Universities and Regional Economic Growth
Since 1980, governments and universities throughout the United States and other advanced countries have set up an array of initiatives to more closely link universities to industrial innovation and to spur local economic development based on university research. Many of the national efforts have been modelled after the U.S. Bayh-Dole Act. At the local level, universities and local governments have set up research parks located near university campuses, public venture capital funds, and business incubators with the goal of encouraging the formation of university spin-offs and regional high-tech clusters linked to university research. These policies have been motivated in part by a recognition that the high-value sectors of the economy are increasingly knowledge based and that universities, as both suppliers of knowledge workers and a source of ideas for industrial innovation, can play an important role in determining the location of new economy industries. Of more immediate and practical concern is that public funding of higher education and university research has been declining. Universities have been eager to find new sources of revenue and to demonstrate their broad economic value to government officials.
Is it realistic to expect universities to be such a powerful attraction for high-tech firms? There are certainly some success stories. The current spatial distribution of biotech firms in the United States owes much to the residence and university affiliation of star scientists who made important early contributions to gene sequencing research. As has been told countless times in simple if incomplete stories of the Silicon Valley, faculty from Stanford University were important in the early growth of the semiconductor industry. Research Triangle Park would not have developed into a large high-tech cluster without early and continued support from the state of North Carolina.
But associations do not imply causality. Case studies of knowledge flows between universities and industry show that these flows are often bidirectional. The expertise and needs of existing firms in an area influence the direction of local academic research. University faculty who make important technological advances often come from industry. Historical analysis of the origins of the Silicon Valley indicates that the region’s propensity for innovation dates to the early part of the twentieth century, well before Stanford University had developed a reputation for excellence in electrical engineering. The consensus view among economic geographers is that the particular locations of regional high-technology clusters are the result of a complex and unique set of historical circumstances that would be difficult to replicate. There is very little evidence of policies encouraging university-industry linkages being successful in creating high-tech agglomerations.
Setting aside the lofty goal of creating another Silicon Valley or Research Triangle Park, is there a general tendency for regions with a university to have a more educated population and above-average earnings per worker? A significant econometrics literature says “yes,” even if the effects are modest in size. Universities appear to have become increasingly important for local economic performance since the mid-1980s, coinciding with the rise of the knowledge economy and efforts to commercialize university research. As a determinant of local economic activity, university research seems to be more important than degree production, which is consistent with the idea that migration plays a large role in determining the spatial distribution of the college-educated population. One study of U.S. metro areas found university research to be a statistically significant determinant of relative earnings growth, but its contribution to variations in economic performance across metro areas was small relative to other variables such as metro area size (agglomeration). At the international level, one study covering 1,500 regions in 78 countries found that a 10 percent increase in the number of universities was associated with a 0.4 percent increase in regional per capita GDP. The authors concluded that these impacts were not simply demand driven, but also seemed to be the result of contributions that universities make to the supply of human capital and to the stock of knowledge in the local region. Their results also suggested that a portion of the long-run impact universities have on per capita GDP occurs because of greater support for democratic institutions in countries with more universities.
Brekke, T., “What Do We Know About the University Contribution to Regional Economic Development: A Conceptual Framework,” International Regional Science Review 44 (2021): 229-261.
This is a compilation of 193 articles from 1994 to 2018 identified from the Web of Science using the search string (“University” or “Higher Education”) and (“Regional Economic Development” or “Regional Innovation System”). The assembled articles were written by scholars from a variety of disciplines and use cases and evidence from the United States and Western Europe. Specific topics covered include (1) alternative mechanisms for knowledge transfer such as university spin-offs, research collaboration, licensing of patented university knowledge, and science parks; (2) the importance of universities as sources of skilled labor used in knowledge production; (3) alternative R&D strategies of firms, e.g., the extent to which they seek knowledge and expertise from a global information network as opposed to local sources, and are willing to use private-sector providers of research services; and (4) the relative strengths of top-down federal regional development policies versus the bottom-up, contextual policies of local and regional governments.
Drucker, J. and H. Goldstein, “Assessing the Regional Economic Development Impacts of Universities: A Review of Current Approaches,” International Regional Science Review 30 (January 2007): 20-46.
The authors review 122 articles that attempt to quantify the effect universities have on regional economic development, drawing heavily from articles written by economists and published in economics journals. The articles reviewed are primarily studies of U.S. universities, although there is some coverage of studies of Canadian and Western European experiences. Studies are classified according to one of four methodological approaches: (1) economic impact studies of individual universities that use detailed information on the university’s expenditures together with a regional input-output model to estimate the direct and indirect effect of the university’s operations on local income and employment; (2) surveys of firms that elicit information on the importance of regional universities in their location decisions and on how they specifically utilize universities in their innovation activities; (3) studies that estimate regional knowledge production functions that express research output as a function of inputs such as the R&D expenditures of industry and universities; and (4) regression-based studies that utilize cross-section data and quasi-experimental design to test for an empirical relationship between the presence of a university and outcomes such as the location of high-technology firms or the level of average local real earnings.
