Recent Research: Do Jobs Follow People, or People Follow Jobs?

When General Electric (GE) announced earlier this month that it was moving to downtown Boston’s Seaport District, significant attention was paid to the generous incentive package handed to the company by Massachusetts. Ultimately, however, it was the human capital and innovative talent in the city that lured GE, according to the Chairman and CEO Jeff Immelt’s comments announcing the move. While conventional thinking suggests that attracting new businesses is a path to population and economic growth, new research suggests that this may not be the case. 

In What Comes First, People or Jobs: Evidence and Lessons for Indiana, Michael Hicks and Dagney Faulk, two economists from Ball State University’s Center for Business and Economic Research, seek to understand the long-term differences between whether or not people followed jobs or jobs followed people. To do so, the authors use data from the Census and Bureau of Labor Statistics to examine population and employment growth in Indiana’s 92 counties in the decades of the 1970s and the 2000s, controlling for variables such as educational attainment, initial population, urban and rural influences, natural amenities, income inequality, intergenerational mobility, size of local government, and other spatial spillover effects.

The findings in their models point to several important features of Indiana’s economy, each with broad policy implications. The authors find evidence of slow conditional convergence. In other words, as long as educational attainment, amenities, and fiscal differences remain in Indiana counties, population will also remain dissimilar.  In the 1970s, Hicks and Faulk found that workers migrated to jobs, while jobs also moved to workers. This type of thinking informed the traditional business attraction policy seen in the state throughout the 20^th century. During their study of Indiana in the 2000s, however, the authors found that the movement of workers to jobs ceased to play a statistically meaningful role in population dynamics. Instead, jobs relocated to be near people. Population growth within a county was no longer dependent upon job growth within the county but on factors such as educational attainment or local amenities. 

Hicks and Faulk note that traditional economic development as practiced at the county level in Indiana to the tune of $1.5 billion per year is largely focused on attracting employment growth, manifesting itself in policies designed such as property tax abatements and other incentive packages. While business attraction may be a useful policy adjunct at the appropriate regional level, the authors’ analysis suggests it does not grow communities at the county or sub-county level. Instead, places should focus on improvements to quality of place and education. 

Brian Whitacre, David Shideler, and Randi Williams, three economists from Oklahoma State University, find similar conclusions about business attraction incentives in a February 2016 Economic Development Quarterly article. In Do Incentives Programs Cause Growth? The Case of The Oklahoma Quality Jobs Program and Community-level Economic Growth, the authors analyze a state incentive program that provides cash payments, rather than tax incentives, to firms for up to five percent of newly created gross taxable payroll in order to encourage firms to locate or expand within the state.

A previous study in 2004 on the Quality Jobs Program made an assumption that a company’s location or expansion decision was based solely on the incentive program.  Largely a result of this assumption, the previous study concluded that the program contributed to Oklahoma’s employment base beyond what would have occurred naturally. This recent study, however, takes an alternative approach and develops a “counterfactual” argument by matching similar communities that differ only by whether or not they participated in the Quality Jobs Program.

Employing both multivariate regression and mixed-pair analysis techniques, Whitacre, Shideler, and Williams conclude that economic growth between 1990 and 2005 was not statistically different between Oklahoma communities with businesses participating in the Quality Jobs program and those that were not participating. The authors’ analysis finds no evidence that cities with at least one business that obtained Quality Jobs funding grew any faster in terms of income, population, or house value than did non-recipient communities. Written 10 years later, these findings are considerably less positive than those in the previous studies.

Taken together, these studies highlight some of the shortcomings of traditional economic development policy. While the “smokestack chasing” model worked throughout the majority of the 20^th century, conventional efforts to attract businesses are no longer yielding the same growth in population or in the local economy. In his CityLab article about the General Electric headquarters move, Richard Florida comments on these limitations, noting, “Ultimately, all these incentives really do is take money out of the pockets of Boston and Massachusetts taxpayers – money that could and should be used to reduce poverty and improve education in the city and state’s many disadvantaged neighborhoods.” Applying the findings of Hicks and Faulk to the nation at large, this type of policy could fare better at improving the potential population and economic growth outcomes in communities than traditional business attraction incentive packages.