conservative nor neoliberal approaches to regulatory reform are likely to solve the problems inherent in the liberal approach to workplace regulation because neither successfully confronts how the political and economic power of business discourages investment in health and safety and worker participation in the determination of working conditions. Nor does either provide a compelling alternative to the statism of the liberal approach. Market-conservative proposals to substitute private action for state power rest on an unrealistic theory of the market that ignores how class inequalities shape the distribution of risk and protection and frustrate efforts by workers to protect themselves from hazards. Neoliberal proposals to rationalize regulation synthesize the values of capitalist production and regulation, but in a way that almost invariably argues for less, rather than more, protection and devalues worker participation—either because economic theory demands it or because public officials, dependent on capitalist investment, are reluctant to challenge these assumptions.

The discussion of labor-oriented approaches to workplace safety and health helps to clarify these issues by suggesting alternative ways of combining public and private action in capitalist democracies. The social democratic approach in particular argues for a radically different approach to the relationships among capitalist production worker participation, and state intervention. State action services and is subordinated to the political and economic organizations of workers. State-enforced rights facilitate rather than replace collective action by workers. Thus, the social democratic approach promises to make workplace safety and health policy simultaneously more effective and more democratic.

The social democratic approach is not about to be adopted in the United States in the near or mid-term. Nonetheless, lessons can be learned from it. First, it underscores the narrowness of the American debate about regulatory reform. In addition, it provides several concrete suggestions about how reformers might approach social problems of this kind in the future. I now turn to this second set of issues by considering the future of social regulation in the United States. At the same time, I return to the theoretical issues raised at the outset of the book.