Hill, K. “University Research and Local Economic Development,” Center for Competitiveness and Prosperity Research, Arizona State University, August 2006.
A selected review of 48 papers through 2004 is provided with a complete background narrative on topics ranging from the long historical role of universities in industrial innovation, to the importance of tacit knowledge as a fundamental reason for the localization of knowledge transfers, to empirical evidence of knowledge spillovers from case studies, patent citations, and econometric studies.
Peer, V. and M. Penker, “Higher Education Institutions and Regional Development: A Meta Analysis,” International Regional Science Review 39 (2016): 228-253.
This is a compilation of 127 case studies on the role of higher education institutions (HEIs) in regional economic development. The studies are drawn from the English and German language literature using Google Scholar, Scopus, and other electronic databases to search using the string (“University” or “Higher Education”) and “Regional Development.” The studies come from a variety of disciplines and are international in scope. The authors code and provide a meta-analysis of the case studies based on the primary policy question addressed in the study. Eight alternative policy goals are identified from strategic policy papers in Austria and the European Union:
(1) Has a regionalization of HEIs caused more of the local (rural) population to participate in higher education programs?
(2) Did HEIs develop a curriculum that meets the needs of local industry and, in so doing, improve local employment opportunities sufficiently to reduce outmigration?
(3) Do HEIs help to attract science and technology companies to the local area? Are households attracted to the area because of cultural and leisure activities provided by the HEI?
(4) How large are the local economic impacts of HEIs associated with their demand for labor, goods, and other resources?
(5) How successful are HEIs at transferring knowledge to local industry through their graduates, spin-offs, and other modes of technology transfer?
(6) What are the migration patterns of highly qualified graduates from HEIs?
(7) How specifically do HEIs contribute to private sector innovation?
(8) What role can HEIs play in the development of regional innovation systems?
Selected Econometric and Case Studies
Abel, J. and R. Deitz, ”Do Colleges and Universities Increase Their Region’s Human Capital?” Journal of Economic Geography 12 (May 2012): 667-691.
The authors use a regression analysis of pooled cross-section data for 283 U.S. metro areas to test whether local human capital levels (as measured by the percent of the working-age population that has a college degree) are influenced by the level of higher education activities in the local area. They find a positive but small effect of local degree production (degrees awarded per 100 members of the working-age population) on local human capital levels, which is consistent with the idea that migration plays a large role in determining the spatial distribution of the college-educated population. The authors find a larger and more significant positive effect of the research intensity of local-area educational institutions (as measured by academic R&D expenditures per enrolled student) on human capital levels.
Breznitz, S. The Fountain of Knowledge: The Role of Universities in Economic Development Stanford, CA: Stanford University Press, 2014.
This short book provides a case study of two universities, Yale and Cambridge, and their comparative performances in commercializing biotechnology research during the 1990s. Both are internationally recognized universities with strong faculty but, by any measure, Yale was much more successful in transferring university knowledge to industry. The author reviews several factors that contributed to this outcome including differences in internal processes, historical differences in commitments to developing university-industry relationships, and differences in external or environmental factors. The author considers internal factors to have been crucial, focusing on the entrepreneurial culture of the university (e.g., the extent to which faculty were encouraged to develop relationships with industry), the quality of technology transfer policies, and the quality and stability of technology transfer offices and their administrators.
Goldstein, H. and C. Renault, “Contributions of Universities to Regional Economic Development: A Quasi-Experimental Approach,” Regional Studies 38 (October 2004): 733-746.
The authors examine a cross-section of data on 312 U.S. metro areas over two time periods (1969 to 1986 and 1986 to 1998) to test whether universities significantly contribute to local economic development. The dependent variable in their analysis is the change over the study period in the ratio of metro-area earnings per job to the national average. The authors use four different measures of university research and teaching activity: whether the metro area has a top-50 research university, total research expenditures of universities in the area, total number of degrees awarded, and number of patents received by universities. Over the period from 1969 to 1986, there is no evidence of either research or teaching activity having a significant effect on the local economic performance. In the period from 1986 to 1998, however, both measures of university research activity are found to be significant and positively related to growth in metro-area relative earnings. Differences in impacts between the two periods are attributed to the fact that knowledge-based economic activity became more important to the U.S. economy in the second period, as did interest among universities in serving as agents of local economic development. While university research activity was found to be a statistically significant determinant of relative earnings growth over the latter period, its contribution to variations in economic performance across metro areas was small relative to other variables, such as metro area size (agglomeration) and industry structure.
