INTRODUCTION

The stated purpose of the Occupational Safety and Health Act of 1970 ("The Act") is ". . . to assure so far as possible every working man and woman in the Nation safe and healthful working conditions and to preserve our human resources . . . ." The Act authorizes the Secretary of Labor to issue safety and health standards that ". . . require conditions, or the adoption or use of one or more practices, means, methods, operations, or processes, reasonably necessary or appropriate to provide safe or healthful employment and places of employment." Moreover, when promulgating health standards, the Act requires the Secretary of Labor to set the standard at the level "that most adequately assures, to the extent feasible, on the basis of the best available evidence, that no employee will suffer material impairment of health or functional capacity . . . ." These Congressional mandates, along with those in section 8 of the Act, which addresses some of the activities the Agency can require by regulation, provide the bases for OSHA's proposed ergonomics program standard, which is designed to minimize occupational exposure to conditions that lead to musculoskeletal disorders.

Before issuing such a standard, however, OSHA must demonstrate that other, non-regulatory approaches are not available that could provide an equal or higher level of net benefits. Executive Order 12866 directs regulatory agencies to assess whether an unregulated private market can achieve the same level of social benefits as that expected to result from Federal regulation:

Section 1. Statement of Regulatory Philosophy and Principles.

(a) The Regulatory Philosophy. Federal agencies should promulgate only such regulations as are required by law, are necessary to interpret the law, or made necessary by compelling public need, such as material failures of private markets to protect or improve the health and safety of the public, the environment, or the well-being of the American people. In deciding whether and how to regulate, agencies should assess all costs and benefits of available regulatory alternatives, including the alternative of not regulating. . . . Further, in choosing among those alternative regulatory approaches, agencies should select those approaches that maximize net benefits.

The discussion below considers several alternatives to the proposed ergonomics program standard. The principal non-regulatory alternatives are private market incentives, information dissemination programs, tort liability options, and workers' compensation programs.

PRIVATE MARKET INCENTIVES

Perfect Competition Model

Economic theory suggests that the need for government regulations would be greatly reduced if private markets worked efficiently to allocate health and safety resources. At issue is whether the private market will produce a level of employee safety and health -- and specifically a level of risk of experiencing work-related injuries resulting from the absence of an ergonomics program -- that will be equal to or lower than that potentially afforded by the proposed OSHA standard. A major conclusion of the "perfect competition model" of economic theory is that, in the presence of full information about market choices and outcomes, and with complete mobility of the factors of production, the private market will produce an efficient allocation of resources; such an outcome will be "efficient" in the sense that no person can be made better off without making someone else worse off. (This definition logically implies that reallocating resources and achieving an increase in the production of one commodity is impossible without simultaneously reducing the production of another commodity.)

In the presence of perfect information regarding occupational risks, labor markets (i.e., producers and workers) should take into account the presence of different degrees of risk across different industries and firms. In such a market, wage premiums would be paid to compensate workers engaged in hazardous occupations for the added risk they confront on the job. According to theory, since wages would vary directly with the riskiness of a job (other things being equal), employers would have an incentive to make safety-related investments or to use more costly but safer work practices. In other words, because employers would have to pay their workers a premium to induce them to work in a risky environment, employers would pay to make that environment less risky by introducing controls and safety precautions. From a firm's perspective, an economically optimal level of safety would be reached when the additional money spent on safety is just equal to the additional savings in labor premiums.

According to theory, a fully competitive market will lead to the efficient allocation of resources only if market participants directly bear (in economic terms, fully internalize) all costs, pecuniary and non-pecuniary. If all of the costs associated with occupational safety and health are in fact internalized, then market decisions about occupational safety and health made by producers and workers will be based on a consideration of the full costs of their economic actions. However, if some of the costs of these actions are externalized, i.e., are not borne by producers and employees but by other parties who are external to the firm, they will not be incorporated into the decisions of managers and workers and the resultant market allocation of resources will not be efficient. Costs and other impacts that are imposed on society and are not borne directly by economic decision-makers are termed "externalities" by economists. The existence of such externalities is one reason for the market's failure to produce an efficient allocation of resources, and these externalities are often used as a justification for regulatory intervention.

In a perfectly functioning market, i.e., one without externalities, firms would decide how much to spend on safety and health based on how much they had to provide in the way of risk premiums to workers and how much they had to spend on other health- and safety-related impacts on the costs of production. Workers, on the other hand, would decide whether they were willing to work in a particular job based on the relative riskiness of that job and the extent to which they believe wage premiums for the losses associated with the injuries and illnesses potentially associated with the performance of that job compensate them for these risks.

