Are College Towns Recession Proof

• Archives

Introduction

Research universities bring a variety of benefits to their surrounding communities such as a supply of highly skilled graduates, innovation via research and development activities, and a large stable employer. But do research universities make their communities resilient to economic downturns? In a recent working paper, we attempt to answer this question by examining the unemployment rates for counties containing state flagship universities over the past three national recessions.

We find that state flagship universities do not necessarily make a community immune to recessions. Although we find that counties containing state flagship universities had significantly lower growth in their unemployment rates during the great recession than control counties, they faced significantly higher growth in their unemployment rates during the COVID-19 recession, and little difference during the dot-com recession. It seems clear, therefore, that the reason for the recession matters in how much resilience universities can provide. In general, we believe that more stable employment and thus stable demand for local goods and services by faculty, staff, and students drove the resiliency effect observed during the Great Recession. However, the additional stability was absent during the COVID-19 recession when most university campuses were closed to students due to social distancing restrictions.

Research Universities and Regional Economies

Research universities are postsecondary institutions that emphasize knowledge production as a core component of their activities through the academic research of their faculty and the training of doctoral students across various disciplines. Research suggests that regional economies reap benefits through place-based education and research and development expenditures. The United States has observed prominent examples of this through the high-tech industry clusters in Silicon Valley (e.g., Stanford and University of California, Berkeley) and Pittsburgh (e.g., Carnegie Mellon and University of Pittsburgh) that were fostered by the shared research efforts and hiring of skilled graduates from local universities.

Not only might research universities offer the potential for economic growth, but they may also guard against decline. Recent research by Luisa Gagliardi, Enrico Moretti and Michel Serafinelli suggests that higher shares of college graduates provided resiliency for manufacturing-dependent Rust Belt regions during manufacturing's decline in the richest industrialized nations over the second half of the 20th century. In addition, Greg Howard, Russell Weinstein, Yuhao Yang find that a regional university roughly offsets the negative effects of manufacturing exposure in part through local consumption by faculty and students. Is it possible that universities provided a cushion against more recent destabilizing events, such as the dot-com recession, the Great Recession, or the COVID-19 pandemic?

Evaluating Flagship College Towns Over the Business Cycle

We investigate whether large research universities recession-proof their regions by examining the unemployment rates of U.S. counties that contain state flagship universities. A state flagship university is the leading institution within a network of state public universities and tends to be the oldest, most selective, highest enrolled, and most research-intensive public university within a state. Examples of these include the University of North Carolina at Chapel Hill, University of Virginia, and University of Maryland, College Park. Unlike many private research universities, which were founded by benefactors or religious organizations in major U.S. cities (e.g., University of Chicago, Johns Hopkins, and Vanderbilt), most flagship universities are located outside of the nation's largest metropolitan areas.

We compare unemployment rates in these flagship counties to a control group of U.S. counties that do not include research-intensive universities but have similar industrial composition, size, and pre-recession unemployment rates. We use difference-in-differences techniques (see the working paper for a more detailed explanation) to evaluate the performance of flagship counties by comparing the difference in outcomes between the treated and control groups before and after each of the past three recessions (Dot-Com, Great Recession, and COVID-19).

Research Universities Do Not Provide Consistent Protection From Recessions

We analyze how flagship counties performed against controls over the past three U.S. recessions. The dot-com recession was a relatively short and shallow downturn in the U.S. economy from March through November 2001. The Great Recession was a more severe global economic and financial downturn from December 2007 through June 2009, which was triggered by the subprime mortgage crisis. Lastly, the COVID-19 recession from February to April 2020 was caused by the interaction of virus contagion fears and statutory stay-at-home policies that forced many parts of the economy to shut down.

Using unemployment rate data from the Bureau of Labor Statistics, we find that there was little difference in the unemployment rates between flagship counties and counties without flagship universities after the dot-com recession. However, unemployment rates in flagship counties were more than 0.5 percentage points lower after the Great Recession. On the other hand, the COVID-19-induced recession was entirely different. After that recession, in May 2020, flagship counties had unemployment rates more than 0.5 percentage points higher.

We believe that any recession-proofing effect from flagship universities on their regions is driven by stable consumption demand for local goods and services. Our small and insignificant finding for flagship counties during the dot-com recession is not surprising given that U.S. consumption only slowed and never declined, so there was no negative consumption shock which could be absorbed. However, the Great Recession was characterized by a broad decline across consumption categories, which was likely absorbed through local consumption by flagship university students whose enrollment tends to be countercyclical. Conversely, counties that are heavily reliant on higher education were badly affected by the COVID-19 recession, as the absence of students further compounded the negative consumption shock from the national business cycle downturn.

West Virginia University's Impact on Monongalia County

Monongalia County, home to West Virginia University, is an example of a flagship county in the Fifth District. Its unemployment rate has been consistently lower than many other counties in West Virginia. (See Figure 2 below.) At the start of the 2001 recession, it had an unemployment rate of 4.1 percent compared to 5.4 percent in West Virginia as a whole. Similarly, at the beginning of the Great Recession in 2007, its unemployment rate was 3.3 percent (compared to 4.8 percent in West Virginia), and in 2020, it was 6.3 percent (compared to 8.3 percent in West Virginia).

By creating a "frankensteined" Monongalia County (called a synthetic control), we can test what the unemployment rate for the county would have been at the end of each of these recessions had it not contained West Virginia University. This synthetic control is generated by weighting West Virginia's other counties so that it has an identical unemployment rate to Monongalia County before each of the past three recessions.

We see very little difference between Monongalia and its synthetic control after the 2001 recession (Figure 3). Then, the unemployment rate of the synthetic control closely tracks Monongalia County's until 2008, when it rises relative to Monongalia during and after the Great Recession (Figure 4). This exercise suggests that Monongalia County would have had an unemployment rate of at least one percentage point higher in the aftermath of the recession if it did not contain West Virginia University. In the third synthetic control chart, we see that the unemployment rate in Monongalia County is higher than its synthetic counterpart, indicating that it would have had a slightly lower unemployment rate after the COVID-19 recession had it not housed the state flagship university (Figure 5).

Figure 3: Synthetic Control for Monongalia County, West Virginia, for the Dot-Com Recession

Note: The vertical line represents recession start.
Source: Bureau of Labor Statistics, authors' calculations

Figure 4: Synthetic Control for Monongalia County, West Virginia, for the Great Recession

Note: The vertical line represents recession start.
Source: Bureau of Labor Statistics, authors' calculations

Figure 5: Synthetic Control for Monongalia County, West Virginia, for the COVID-19 Recession

Note: The vertical line represents recession start.
Source: Bureau of Labor Statistics, authors' calculations

Takeaways

Universities cannot consistently shield their regions from recessions. Although counties with state flagship universities did not see their unemployment rate increase as much as their non-flagship counterparts during the Great Recession, they experienced higher unemployment rates after the COVID-19 recession. These results support the hypothesis that stable demand for local goods and services drives resiliency, which was absent during the COVID-19 recession when most university campuses were closed to students. Understanding how anchor institutions (e.g., universities, hospitals) affect their local labor markets can help us better understand the growth and dynamics of local and regional economies.

Views expressed are those of the author(s) and do not necessarily reflect those of the Federal Reserve Bank of Richmond or the Federal Reserve System.