Spillovers from Academic and Industrial R&D Examined

Is a strong university research component critical to local tech-based economic development? Many argue this position, including SSTI (Using Research and Development to Grow State Economies). Using data on royalties, licenses, and job creation figures, others have demonstrated the economic contributions of university R&D. (See for instance, the annual licensing survey from the Association of University Technology Managers.)



But can strong research universities contribute more to local technological innovation than large industrial research laboratories – or can communities achieve the same degree of success in building tech-based economies by concentrating resources toward attracting industrial research laboratories?



At issue, are the local “spillover” effects of industrial and academic research. Spillovers may include, for example, new company formations, job creation, increased research contracts, etc. Which type of research on average, industrial or academic, results in more spillovers to the local economy? John Adams, with the Department of Economics at the University of Florida, explores the issue in Comparative Localization of Academic and Industrial Spillovers, one of the latest Working Papers released by the National Bureau of Economic Research.



Using data from a sample of R&D labs owned by U.S. manufacturing firms, Adams finds the spillover effects of academic research to be more localized than industrial R&D on average. He concludes from his analysis that "localized academic spillovers reflect open science and the industry-university cooperative movement, which encourages firms to work with local universities, so that localization coincides with the public goods nature of science. This situation contrasts with relations with other firms, where contractual arrangements are needed to access proprietary information, often at a considerable distance.”



The policy implications for state and local tech-based economic development efforts, while not addressed in Adams' paper, may be interpreted as supporting efforts to 1) increase academic research activity within local universities, and 2) encourage industry-university collaborative partnerships, which many states do through their centers of excellence and university-industry research grant programs.



Adams concludes that industrial firms tend to turn to local universities more often for applied research, technical assistance, and students/employees. In contrast, industrial partnerships are more selective and more driven by “personnel movements and collaborative ventures.”



Other conclusions in the paper would lend support to tech-based economic development strategies of attracting more federal R&D grants in local academic institutions, stressing investment in basic or early stage research, and encouraging more university patenting. Adams reports the greater the amount of federally supported university R&D near a firm drives localization of academic spillovers and, subsequently, industrial spillovers.



Spillovers from university patenting also was found to be more localized than industrial patenting. The paper also suggests “spillover from new products, although weaker than patent spillovers, are still more localized” for universities than industrial product development.



Adam's last point suggests that state and academic policies and programs that encourage faculty tech-based entrepreneurship and technology commercialization may prove particularly beneficial for building local tech-based economies.



Comparative Localization of Academic and Industrial Spillovers (NBER Working Paper No. 8292) can be purchased online from NBER for $5.00 at: http://www.nber.org/papers/w8292

[Editor’s note: For much of the econometric literature, an acceptable distance for “localization” can be as much as 100-200 miles of the university or industrial plant. Readers should not think spillover effects are limited only to those communities in which the institution or research lab is located.



While much has been written on understanding and measuring spillovers, additional research exploring which policy and public program approaches were most successful in encouraging academic research and the resulting spillovers would be a useful next step toward refining public tech-based economic development strategies. Policy analysis into those public programs which increase the localization of industrial R&D spillovers also would be beneficial.]