Voluntary Initiatives

OSHA's survey data (ERG 1999) and information from individual companies show that many firms have begun to respond to the variety of risks posed by work-related musculoskeletal disorders and their associated costs by implementing ergonomics programs for their employees. In fact, many programs go beyond the program OSHA is proposing in its ergonomics program standard. The fact that employers are implementing these programs means that economic incentives have been adequate, at least in some cases, to motivate employers in this direction. Nevertheless, the number of employers who have responded in this way is relatively small: for example, OSHA's survey data suggest that no more than roughly 20 percent of all surveyed firms have implemented training in the identification and prevention of ergonomics-related hazards. Firms that have begun to implement ergonomics programs cannot be distinguished by industry, SIC code, or other obvious factor from those that have not done so, i.e., some firms have implemented such programs, while other firms that face similar workplace hazards and belong to the same industry have not.

The Benefits chapter of this Preliminary Economic Analysis shows that more than 600,000 preventable work-related lost workday musculoskeletal disorders occur each year. When the results of the benefits and cost analyses are reviewed together, it is clear that preventing these disorders will result in more than one dollar in direct cost savings for every dollar spent on controls. The preponderance of the evidence presented in this analysis thus indicates that market forces have failed to curb occupational exposures to ergonomic-related safety and health hazards adequately. Although almost all of the costs of this standard are borne by employers, employers will experience only 35 percent of the benefits of the standard, i.e., those benefits resulting from reductions in workers' compensation costs and in "indirect costs." Most of the benefits will accrue directly to employees and will therefore only be reflected in markets to the extent that employees pass these benefits back to employers in the form of lower risk premiums in wage rates.

There are a variety of reasons why workers may not be paid the risk premiums that would theoretically be necessary to assure that markets provide efficient levels of expenditures on safety and health. First, workers have imperfect knowledge about the nature and magnitude of occupational risk factors. This is particularly true of health risks such as those associated with exposure to asbestos and lead, for which OSHA currently has workplace standards, and health risks for other toxic substances or dangerous hazards, for which mandatory workplace standards may not presently exist. Many workers are likely to be unaware of the work-relatedness of the musculoskeletal disorders caused by working at jobs involving exposure to risk factors such as repetitive motion, forceful exertion, etc. Studies (Fine, et al., 1986; Cannon, et al., 1981; Silverstein, et al., 1997) have shown that there are far more work-related musculoskeletal disorders than are recorded in the medical files of specific firms. Although some of the workers experiencing these work-related musculoskeletal disorders may have recognized them as work-related and chosen for a variety of reasons not to report them, there nevertheless remains a sizeable number of work-related musculoskeletal disorders that were not recognized by workers as work-related. If workers do not recognize a risk as work-related, it will not be adequately addressed by markets.

Similarly, in the area of safety management, many workers may not be fully aware of the risks associated with, for example, work at heights, activities involving electrical equipment, and operations around machinery, particularly if accidents associated with these types of workplace exposures occur relatively infrequently. If an individual worker seldom observes a serious accident, he or she may assign a downwardly biased risk probability to unsafe events; as a result, the worker may demand a wage that fails to fully compensate for these risks.

Incomplete Information and Transaction Costs

Even if they have information regarding the risks associated with musculoskeletal disorders, workers may be unable to adequately incorporate this information into their decisions about leaving, or staying, on the job. For example, workers may excessively discount long-term occupational risks when making employment decisions. Assessing occupational risks for the purpose of determining wage risk premiums is made even more difficult when the differences in risk between two firms are significantly different but are small enough not to be readily observable over a time frame of one to several years. If differences in occupational risk between establishments are not fully incorporated into a worker's employment decisions, any wage premiums paid for risky jobs will not accurately reflect the relative occupational risk of different firms, and thus firms will have little motivation to individually reduce risk. Many employers are also unaware of the importance of some ergonomics hazards in the workplace or of the availability of cost-effective ways of ameliorating or eliminating these risks. However, an awareness of ergonomics hazards is growing among employers and employees, albeit slowly.

