The Role of Invention in U.S. Metropolitan Productivity
Jonathan T. Rothwell
Arizona State University (ASU)
University of North Carolina Charlotte
April 9, 2013
It is well established in economic literature that innovation and invention are primary engines of economic growth, but there has been surprisingly little empirical work on the specific mechanisms, with credit going to such diverse attributes as human capital, population density, institutions, and geography. Patent records offer a more direct measure of inventive activity at national and regional scales through an institution that retains at least some of the value of invention for inventors and their employers. This article examines the relationship between inventive productivity and economic productivity at the level of U.S. metropolitan economies, using a new patents database that links inventors to their metropolitan areas. We combine these data with the economic characteristics of metropolitan areas over the 1980 to 2010 period in a panel estimation framework. We find that patenting Granger causes higher productivity in metropolitan areas. Higher-quality patents, measured through claims, enhance the measured effect. Patenting is associated with higher productivity in the tech and manufacturing sectors, but not the local services or legal services industries. Long-run wages appear to be higher in more patenting-intensive metropolitan areas. The presence of high-productivity industries, high-tech workers in a highly educated regional economy, and a large population also contribute to metropolitan productivity growth apart from whatever contributions they may make to patenting.
Number of Pages in PDF File: 51
Keywords: patents, innovation, productivity, urban economics, economic development, growth, technology
JEL Classification: O1, O18, O3, O31, R11, I22, O40working papers series
Date posted: November 14, 2012 ; Last revised: May 9, 2013
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo7 in 0.313 seconds