This paper studies the influence of R&D in the U.S. federal laboratory system, the world's largest, on firm research. Our results are based on a sample of 220 industrial research laboratories that work with a variety of federal laboratories and agencies and are owned by 115 firms in the chemicals, machinery, electrical equipment, and motor vehicles industries. Using an indicator of their importance to R&D managers, we find that cooperative research and development agreements (CRADAs) dominate other channels of technology transfer from federal laboratories to firms. With a CRADA industry laboratories patent more, spend more on company-financed R&D, and devote more resources to their federal counterparts. Without this influence, patenting stays about the same, and only federally funded R&D increases, mostly because of government support. The Stevenson-Wydler Act and amendments during the 1980s introduced CRADAs, which legally bind federal laboratories and firms together in joint research. In theory the agreements could capitalize on complementarities between public and private research. Our results support this perspective and suggest that CRADAs may be more beneficial to firms than other interactions with federal laboratories, precisely because of the mutual effort that they demand from both parties.

This content is only available as a PDF.