A second problem is that the operation of labor markets is such that there are large transaction costs associated with changing jobs. Changing jobs is always costly, and in some circumstances such a change can be very costly. Factors that may make job changes particularly costly include non-transferability of occupational skills, the difficulty of acquiring sufficient human capital to seek alternative employment opportunities, the costs and uncertainty associated with relocating to take advantage of better employment opportunities, the existence of institutional factors such as the non-transferability of pension plans and seniority rights, and the risk of prolonged periods of unemployment. Another factor is that mobile workers tend in general to be younger workers, who are less likely than older workers to be aware of job risks [Viscusi and Moore, 1991]. Often, differences in occupational risk between two firms must be very marked before a worker will change jobs on that basis. Market-determined wage rates are, therefore, unlikely to fully compensate workers for expected occupational health and safety risks, including those related to the risk factors of concern here.

Finally, an important conclusion from economic theory is that no normative significance can be attached to the level of social welfare achieved as a result of an efficient market outcome, e.g., a market could still be considered efficient even if half of the population was fabulously wealthy and the other half starving. In other words, social goals, such as assuring workers a safe and healthful workplace, are not necessarily achieved even by perfectly functioning efficient markets. Thus, while a fully competitive market might produce an "efficient" level of risk in terms of occupational hazards, such a risk level might still be substantially higher than the level of risk considered socially equitable.

INFORMATION DISSEMINATION PROGRAMS

Another alternative to an OSHA standard on ergonomics would be the voluntary dissemination of information about the safety and health risks associated with exposure to the risk factors found in workplaces where ergonomics programs are absent. Better informed workers could more accurately assess the occupational risks associated with different jobs, thereby facilitating those market interactions that result in wage premiums for relatively risky occupations. The proposed OSHA standard recognizes the link between the dissemination of information and workplace risks by requiring that workers engaged in problem jobs be provided with information and participate fully in job hazard analysis and control, training, program evaluation, and other procedures for preventing and responding to ergonomics risks.

In recent years OSHA has helped to promote the creation and adoption of ergonomics programs in the employer community through an assortment of Agency-sponsored programs. Among these OSHA-funded initiatives are the Voluntary Protection Programs; the OSHA Consultation Program; courses taught by experts in effective ergonomics practices and funded by OSHA training grants and awards; an ergonomics home page on the OSHA web site with links to technical support services; and targeted inspection programs. These OSHA-sponsored programs, discussed in Section II of the preamble, have contributed substantially to improvements in ergonomic safety and health in affected workplaces.

There are several reasons, however, why reliance on voluntary information dissemination programs will not yield the level of social benefits achievable through compliance with the proposed ergonomics program standard. First, absent a standard, there are few incentives or mechanisms to ensure that appropriate information will actually be distributed among workers. Furthermore, many hazards are highly specific to individual tasks and work environments. Accurate knowledge about these occupational safety and health risks would thus require each employer to make available specific information about the risk factors present in his or her facility. The lack of incentives or mechanisms for information dissemination is an even greater issue in the case of such facility-specific information.

Second, even if workers are better informed about workplace risks and hazards, other factors, such as barriers to labor mobility, that contribute to market failure would still remain. Finally, as argued above, workers may not be able to evaluate information about long-term health risks accurately when making employment decisions. Better information, therefore, will not ensure that the market will produce wage risk premiums in a manner that is consistent with an efficient allocation of resources. Thus, while improved access to information about occupational risk factors can provide for more rational decision-making in the private market, voluntary information programs will not produce the level of occupational risk envisioned by the OSH Act.

TORT LIABILITY OPTIONS

Greater worker use of tort law in seeking redress from the diverse injuries and illnesses found throughout the nation's workplaces is another example of a possible non-regulatory alternative to the proposed OSHA ergonomics program rule. If workers were able to sue their employers for damages caused by work-related musculoskeletal disorders, the need for an OSHA ergonomics standard would be reduced or eliminated. A tort may be described, in part, as a civil wrong (other than breach of contract) for which the courts provide a remedy in the form of an action for damages. The application of the tort system to occupationally related injuries and illnesses would mean that a worker whose disability resulted from exposure to a workplace risk factor would sue the employer to recover damages. The tort system could thus shift the liability for the direct costs of occupational injury from the worker to the employer, at least under certain specific circumstances.

