Richard Florida is a co-founder and editor at large of CityLab and a senior editor at The Atlantic. He is a university professor in the University of Toronto’s School of Cities and Rotman School of Management, and a distinguished fellow at New York University’s Schack Institute of Real Estate and visiting fellow at Florida International University.
A new study exposes the futility of the $45 billion that states spend on economic development incentives.
Even before being sworn in as President, Donald Trump jumped up and down at the chance to showcase the great deal he and Vice-President Mike Pence made to keep a Carrier plant in Indiana. The company pocketed $7 million in tax breaks in exchange for about 800 jobs. But the broad consensus among economists who study the subject is that such business incentives do little to alter the location decisions of companies. In fact, they’re often worse than ineffective—they’re counterproductive. My own take on this site is that they are a useless waste of taxpayer dollars.
Timothy Bartik of the W.E. Upjohn Institute for Employment Research, who is perhaps the leading student of incentives and economic development, has a new report that provides the most detailed assessment of incentives across states and their effects on economic development. His database contributes a useful tool for state and local economic developers and others to take a hard look at incentives, what they cost, and whether they are worth it or not.
Bartik’s fastidious research compiles data on economic development incentives for 45 industries that compose more than 90 percent of U.S. labor compensation in 33 states. That makes up more than 90 percent of U.S. GDP from 1990 to 2015. It collects detailed information on five different types of incentives: job creation tax credits, property tax abatements, investment tax credits, research and development tax credits, and customized job training credits. He compares their impact on three different types of taxes—property taxes, sales taxes, and income taxes. The study includes data from 47 cities within those states; Bartik plans to analyze that data more specifically in a forthcoming report.
Business incentives are costly
Business incentives have boomed in the past several decades, growing by more than three times since the 1990s, though their rate of growth has slowed since 2000. The biggest single type of incentives are job creation tax credits and property tax abatements, which together accounted for more than 70 percent of all incentives given across states.
Bartik arrived at a total figure of $45 billion in 2015 for export-based industries—industries that sell their products and services beyond their local economy. That’s somewhat smaller than previous estimates, which after adjustments to 2015 dollars ranged from $65 billion from incentives expert Ken Thomas in 2005 to $90 billion by The New York Times in 2012. This $45 billion amounts to about 1.5 percent of all industry value-added for traded or exporting industries in the United States, and about 30 percent of what they pay in state and local taxes.
Business incentives are already focused on manufacturing
Trump may be out to restore the greatness of American manufacturing, but business incentives already flow disproportionately to factories. Incentives given to manufacturing firms amount to more than 35 percent of the state and local business taxes paid by manufacturing firms, and amount to nearly 1.6 percent of total value-added in this manufacturing sector. That compares to just 25 percent of average state and local business taxes and 1.3 percent of value-added for traded or exporting businesses outside the manufacturing sector, like for example, computer software.
Business incentives vary widely by state—and so do their effects
There is substantial state-by-state variation in the use of business incentives. (Bartik measures this as a share of total state value-added made up by incentives). New Mexico (4.2 percent), New York (3.5 percent), and Louisiana (3.3 percent) are the states with the highest incentives. Conversely, incentives are lowest in Washington (0.09 percent), Nevada (0.23 percent), Virginia (0.27 percent), Maryland (0.36 percent), and California (0.47 percent).
“Even among adjacent states, business incentives often vary by a factor of 2 or 3 to 1,” Bartik writes. “In many cases, states have higher incentives than nearby states even though economic conditions are no worse and gross business taxes are similar or lower.” Incentives are not the result of economic or fiscal conditions but reflect state politics and past practices. In other words, some states like to give away more money to business than others—for no obvious economic reasons.
Business incentives are ineffective and wasteful
Instead of focusing on highly innovative, high-paying industries that create a lot of jobs, most incentives, Bartik finds, are poorly targeted; there is little rhyme or reason to which firms or which industries get them. As he points out, incentives do not vary in light of the industry characteristics that would predict greater local benefits—obvious ones, like how much R&D they do, how many people they employ, or the wages they pay. So, for example, a 10 percent increase in an industry’s wages predicts just a 3 percent increase in incentives. Incentives track more with a state’s gross taxes than with unemployment rate or wage levels.
Moreover, states shell out too much money up front. They front-load incentives, instead of watching how businesses that are getting the incentives perform over time. Companies are able to take the incentives and can be slow to deliver on jobs or follow-on investments, or in some cases just pull up shop all together. Other states continue to shell out incentives, even after companies fail to meet their own projections for job creation or investment.
Basically, states hand over the money to companies and then do not even follow up to see if they are working or not. As Bartik puts it:
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. States devote relatively few resources to incentives that are services, such as customized job training. Based on past research, such services may be more cost-effective than cash in encouraging local job growth.
The study concludes that there is little connection between the level of incentives a state forks over to business and its economic fortunes. “Incentives do not have a large correlation with a state’s current or past unemployment or income levels or with future economic growth,” Bartik writes. He also notes, however, that these results are preliminary and might change in more full-blown models with more control variables.
These conclusions are in line with a wide body of research on the wastefulness of business incentives. A 2002 study of some 350 companies that received incentives found a negative effect on their ability to create jobs. Companies that received incentives expanded more slowly than others, and the overall effect of incentives was a reduction of 10.5 jobs per establishment. My own research published on this site found virtually no association between business incentives and any measure of economic performance, including wages, incomes, and unemployment.
Ultimately, business incentives do little to alter the locational calculus of most companies. As I wrote here in 2012:
The broad body of evidence on incentives… finds that incentives do not actually cause companies to choose certain locations over others. Rather, companies typically select locations based on factors such as workforce, proximity to markets, and access to qualified suppliers, and then pit jurisdictions against one another to extract tax benefits and other incentives. A 2011 Lincoln Institute of Land Policy study found property tax incentives to be counterproductive, being all too frequently given to companies that would have chosen the same location anyway. So instead of creating new jobs or spurring employment, the main effect of incentives is simply to deplete a community's tax base. Since poorer states and communities are more likely to use incentives in the first place, the end result is to undermine the resources and revenues of the places that can least afford it.
But, thanks to Bartik’s tool, you don’t have to take my word for it.
You can also use his models to assess the savings that would come from eliminating or reforming incentives. For example, limiting incentives to just one year would reduce the overall cost of incentives by roughly two-thirds. Getting rid of refundability—the ability of businesses to receive incentives even if they have no state corporate income tax liability—would reduce the costs of incentives by a third.
Let’s hope the information and transparency his database provides can help put an end to such an ineffective use of public resources.