Health, safety, and environmental regulation.

Gray, Wayne B.

Since the early 1970s, the United States has seen a substantial increase in federal government regulation, through the establishment of several new regulatory agencies. The Occupational Safety and Health Administration (OSHA) and the Environmental Protection Agency (EPA) have aroused particular controversy, attacked by some for being too strict and imposing large costs on business, and by others for being too lenient and allowing many hazards to remain. Congress is currently considering legislation affecting both agencies, and the debate about how much regulation is appropriate continues.

My research on OSHA and EPA has considered both the benefits and costs of regulation. I have examined the effectiveness of OSHA enforcement in reducing injuries, violations, and worker exposures to health hazards. My studies of the costs imposed by regulation have focused on the impact of regulation on productivity.

OSHA Safety Regulation

Traditionally, U.S. regulatory agencies have followed a "command and control" strategy, establishing standards designed to reduce hazards and requiring firms to meet them. Since compliance is costly, some firms are likely to violate standards unless the agency's enforcement activity provided a sufficient deterrence. Critics of OSHA's safety regulation have argued that infrequent inspections and small penalties provide little deterrence. In addition, many injuries are caused by factors not covered by standards, so even complete compliance would not eliminate injuries completely.

John T. Scholz and I have found evidence that OSHA inspections do reduce injuries.(1) This impact appears to be restricted to inspections followed by penalties; not being penalized means OSHA found nothing gong, and nonpenalty inspections had little or no effect on injuries. Plants penalized by OSHA averaged a 22 percent reduction in injuries over the following few years. We argue that the presence of penalties serves to focus the firm's attention on safety issues, eventually leading to hazard abatement. Having multiple inspections of the same plant within the same year, with or without a penalty imposed, does not seem to reduce injuries further. Imposing larger penalties also does not seem to have much impact on injuries, at least within the range that OSHA usually employs.(2)

We also have examined the effectiveness of different types of inspections. For example, many inspections are based on worker complaints rather than being planned by OSHA. Some have argued that if workers are ignorant of true hazards and simply use complaints to harass employers, then these inspections may be a waste of OSHA resources. However, we find that complaint inspections are about as effective as planned inspections in reducing injuries.(3) Complaints appear to be especially effective in larger plants, which tend to be more heavily unionized. Complaints also seem to be less reliant on penalties for their effectiveness, perhaps because the complaining workers can take advantage of the inspection to force hazard abatement, even when the firm is not penalized.

OSHA Health Regulation

Measuring the effectiveness of OSHA regulation for health hazards is difficult, since work-related illnesses can take years or decades to develop. This means that we must rely on indirect indicators of potential future illnesses, such as worker exposure to hazardous substances, rather than direct measures of current injuries. Since OSHA's health standards often are aimed precisely at lowering worker exposures to hazards, violations of health standards also could indicate future health problems.

In work with Carol A. Jones, I examined OSHA's effectiveness, using dam that linked all OSHA inspections at a given plant over time. First inspections tend to find the most violations, and subsequent inspections of the same plant find progressively fewer problems, suggesting that plants are responding to the earlier inspections by reducing hazards.(4) For health and safety inspections, the first inspection seems to lead to a 50 percent reduction in violations.

We find similar effects when we focus exclusively on health inspections, with a 45 percent reduction of violations after the first inspection.(5) OSHA inspectors also collect test samples to document workers' exposure to health hazards. We find that worker overexposure to hazardous substances falls, with a decline of 37 percent after the first inspection, and continues to decline after subsequent inspections. These results suggest that OSHA is effective in reducing health hazards, at least in those plants that have had health inspections. Because most of the reduction in hazards is attributed to the first inspection of a plant, our results also provide a suggestion for improving enforcement effectiveness: focus more on uninspected plants, rather than inspecting the same plants repeatedly.

Regulation and Productivity

It can be difficult to identify the costs imposed by regulation. Compliance with a particular regulation may involve new capital equipment, the energy and labor needed to operate the equipment, and the managerial and engineering expertise to decide which equipment is needed. For environmental regulation, there are surveys of pollution abatement expenditures, but these may not consider all the different types of costs. If a plant chooses to completely redesign its production process to reduce emissions, it may be difficult to identify what fraction of the redesign costs should be attributed to pollution abatement.

One way to go beyond reported compliance costs is to examine productivity measures. This is especially true for measures of total factor productivity, which account for the contribution of both capital and labor to output. If a changed production process means higher production costs, then productivity will fall. Examining productivity also can help measure compliance costs of other regulatory areas, such as OSHA, where no cost surveys are done.

My research on regulation began by examining the reasons for the slowdown in manufacturing productivity during the 1970s. I compared the productivity experience of different manufacturing industries with the amount of OSHA and EPA regulation directed toward them.(6) Regulation measures included enforcement activity for both OSHA and EPA, as well as pollution abatement costs. Industries that faced more regulation had slower productivity growth and a greater productivity slow-down in the 1970s. About 30 percent of the productivity slowdown in the average manufacturing industry could be attributed to OSHA and EPA regulation.

Modeling Regulation

The decisions of regulatory agencies and regulated firms can be interconnected, especially when the regulators want to avoid adverse publicity associated with plant closings. Mary Deily and I examined this issue using data on air pollution enforcement and plant closings in the steel industry, where such pressures were acute: 43 percent of the plants we studied closed by 1986.(7) Plants facing greater expected enforcement were more likely to close, although the magnitude of the effect was small (raising enforcement by 12 percent would increase the probability of closing by only 1 percentage point). Enforcement decisions appear to be more sensitive to the likelihood of a plant closing: a 10 percentage point increase in the probability of closing reduces enforcement by 6.5 percent.

1 W. B. Gray and J. T. Scbolz, "Does Regulatory Enforcement Work? A Panel Analysis of OSHA Enforcement," NBER Working Paper No. 3 774, July 1991, and Law and Society Review 27 (1993), pp. 177-213; also J. T. Scholz and W. B. Gray, "OSHA Enforcement and Workplace Injuries: A Behavioral Approach to Risk Assessment," NBER Working Paper No. 2813, January 1989, and Journal of Risk and Uncertainty 3 (1990), pp. 283-305.

2 W. B. Gray and J. T. Scholz, "Analyzing the Equity and Efficiency of OSHA Enforcement," Law and Policy 13 (1991), pp. 185-214.

3 W. B. Gray and J. T. Scholz, "How Effective Are Complaint Inspections?" report submitted to OSHA, June 1992.

4 W. B. Gray and C. A. Jones, "Longitudinal Patterns with OSHA Health and Safety Regulations in the Manufacturing Sector," NBER Working Paper No. 3213, December 1989, and journal of Human Resources 26 (1991), pp. 623-653.

5 W. B. Gray and C. A. Jones, "Are OSHA Health Inspections Effective? A Longitudinal Study in the Manufacturing Sector," NBER Working Paper No. 3233, January 1990, and Review of Economics and Statistics 73 (1991), pp. 504-508.

6 W. B. Gray, "The Cost of Regulation: OSHA, EPA, and the Productivity Slowdown," NBER Working Paper No. 1405, July 1984, and American Economic Review 77 (1987), pp. 998-1006.

7 M. Deily and W. B. Gray, "Enforcement of Pollution Regulations in a Declining Industry, "Journal of Environmental Economics and Management 21 (Fall 1991), pp. 260-274.