With limited exceptions, however, the tort system has not been a viable alternative to regulation in dealings between employees and employers, for a number of reasons. For example, all states have legislation making workers' compensation either the exclusive or principal legal remedy available to employees. Generally, tort law can be applied only to third-party producers or suppliers of hazardous products or equipment, e.g., asbestos products. It is often difficult, however, to demonstrate that workplace injuries have been caused by defective or negligently designed products or equipment, a necessary prerequisite for the successful application of tort liability against the supplier or producer of the equipment. For example, according to an article in the Wall Street Journal that appeared several years ago, as of 1994 over 3,000 lawsuits had been filed against equipment makers alleging that unsafe design has resulted in repetitive stress injuries (Felsenthal, 1994). To date, however, such suits have been largely unsuccessful because of: (1) the lack of scientific links associating injuries with specific equipment use, (2) findings by the courts that different musculoskeletal disorders are different injuries, each with numerous possible causes, and (3) conclusions that musculoskeletal injuries resulted from equipment use patterns rather than the design of the equipment itself.

As noted in the previous example, there are many cases where tort liability would be ineffective. One case involves improperly maintained equipment for which the manufacturer would not be liable. Another case would involve a determination by interested parties that the design or manufacture of the equipment did not contribute to the injury. Under those circumstances, there are no grounds to sue the manufacturer. Finally, if the manufacturer were no longer in business -- as employees can discover when old equipment is implicated in an accident -- a lawsuit would be impossible.

In addition, tort liability can introduce additional inefficiencies into the economy. On the one hand, a large corporation can employ the best lawyers and use a multitude of time-consuming legal options available to frustrate an injured employee's attempt to obtain legal redress. On the other hand, a larger corporation will often settle suits out of court even if it expects to win the suit because the out-of-court settlement will cost the firm less than would the legal fees of the best lawyers. For example, a construction worker who suffered severe head injuries after falling 30 feet from a scaffold collected $13.1 million in a settlement negotiated with the companies involved in the construction project (Stelton v. Merchandise Mart Properties Inc., Ill CirCt, No. 91 L 14568, 2/26/1996) (BNA, 1996b).

In another tort case involving an accident at a construction project, a pipefitter unsuccessfully sued the owner and subcontractor of the site for a back injury he suffered when a handrail he was holding became dislodged. The Illinois Supreme Court, reversing the verdict of a lower court, ruled that the worker failed to present evidence that the removable handrail was unreasonably dangerous or violated any industry or safety standard. Nor could the complainant show that the companies were aware of the danger. In fact, according to the testimony of the subcontractor, the dangerous condition may have been the result of a defect in the design of the handrail, for which the manufacturer, rather than the company, would not have been liable. Other evidence showed that removable handrails were common in the industry. Thus, in this example, the absence of clear standards and the issue of determining fault among several parties assured that the tort liability case failed and thus provided no cost internalization (Jordan v. National Steel Corp., Ill., Nos. 84271 and 84282, 10/1/98) (BNA, 1998b).

An example of a case where significant damages were awarded to workers or former workers for injuries resulting from exposure to harmful substances is one involving 11 refinery workers and a longshoreman who received $25.8 million in 1996 for illness or death from alleged exposure to asbestos at their job sites (BNA, 1996a). A Texas county court jury awarded $7.4 million to cover actual damages related to medical and other costs, including pain and suffering, and $18.3 million in punitive damages after hearing evidence that the employees -- sheet metal workers, electricians, laborers, carpenters, insulators and pipefitters -- developed cancers and other diseases and illnesses due to exposure to asbestos products. A widow of a lung cancer victim received $685,000 from the jury decision and her son and daughter each received $50,000 from the case. Other awards in the lawsuit ranged from $190,000 to $1.5 million. The defendant, a leading manufacturer of asbestos insulation, fiberglass and other related building materials, was held directly liable for causing injury to the workers without warning them or their immediate employers of the risks.

To provide an example where a traumatic occupational accident prompted the victim's family to seek resolution through the court system, in 1997 a subway construction worker in Los Angeles was crushed by a refuse container that fell when a chain used to haul the container broke. The victim was employed by a contractor hired by the transit authority operating the subway. The victim's family sued the subway authority, alleging negligence because the authority "selected a 'dangerous' contractor and failed to provide appropriate supervision to ensure the project was safe," according to the plaintiff's attorney [BNA, 1998a]. The transit agency settled the lawsuit for a reported $6 million, one-third of which was to be paid by the transit agency and the remainder to be paid by its insurance carrier.

