10.09.2012
Economic Trends
Regional Differences in Science and Engineering Schooling and Employment
Differences in human capital across regions are associated with differences in economic performance. For example, many studies have documented that regions with higher human capital, typically measured in terms of educational attainment, experience higher income growth. The correlation is attributed to many channels, but key among them is the view that more educated locations are more innovative and can take better advantage of new technologies.
Of particular interest in some policy quarters is the ability of a region to assemble a college-educated, technically sophisticated workforce. At the state level, one measure of this capacity is the production of baccalaureates with science or engineering (S&E) degrees in a state. The number of bachelor’s degrees (BAs) conferred in science and engineering in the United States in 2009 is 17 per 1,000 individuals aged 18-24 years old. This compares to the roughly 53 BAs per 1,000 across all degree fields. Fourth district states produce S&E graduates across nearly the full range of the distribution, with Kentucky on the lower end and Pennsylvania on the upper end.
States on the lower end of the distribution produce about 10 S&E degrees per 1,000 individuals aged 18-24, while on the upper end degree production rises to the mid-to-high 20s per 1000. Some of the difference in production rates is due to differences in the rate at which young people attend 4-year colleges across states, and some of the difference reflects the fact that certain states are net exporters or importers of college services. For example, states in the upper end of the distribution such as Iowa, Massachusetts, Pennsylvania, and Rhode Island serve significant numbers of students that come from out of state, as does the District of Columbia (Table 234: Digest of Education Statistics, National Center for Education Statistics).
A natural question is whether S&E degree production in a region is related to the development of high technology industries in a region? The National Science Foundation (NSF) identified 46 industries that require a substantial fraction of the workforce to have “in-depth knowledge” of science, engineering, or mathematics. These industries include information technologies, biotechnology and pharmaceuticals, scientific equipment and instrument manufacturing, computer system design, engineering services, and aerospace manufacturing and design, among others. The location of such industries is not uniform across the country. As a share of a state’s employment, these high-tech industries vary from a low of around 6 percent to a high of around 15 percent across the states. States with particularly large shares of high-technology employment include Maryland, Massachusetts, New Jersey, Virginia, and Washington.
The relationship between S&E degree production and high-technology employment shares is positive, though not particularly tight. States that produced a high number of S&E degrees in the 1990s had, on average, somewhat higher employment shares in high-technology industries in 2008. The relatively weak relationship is not too surprising for a number of reasons. First, as discussed above, some states are net exporters (or importers) of college educations. Export states will confer an above-average number of degrees, but it is likely some of these students will return to their home states or move to another state. Second, the labor markets for many highly trained scientists and engineers are national in scope, reducing the link between local supply and local demand. Highly educated workers are quite mobile. Third, many S&E graduates will work in industries (and in occupations) that require strong science backgrounds but that are not high-technology industries, as defined by the NSF. An example is medical personnel, as hospital and medical service industries are not included as high technology industries. Another example is higher education. Indeed, if one examines the relationship between S&E degree production in the 1990s and the share of the population aged 30-44 in 2010 with a science or engineering degree regardless of the industry they work in, the relationship tightens up. The point is that S&E degree production in a state is related to the future science and engineering talent in the workforce, though by no means is there a one-to-one correspondence.
Still, one needs to be cautious in interpreting the relationship between degree production and high technology employment. On the one hand, one might want to infer that a higher supply of S&E graduates in a state fosters the development of high-technology industries in the location. This is a traditional workforce development story. Investments in the training and development of particular workforce skills allow for the growth of industries that utilize such labor inputs. On the other hand, it is also plausible that higher education institutions and students in a region are simply responding to shifts in demand or expected shifts in demand by high-technology firms. In this case, it is demand for high-technology workers that leads to a rise in the S&E graduates. Looking at simple correlations cannot reveal the underlying drivers of the relationship. That said, it would not be surprising if it is a combination of both stories that form the basis of the observed relationship.