The oft-stated goal of government economic development incentive programs is to create jobs. Yet policymakers and program advocates seldom conduct careful analysis of these programs to determine how well they work. Given how many states use these programs liberally yet fail to even keep up with national job-creation rates, one suspects these efforts are largely ineffective. So why do these programs continue to multiply?

In this new book, Nathan Jensen (University of Texas, Austin) and Edmund Malesky (Duke University) advance original arguments that explain the ubiquity of these incentives and offer technically feasible and politically practical reforms to rein in these programs. The issue is topical: in recent decades government pandering with incentives has grown in response to threats by sports teams, movie companies, and manufacturing firms to relocate their operations.

Zero-sum games / Government incentives can be discretionary (“deal-closing funds”) or statutory. Oftentimes they are targeted by policymakers to assist only a small number of firms in the vast universe of enterprise—sometimes even a single firm (think of the goodies different states and localities recently offered for Amazon’s HQ2). These incentives have grown in size, with at least 17 single-firm state packages eclipsing the $100 million mark in recent decades. Examples include South Carolina’s $130 million-plus offer to attract a BMW plant (1992) and Georgia’s $258 million to land a Kia plant (2006). Less well known is the aggressiveness of city and county programs. One example Jensen and Malesky share is Lenoir, NC offering a quarter-billion dollars in incentives (mainly tax breaks) over 30 years to woo a Google server farm. That equals roughly $1 million for each center employee.

The economic inefficiency of these programs is a recurring theme in the book. The authors tell the story of an “incentive war” between Kansas and Missouri that resulted in employers of 3,200 workers moving from the east side of the Kansas City metroplex to the west, and employers of 2,800 workers moving from the west to the east, which the authors term “the very essence of a zero-sum game.” There is no clear evidence that incentives create net public benefit, however, as numerous academic studies have shown them to be ineffective and redundant.

Why use such a flawed policy? And why do politicians herald these incentives rather than hide them? Jensen and Malesky present “puzzles,” building on Gordon Tullock’s insight that voter ignorance about incentives is rational given the absence of knowledge of their true costs. Incentives are the perfect “pandering tool” for politicians engaged in “credit-claiming and blame-avoiding roles” with voters. The “consistent use of incentives in press releases by governors’ offices around the country and in campaign materials,” they write, “suggests that politicians see the use of incentives as an asset, not a liability.”

Given that, we must conclude that the main purpose of these incentives is for political gain. Apparently, politicians use incentives to signal alignment with voter interests, assuming that voters have imperfect knowledge of incentives’ importance. The counterfactual is unobserved: voters don’t know that most incentives are given to firms that are already planning investments. They may lower firms’ costs but they seldom create jobs.

Pandering / Jensen and Malesky develop a “theory of pandering” to explain this bad public policy. They begin by rejecting the argument, commonly found in the popular press, that incentives are driven by corruption or as legal means to obtain campaign contributions. They dismiss the corruption charge because “politicians do not hide their allocations of incentives to firms. [This is] far from what we would expect from under-the-table exchanges of campaign contributions for financial support.”

Indeed, elections provide politicians with an incentive to publicize incentives. Voters prefer incumbents who take credit for creating jobs over those opposed to incentives. Voters also prefer incumbents who try to attract jobs with incentives even if they fail, instead of critics who vow to eschew the practice. Politicians exploit their “information advantage” by providing too many and too generous incentives. This “information asymmetry between voters, politicians, and firms” can lead incumbents to use incentives to take credit or reduce the blame for economic outcomes. Politicians will use incentives, regardless of investors, if voters believe they are effective.

Consider Donald Trump’s highly publicized move in late 2016 to retain manufacturing jobs in two counties in Indiana. The authors write, “From the start, there was some fuzziness in the numbers” of jobs ostensibly created or saved by Trump and Indiana officials’ efforts. In fact, Bureau of Labor Statistics data show total employment in one county declined after the announcement, while the other county experienced an employment increase of only 0.3% versus the national average of 2.5%. Capital movement influenced by globalization, they write, “can provide politicians with opportunities to pander to the public and take credit for new investment. … Rather than making domestic politics irrelevant, globalization can lead to increased political activity.” Incentives give politicians reason to take credit.