In the context of the discussion in this section on the effectiveness of tort claims in redressing or preventing work-related injury and illness, a reasonable question that can be raised is the adequacy of such monetary awards in compensating the victims and their families for the premature loss of wages by the primary earner in the family. In fact, legal proceedings generally fail to fully internalize accident costs because of the substantial legal fees associated with bringing court actions. In deciding whether or not to sue, the victim must be sure that the potential award will exceed both the expense and hardship of bringing the lawsuit. Legal expenses commonly include a 33-percent contingency fee for the plaintiff's lawyer, plus court fees and the costs of accumulating evidence and witnesses. The accused firm must also pay for its defense.

In sum, the use of legal action as an alternative to regulation is limited because of the expense and because the vast majority of worker injuries and illnesses cannot be shown to be directly linked to improperly designed or defective equipment or hazardous products or materials. Even in those instances where equipment design is at issue, experience to date with third-party suits indicates the unwillingness of the courts to assign liability for certain types of injuries to equipment manufacturers. The tort system, therefore, does not serve adequately to protect workers from exposure to safety and health risks, including ergonomic risks in the workplace. In practice, the workers' compensation system is thus the only source of compensation for injured workers.

WORKERS' COMPENSATION PROGRAMS

The workers' compensation system came about as a result of the perceived failure of the private market (i.e., both the liability and insurance systems) to compel employers to reduce occupational health and safety risks; the system seeks to shift the burden of the costs of occupational injuries and illnesses from workers to employers. By so doing, workers' compensation requirements can assure that employers consider the costs of occupational injuries and illnesses even if employees fail to understand their risks or are unable to demand full wage compensation for such risks.

Originally designed to force employers to internalize the social costs of such injuries and illnesses, the workers' compensation program has in practice fallen short of this goal and has failed to adequately compensate workers for occupationally related injuries and illnesses. Compensation tends to be especially inadequate in permanent disability cases, in part because of time limits on benefit entitlements and the failure of the system to adjust benefits for changes in a worker's expected earnings over time. Several states restrict permanent partial and temporary total disability benefits either by specifying a maximum number of weeks for which benefits can be paid or by imposing a ceiling on dollar payments. Data provided in the Benefits chapter (Chapter IV) of this analysis show that workers' compensation payments replace only 64 percent of the value of the worker's lost after-tax income and non-wage benefits for temporary total disabilities and 42 percent of the value of lost after-tax income and non-wage benefits for permanent partial disabilities. In addition, under workers' compensation, no award is made for pain and suffering.

Most workers' compensation programs routinely include the employer's injury experience as a factor in determining the level of the employer's premiums. However, the majority of firms (approximately 80 percent) are not rated individually for their safety and health record, i.e., are not "experience-rated." For example, small firms often are ineligible for experience rating because of the high year-to-year variance in their claim rates. Such firms are class-rated, and rate reductions are granted only if the experience of the entire class improves. Even when firms have an experience rating, the premiums paid may not accurately reflect their true economic losses. Segregation of loss experience into classes is somewhat arbitrary, and an individual firm may be classified with other firms that have substantially different accident rates. Then again, a firm's experience rating is generally based on the benefits paid to injured workers, not on the firm's safety record. Thus, employers may have more of an incentive to reduce premiums by contesting claims than by initiating safety measures.

For employers who rely on workers' compensation insurance, the payment of premiums represents the employer's major cost for various forms of occupational injuries and illnesses. However, the mechanism for determining an employer's workers' compensation premium frequently fails to reflect the real costs associated with a particular employer's record. As a result, efforts made by an employer to reduce the incidence of occupational injuries and illnesses are not necessarily reflected in reduced workers' compensation premiums. Similarly, firms that devote fewer resources to promoting worker safety and health do not incur commensurately higher workers' compensation costs. Consequently, the workers' compensation program does not provide direct incentives for most employers to reduce the occupational health and safety risks in their workplaces.

Finally, workers' compensation is an insurance mechanism through which participants spread and share the risk of injury and illness claims, and the costs associated with occupational injuries and injuries are often spread throughout the economy through risk sharing stemming from participation in health insurance programs, i.e., because many workers go to their private physician rather than the company's physician for work-related injuries and illnesses. Several studies (Cannon, et al., 1981; Silverstein, et al., 1995; and Fine, et al., 1986) provide plant-specific evidence of this tendency. The social burden of adverse health effects is also shared by taxpayer-supported programs such as welfare, social security disability payments, and Medicare. Employers have, therefore, less incentive to avoid such losses than they would if they were directly liable for all such claims. This transfer of risk is another reason why the market fails to internalize the social costs of occupationally related injuries and illnesses such as musculoskeletal disorders.

