Commentary

Incentive Programs Fail: Evidence is All Too Clear

Center Scholar Speaks to Tax Group in Florida

Editor's Note: These remarks were originally given before the Florida Tax Watch Spring Board Meeting on May 26, 2017.

Thank you. It’s a pleasure to be here today to talk about economic development. There are few areas of research where I find such widespread agreement in academic and other studies. In short, state and local “development” programs are ineffective and expensive. They don’t work, they’re unfair to those who pay full freight, and cost billions of dollars that could be better used elsewhere.

The modern “war between the states” over jobs has been active since the Great Depression. Fifty states run some 1,829 incentive programs designed to create jobs and wealth, according to the Council for Community and Economic Research and economists have written hundreds of studies trying to measure their effectiveness.

A “meta” review of the economic development literature was completed by economists from the University of Iowa in 2004. Their study was a literature review of the literature reviews in the field. Titled, “The Failures of Economic Development Incentives,” the professors argue that assertions that these programs improve economic growth or lead to net positive fiscal benefits for governments are probably false.

Since this study was released, I found nothing in the literature to suggest anything has changed. In fact, if anything, the frequency and clarity of published evidence against such programs has only increased. Here’s some of what we’ve learned recently about different economic development programs:

  1. State tourism promotion: My 2016 co-authored study looked at 39 years of data from 48 states. We found that for every $1 million increase in promotion spending, there was an increase of just $20,000 in extra economic activity in the average state’s hotel and motel industry. Other sectors fared worse.
  2. Subsidies and loans: Another 2016 paper from Texas professor examined some high-profile jobs and capital-creation programs in Maryland and Virginia and found “no impact on growth or job creation.”
  3. Sports facilities: An April 2016 academic paper found the opening of a new facility does not increase new businesses, though it registers a slight uptick in additional workers. Most other academic papers on professional sports stadiums and teams also show zero to negative impacts.
  4. Tax Increment Financing: A 2015 study from Ball State University argued that TIFs in Indiana were “associated with less employment, less taxable income and slightly higher tax rates.”
  5. Tax Credits: There have been five scholarly studies since 2005 on Michigan’s huge refundable tax credit program called MEGA, which offered up billions in incentives during its lifespan. Four found an impact from zero to negative. One found a positive impact, but it was tiny.

The research is mixed but leans negative. The record gets worse when you consider opportunity cost. That is, if billions weren’t spent trying to create economic development, might they be used more effectively with across-the-board tax cuts or infrastructure improvements? A 2000 tally by professor Kenny Thomas said the annual value of incentives offered reach as much as $50 billion.

Why do incentive programs fail? I suggest three big reasons. 1) You can’t give anyone something you don’t first take from someone else. 2) They encourage rent seeking (a growth retardant) and distort decision-making, turning some market entrepreneurs into political ones. 3) The opportunity cost is high. Cutting taxes for a few politically well-connected firms makes it harder to cut taxes for everyone.

A better way to development is to take the “fair field and no favors” approach. People and markets were creating jobs and wealth long before government officials thought it should be their responsibility.

Michigan Capitol Confidential is the news source produced by the Mackinac Center for Public Policy. Michigan Capitol Confidential reports with a free-market news perspective.