Association for Public Policy Analysis & Management

The social efficiency of regulatory programs is highly controversial, and this is particularly the case for the Occupational Safety and Health Administration (OSHA). While the goal of OSHA is to assure safe and healthful working conditions, improvements in worker safety may not outweigh the cost of compliance borne by firms. Moreover, the effect of inspections on worker safety is difficult to measure empirically, as OSHA specifically targets high-risk industries and establishments for inspection. This generates a negative correlation between inspections and worker safety, which confounds any positive, causal effect of the former on the latter.

To identify the effect of inspections on worker safety, this study exploits quasi-experimental variation in inspections generated by OSHA’s Site Specific Targeting (SST) plan. The SST plan, implemented in 1999, prioritized inspecting establishments with injury rates higher than a cutoff. The cutoff is associated with a discontinuous increase in inspections, which is used to identify the effect of inspections on worker safety. The data are from the OSHA Data Initiative (ODI), which collected establishment-level data on accidents and injuries directly from employers, matched to OSHA inspection history from OSHA’s Integrated Management Information System (IMIS).

Using local linear regression, this cutoff is associated with a 22.7 percentage point increase in programmed inspections. The discontinuity in inspections is used to identify the effect of inspections on worker safety. The estimated effect of an inspection on cases involving days away from work, job restrictions, and job transfers is -1.792 per 100 full-time equivalent workers, representing a decrease of approximately 20 percent. This estimate is statistically significant at the five percent level and robust to the inclusion of establishment characteristics. Moreover, the effect on the DART rate is most pronounced among manufacturing establishments below the 90th percentile of the DART distribution post-inspection.

This study contributes to a literature on the effects of OSHA inspections on worker safety. The literature finds a wide range of effects, depending on the identification strategy, firm size, definition of treatment (inspection versus citation or penalty), and worker safety outcome (overall injuries versus specific types) (Kniesner and Leeth, 2004). While this study uses a different identification strategy and analysis period from related literature, the estimated effects from this study are large compare to related studies.