Kantor, S. and A. Whalley, “Knowledge Spillovers from Research Universities: Evidence from Endowment Value Shocks,” The Review of Economics and Statistics 96 (March 2014): 171-188.
The causal effect of research universities on average labor income in local noneducation industries is tested using data from 1981 to 1996 in 85 U.S. counties that had a population of at least 250,000 people in 1981 and at least one of 135 leading research universities as identified by the Carnegie Classification of Higher Education Institutions. The authors make a serious effort to control for endogeneity and reverse causality issues that can result from the fact that human capital and innovation-intensive industries that have high average earnings are also likely to have relatively large demands for teaching and research services from local universities. The authors deal with these issues by using as an independent variable the (exogenous) variation in university expenditures that arises from differential effects that national stock market changes have on counties because of differences in the endowments of their universities. Variations in the expenditures of research universities are found to have a statistically significant and positive effect on average labor income in local noneducation industries. The effects are much larger in counties with universities that have above-average enrollments of graduate students. The effects on labor income are also greater in industries that are technologically close to the higher education sector as identified by having in their own patents an above-average percentage of patents issued to universities. The findings in the paper suggest that the long-run effects of universities on the local economy may grow over time as the composition of local industry changes to take advantage of knowledge spillovers from local universities.
Kenney, M. and D. Mowery (eds.) Public Universities and Regional Growth: Insights from the University of California Stanford, CA: Stanford University Press, 2014.
This is a collection of six case studies of university-industry relationships involving campuses in the University of California system that led to important technological advances in high-profile tech industries and had important consequences for the location of new industries in the state of California. The six industries and associated UC campuses studied are (1) the microelectronics and semiconductor industries in San Francisco, Los Angeles, and Santa Barbara and the role played in their development by UC campuses in those areas; (2) the role of UC San Francisco and UC San Diego in developing the California biotechnology industry; (3) the history of the electrical engineering department at UC Berkeley and its long and varied history of interactions with Bay Area technology firms; (4) the wireless communication industry in San Diego and its relationship with UC San Diego; (5) the role played by UC Santa Barbara in building a scientific instruments cluster in Santa Barbara; and (6) the relationship between the Napa Valley wine industry and the department of viticulture and enology at UC Davis. Two recurring themes in these studies are stressed by the editors of the volume. First, while patents, licensing, and other formal channels of university technology transfer have been important in some industries (e.g., biotechnology), they play a more modest role in many other cases. Scientific advance and knowledge transfer often occurs through more informal channels such as sabbatical exchange between university and industry scientists, conferences, publications, and industry employment of university-trained scientists. Second, the flow of knowledge between universities and industry is often bidirectional. The expertise and needs of existing firms in an area often influence the direction of local academic research. Many of the case studies contain examples of how university faculty that made important technological advances first came from industry.
Roberts, E., F. Murray, and J. Kim, “Entrepreneurship and Innovation at MIT: Continuing Global Growth and Impact—An Updated Report,” Foundations and Trends in Entrepreneurship 15 (April 2019): 1-55.
This report provides a summary of findings from surveys administered to MIT alumni beginning in 2003 in which information is collected on the entrepreneurial activities of MIT graduates. Alumni are asked whether they started a company (or companies) after they graduated and, if so, how long the company survived and what was its industry classification, employment, and sales. Graduates are also asked whether they had ever filed patents or served on scientific advisory boards of companies. The survey data show a clear upward trend in entrepreneurial activity among graduates. The data indicate that 6.0 companies were founded per 100 alumni during the 1960s, rising to 10.7 in the 1980s, 13.4 in the 2000s, and a projected 18.0 companies per 100 alumni in the 2010s. Surveys similar in design have been created and administered to graduates of Stanford University and universities in China and Israel.
Valero, A. and J. Van Reenen, “The Economic Impact of Universities: Evidence from Across the Globe,” Economics of Education Review 68 (February 2019): 53-67.
The authors report the results of a regression analysis of the effects of the presence and number of universities in a region on per capita GDP using a pooled cross-section data set from 1950 to 2000 that spans 1,500 regions in 78 countries. The authors find that a 10 percent increase in the number of universities is associated with a 0.4 percent increase in regional per capita GDP. The economic impact of a university is largest in the region in which it was founded, but spillovers also are identified in neighboring regions. In seeking to determine the sources of these impacts, the authors conclude that the impacts are not simply demand driven (universities are themselves employers and purchasers of goods and services), but also seem to be the result of contributions universities make to the supply of human capital and to the stock of knowledge in the local region. The results also suggest that a portion of the long-run impact that universities have on per capita GDP occurs because of greater support for democratic institutions in countries with more universities.