Patents,R&;D and lag effects: evidence from flexible methods for count panel data on manufacturing firms

This paper investigates the relationship between patents and research and development expenditures using new longitudinal patent data at the firm level for the U.S. manufacturing sector from 1982 to 1992. The paper also develops a new class of count panel data models based on series expansion of the distribution of individual effects. Estimation results from various distributed lag and dynamic multiplicative panel count data models show that the contemporaneous relationship between patenting and R&D expenditures continues to be strong, accounting for over 60% of the total R&D elasticity. The lag effects are higher than have previously been found for the 1970s data. We would like to thank Chris Bollinger, Bronwyn Hall and Paula Stephan for useful comments on the previous version of the paper. Earlier versions were presented at the 11th International Conference on Panel Data, Texas A&M University, the Midwest Econometrics Group Meeting, and the Annual Conference of the Southern Economic Association.