Voters prefer candidates who try to attract jobs with incentives, even if they fail, instead of candidates who vow to eschew this practice.

This process is also visible at the local level, where cities with mayor–council systems offer more generous incentives but are less likely to mandate performance requirements and benefit–cost analyses. The authors find mayors “are more prone to use incentives for electoral gain.” Incumbents facing electoral pressure are more likely to use incentives than city managers shielded from the ballot.

Regressive incentives / How do local politicians pay for these programs? Oftentimes, regressive sales and excise taxes shift burdens onto the poorest taxpayers. Call it “economic development by sales tax,” bad public policy that drives economic inequality. Incentives create a reverse Robin Hood effect, as wealth is transferred from the poor and middle class to wealthy residents. For instance, Ferguson, MO politicians filled their budget hole from funding new incentives by increasing the revenue from fines and penalties, fueling racial acrimony through increased policing. Critics of incentives should explore this common ground with citizens who are troubled by inequality and injustice.

Incentives’ use is not restricted to western-style governments. Authoritarian regimes, especially those linked with meritocratic performance at the local level, are associated with higher levels of use of these programs than their democratic counterparts. Incentives are more likely to be provided to foreign investors if the regime has strong protections for meritocratic promotion for sub-national leaders. Central government elites want gross domestic product, government revenue, and employment growth, and are agnostic about how those gains are achieved. The authors term this phenomenon “upward pandering” and note it exists in single-party states with quasi-meritocratic institutions until the point when officials are no longer eligible for promotion. Aging Vietnamese officials, Jensen and Malesky observe, abandoned the use of tax incentives once they became ineligible for promotion. Interestingly, “personalist regimes” such as Russia under President Vladimir Putin offer far less in incentives because loyalty trumps performance.

Jensen and Malesky pose a series of questions about incentives that should be answered by every politician contemplating their use. Are they worth the cost? Are they effective at attracting or retaining investments? Can governments target firms and pick winners? Do they generate jobs and are those jobs worth the cost? Are the incentives the only option for generating economic development? Most incentive-happy politicians answer in the affirmative.

One recent exception, to some extent, was Michigan governor Rick Snyder. As a businessman turned candidate in 2010, he criticized tax incentives during his election campaign, though he did simply relabel some of the incentives “grants” when he continued them once he took office. But he also signed a 2012 executive order dissolving the Michigan Economic Growth Authority (MEGA), established in 1995 over the objections of the Mackinac Center, a market-based think tank, and state legislators. (Disclosure: I was among those critics.) MEGA’s demise is significant for two reasons. First, BLS records show total Michigan employment declined (4,450,800 to 3,893,700) under Snyder’s predecessor, Jennifer Granholm, an incentives proponent. Since the U.S. economy began expanding in June 2009, a period that largely coincides with Snyder’s tenure, Michigan has been one of the 17 states with a job-creation rate above the U.S. average. Invert those circumstances by postulating a state that does not use incentives and records negative jobs growth, or a state that ends a program like MEGA and trails the nation in jobs creation, and what would political commentators say? Incentive critics should always present their case to the public using the easiest-understood argument: lack of jobs means incentive programs should end.

Researchers, good-government activists, and policymakers can use other strategies to challenge these programs. State-level researchers focus on local programs. One example: Jacob Bundrick of the Arkansas Center for Research in Economics found “no evidence that the [state’s] Quick Action Closing Fund (a discretionary program) provides meaningful increases in employment or net business establishments at the county level.” Clawbacks allow tax dollars to be recouped when programs do not meet political promises. Written performance criteria provide greater transparency. GASB Statement No. 77 (2015) requires disclosure of the true cost of tax abatements.

Politicians’ embrace of incentives means researchers should embrace this book for its insights about political pandering.