As part of the effort to reduce the size and economic impact of the federal establishment, congressional conservatives are proposing legislation to restrict the regulatory activity of the Occupational Safety and Health Administration (OSHA). These proposals push OSHA toward a purely consultative role, at a corresponding cost in direct regulatory capability. The Clinton administration's reinvention of government initiative is also moving OSHA toward a consultative role based on a strategy of cooperative compliance or industry self-regulation with a strong coercive foundation. Since both camps appear to agree that self-regulation can assure a safe and healthy workplace, the remaining debate concerns the extent to which coercive regulation is still needed. National survey data on the industrial provision of occupational safety and health services in the manufacturing sector were used to measure changes in industrial safety and health activity between 1972-74 and 1981-83. In conjunction with data on OSHA command-and-control regulatory activity from 1972 to 1979, these data permitted an examination of the relationship between command-and-control regulatory activities and changes in industrial behavior that could be regarded as a form of self-regulation. This analysis showed that coercive regulation by OSHA in the 1970s was significantly related to industry self-regulation efforts, although the relationship varied by industrial facility employment size and type of regulatory coercion. These results indicate that coercive regulation should be retained as an industrial incentive in any self-regulation policy paradigm. The results also provide evidence that OSHA regulatory policy should be based on anticipated differences in industrial response to various coercive measures.