The workers' compensation system does provide economic incentives for larger firms, especially those that self-insure for workers' compensation, because these firms internalize the true costs of the work-related injuries incurred by their workers. Thus larger firms can be expected to have done more about safety and health risks than smaller firms. Results from OSHA's ergonomics survey (ERG 1999) bear out this theoretical proposition: they show that only 25 percent of firms with fewer than 20 employees have analyzed their jobs for risk factors, while fully 86 percent of establishments with 500 or more employees, i.e., the largest firms and those most likely to self-insure, have done so. The same pattern holds for following through on these job analyses: 76 percent of the largest establishments have implemented at least some engineering controls to reduce risk factors, while only 13 percent of firms with fewer than 20 employees have done so. Similarly, data from the same survey on the question of the existence of safety and health programs in workplaces indicate that larger employers are more likely to invest in these programs. These data suggest that, where adequate economic incentives are present, firms find it in their interest to address the risk factors responsible for musculoskeletal disorders and other safety and health hazards. However, as this discussion has shown, such incentives are absent for most firms.

In summary, the workers' compensation system suffers from several defects that seriously reduce its effectiveness in providing incentives for firms to create safe workplaces. First, because the scheduled benefits are often significantly less than the actual losses experienced by injured workers, the existence of workers' compensation programs limits an employer's liability to levels significantly below the actual costs of the injury or illness. Second, premiums for individual firms are often unrelated or only loosely related to that firm's risk environment. The firm, therefore, does not receive the proper "signals" and consequently fails to invest sufficient resources in reducing workplace injuries and illnesses. The economic costs not borne by the employer are imposed on the employee directly or on society through social welfare programs.

Summary

OSHA has determined that workers are exposed to a significant risk of incurring musculoskeletal disorders from a range of job-related factors (see the section of the preamble entitled Significance of Risk). The private market has not been effective in reducing this level of risk due to a lack of information about health and safety risks, limits on worker mobility, and other factors that contribute to the failure of markets to provide an efficient allocation of resources. Options for improving the operations of markets include information dissemination programs, tort liability options, and workers' compensation programs. After considering each of these options, OSHA preliminarily has concluded that none of them will provide the level of social benefits achievable by the proposed ergonomics program standard.

The Bureau of National Affairs, Inc. [BNA, 1996a] OCCUPATIONAL SAFETY & HEALTH REPORTER. Washington, D.C. Volume 25, No. 32, January 17, 1996. (Ex. 26-1637)

The Bureau of National Affairs, Inc. [BNA, 1996b] OCCUPATIONAL SAFETY & HEALTH REPORTER. Washington, D.C. Volume 25, No. 38, February 28, 1996. (Ex. 26-1638)

The Bureau of National Affairs, Inc. [BNA, 1998a] OCCUPATIONAL SAFETY & HEALTH REPORTER. Washington, D.C. Volume 28, No. 6, July 8, 1998. (Ex. 26-1639)

The Bureau of National Affairs, Inc. [BNA, 1998b] OCCUPATIONAL SAFETY & HEALTH REPORTER. Washington, D.C. Volume 28, No. 20, October 21, 1998. (Ex. 26-1640)

Cannon, L.J., E.J. Bernacki, and S.D. Walter. [Cannon, et al., 1981] "Personal and occupational factors associated with carpal tunnel syndrome." Journal of Occupational Medicine 23(4):255-258. (Ex. 26-1212)

Eastern Research Group [ERG, 1999] Tabulations from OSHA's 1993 Ergonomics Survey (Ex. 28-7).

Felsenthal, E. [Felsenthal, 1994] "An Epidemic or a Fad? The Debate Heats Up Over Repetitive Stress." Wall Street Journal. July 14, 1994. (Ex. 26-1636)

Fine, L.J., B.A. Silverstein, T.J. Armstrong, C.A. Anderson, and D.S. Sugano. [Fine, et al., 1986] "Detection of cumulative trauma disorders of upper extremities in the workplace." Journal of Occupational Medicine 28(8):674-678. (Ex. 26-920)

Silverstein, B.A., Stetson, D.A., Keyserling, W.M., Fine, L.J. (1997). "Work-related musculoskeletal disorders: comparison of data sources for surveillance". American Journal of Industrial Medicine, 31:600-608. (Ex. 26-28).

Viscusi, W.K. and M.J. Moore. [Viscusi and Moore, 1991] "Worker learning and compensating differentials." Industrial and Labor Relations Review 45(1):80-96. (Ex. 26-1635)