Asheville – Think tankers used to look on incredulously at government officials bragging about the silver-bullet power of economic development incentives. To the layman, growing the economy by taxing money out of the free market to redistribute to wealthy business owners makes as much sense as eating extra cookies to lose weight. For people who need statistical analysis, Nathan Jensen of the University of Texas at Austin and Edmund Malesky at Duke University give assurance that the bulk of scholarly research concludes economic development incentives are “uncertain, expensive, distortionary, and usually ineffective.” This is even more pointed in that it is bizarre for a preponderance of economists to agree on anything.
That a seven-member city council could not have more wisdom than Friedrich Hayek’s millions of decision-makers responding to and with price signals in real time, otherwise known as Adam Smith’s invisible hand, was one of those things one had to be in government to not understand. The story is told how the British parliament authorized £152 million in incentives for BMW when that company bluffed about leaving but divested a year later, after authorizing £162.5 million for Chrysler, which had also bluffed and then divested two years later.
Jensen and Malesky realized there had to be something more than meets the eye in what for all practical purposes was a race to the bottom for politicians. The way different governments try to outbid each other for incentives that don’t work certainly looked like a prisoner’s dilemma. Their investigations involved rigorous statistical analysis of available data as well as surveys of large samples to test various hypotheses.
First, they asked the obvious: Are corporate recipients of economic development incentives contributing to awarding politicians’ campaigns? That was an easy answer that only required searching open records of campaign contributions. Using former Texas governor Rick Perry as a case study, Jensen and Malesky found no evidence of quid pro quo but added they didn’t have the time to search the many creative ways, legal and illegal, for contributing indirectly.
After all, the notion that politicians were accepting bribes would not be consistent with the big announcements, the groundbreaking ceremonies, and the ribbon-cuttings, either. Then, again, it doesn’t make sense that they should brag so much to their constituents about doing something most economists agree is profitable, at least in the short-term, for the recipient business but net negative for the economy as a whole. Jensen and Malesky point to various court cases and forms of legislation, such as those that require the identification of revenue sources for incentives, that betray politicians are also aware that “corporate welfare” is a losing prospect.
So, the next hypothesis they tested was that politicians were leveraging asymmetrical knowledge to improve their chances of re-election. Whereas it is the politicians’ business to know about the research, lawsuits, and legislation; the average, rational voter doesn’t have the time to bother. So, politicians need only accentuate the positive, eliminate the negative to create an illusion for voter buy-in.
Part of their test of this hypothesis involved surveying thousands of Americans to see how an explicitly hypothetical incentives package would influence their likelihood to re-elect their governor. They found people were more likely not only to vote for the incumbent if he successfully wooed a large employer but also if he had unsuccessfully wooed them. The latter scenario explains why governments will bid for companies that are not likely to locate in their jurisdictions, with the adverse effect of raising bids so the “winner” pays more. The statistical analyses can be found in their book, Incentives to Pander: How Politicians Use Corporate Welfare for Political Gain.
What avoids mention in the grandstanding is that, even if incentivized corporations meet their targets, when opportunity costs, wear and tear on infrastructure, and public service utilization are taken into effect, the deals net overall losses for the awarding jurisdictions. To pay for the incentives, then, taxes, fines, and fees are increased and/or government services are cut. The authors found raising sales taxes and cutting education to be the go-to sources of revenue replenishment, both of which are disproportionately burdensome on the poor.
Police tensions in Ferguson, they argue, were easily ignited because the town had been, among other things, trying to recoup incentives losses from its tax increment financing (TIF) districts through police fines. Chicago and California both toyed with the idea of canceling all their TIFs, and California, championed largely by the California Teachers’ Association, followed through. Now, many states require school board approval and give school districts grants to compensate for TIF losses – which, ironically, works as a perverse incentive motivating teachers’ unions to support more TIFs.
As for other reforms, the authors admit that clawbacks can, “police fraud or legitimate poor performance of investments, but they do not address that many incentive dollars are duplicative, wasted on firms that would have come anyway.” What’s more, their analysis shows that of those firms that are not weeded out with clawbacks, only one job is created per $1 million in incentives, when compared to job creation by un-incentivized companies.
A more effective reform would be greater transparency. The authors saw hopeful returns when they administered the survey to another sample population, but added, “As with all government policies, incentives come with trade-offs. The money dedicated to incentives would not be available for government spending or tax cuts for individuals.”