Recent Research: Rural regions may not be so far behind in innovation capacity

Differences in per capita innovation capacity between urban and rural regions are not as large as previously believed according to a recent working paper from the National Center for Science and Engineering Statistics (NCSES). The study’s conclusions reduce the difference by a factor of three.

The researchers made modifications to earlier definitions of the “inventive class” to better distinguish occupations that are likely to contribute to innovation as measured by patent production. This new definition of the inventive class includes core occupations of science, engineering, and technology as identified by NSF, and also includes other occupations that have been shown in earlier work to be associated with patenting. The authors assert that most prior studies comparing regional invention capacity fail to consider differences in the composition of inventive populations in different regions. They make the case that the set of occupations associated with patenting in rural regions is different from the set of occupations associated with patenting in urban regions.

Using data from the 2000 U.S. census, the authors then examine the relationship between the composition (the share of the total population in the inventive class of a region) and the productivity of the newly defined inventive class (patents/inventive class member). For each region, U.S. Patent and Trademark data on utility patents issued in the span of 2000-2005 is aggregated to account for the time lag between patent filing and issuance. This decomposition allows for standardization across compositional and productivity factors, that is, what observed patenting rates would look like if compositional factors — or productivity factors — were identical across regions. The researchers controlled for many variables across the regions, including the number of patents awarded, population of the inventive class, university research spending, and workforce education levels. They find that when inventive class population share is held constant across the two populations, large urban regions are only twice as productive as rural regions. This smaller than expected difference may influence policy makers as they make decisions on investments to grow innovation economies.

Dotzel KR, Wojan TR; National Center for Science and Engineering Statistics (NCSES). 2022. An Occupational Approach to Analyzing Regional Invention. NCSES 22-202. Alexandria, VA: National Science Foundation. Available at https://ncses.nsf.gov/pubs/ncses22202.