What happens when firms patent? New evidence from US economic census data∗

We build a new concordance between the NBER Patent Data and US Census micro-data, and use it to examine what happens when firms patent. We find strong evidence that increases in patent stock are associated with increases in firm size and scope as well as with changes in factor intensity. We find somewhat weaker evidence that changes in patenting are positively correlated with changes in total factor productivity. We also analyze a sample of first-time patentees, and find significant increases in firm size, scope, and factor intensity subsequent to their first patent application. Evidence for increases in total factor productivity is slightly weaker. Together, these results suggest that patenting is indeed associated with real changes within firms, and in particular that patenting is strongly associated with firm growth through increases in scope. These findings are robust to alternative measures of size, scope and productivity, and to various sample selection criteria.