Steve Rose may not care what the research tells us, but the research is mounting. Studies by UNC-Chapel Hill, the St. Louis Development Corporation and now the Upjohn Institute for Employment Research all confirm that economic development subsidies just don’t work.
Using data from 47 cities in 33 states over the course of 26 years (1990 to 2015) the Upjohn study confirmed what Show-Me Institute analysts have been arguing for years: economic development incentives do not grow the economy, create jobs, or boost tax revenue. In his 2017 paper, Upjohn author Tim Bartik concludes (page 116),
The existing research on incentives is that in some cases they can affect business location decisions, but that in many cases they are excessively costly and may not have the promised effects. The new research suggests that much of this consensus is justified.
None of this will surprise frequent readers of this blog. Politicians may like attend ribbon-cuttings and crow about creating jobs, but little of this actually pans out. Instead, cities and states end up hollowing out their tax base, collecting ever less for important services such as public schools, libraries and mental health funds. Simply taxing money out of the economy and then turning around and spending it doesn’t grow the economic pie! The new study confirmed as much:
Incentives are still far too broadly provided to many firms that do not pay high wages, do not provide many jobs, and are unlikely to have research spinoffs. Too many incentives excessively sacrifice the long-term tax base of state and local economies. Too many incentives are refundable and without real budget limits.
The 33 states that Upjohn considered account for 92 percent of the U.S. gross domestic product, and the 45 industries it analyzed account for 91 percent of U.S. labor compensation. Kansas City has undertaken its own analysis, of sorts, of its economic development policies, but has hired a trade group of development financiers to do the work. Seriously.
Politicians and those aligned with wealthy developers may not like or care about the research—but it’s becoming increasingly difficult to wave it off and pretend it doesn’t exist.