Perhaps there’s no government policy or program that's as widely reviled, yet universally pursued, as tax incentives.

Study after study demonstrates that when states and cities give out tax breaks to companies looking to relocate or expand, they typically get very limited bang for their bucks, if any. Yet such incentives remain central to development strategies in most jurisdictions.

A study published last summer found that three-quarters of local economic development dollars are devoted to tax incentives, with the amount spent on them tripling since the 1990s.

“There’s still a lot of economic development brainpower devoted to the Amazon HQ2, let’s-win-the-lottery approach,” says John Lettieri, president and CEO of the Economic Innovation Group, a research and advocacy firm.

The big, swing-for-the-fences deals — the pursuit of Amazon HQ2, Wisconsin’s $4 billion deal with Foxconn, the Taiwanese electronics manufacturer — hardly ever pay off as promised. But run-of-the-mill incentive packages often turn out to be mistakes, as well.

A forthcoming study by economists at Columbia and Princeton found that the average firm-specific subsidy is $160 million, presented in hopes of creating 1,500 jobs. That’s $107,000 per job. And not all the promised jobs pan out.

To be fair, the study found that, on average, the companies that receive tax incentives do create the number of jobs specified in their deals. But that’s it. The study found that there’s generally no indirect benefit. That is to say, attracting specific companies does not give any significant boost to the broader local economy.

“It’s not exactly that there are never spillover effects, but on average we don’t see any,” says Cailin Slattery, a Columbia Business School economist and co-author of the study. “Sometimes there are deals that have broader economic effects, but there’s not the energy you expect to see.”

Plenty of other studies have found that tax incentives go to companies that would have created the jobs anyway, or are more likely to move for other reasons, such as a skilled local workforce, infrastructure or access to markets, than they are for a tax break.

But once a firm has narrowed down its choices to, say, three locations, those places know they’re in a bidding war. None of them wants to lose out on a deal because they refuse to ante up an incentive. And companies know they can play jurisdictions off each other. “I don’t see that there’s going to be any kind of disarmament, for want of a better term, by state and local governments,” says Randy Bauer, a former Iowa state budget director.

If incentives are inevitable, state and local officials should think more carefully about what actually works.

Why Incentives Are Irresistible

Promoting the economy and creating jobs is seen as not only part of the portfolio but an essential measure of success for elected officials these days. Nothing looks more like a win than being able to announce the creation of 500 new jobs, or 5,000.

Building up clusters of homegrown industries is likely to pay off better in the long run, but who wants to think about the long run?

“Building homegrown industries and diversifying your economy works completely against political timelines,” Lettieri says. “They want the ribbon cutting today and 1,000 jobs a lot more than the founder of a company announcing they’re hiring five people, and it might grow to 1,000 jobs over time.”

There are definitely factors that lay the groundwork for entrepreneurial activity and economic growth. Successful places often create collaborative cultures between local government, industry and philanthropic and educational institutions. Having a strong research university is a huge plus, but even community colleges can play an important role.

In South Carolina, for example, attracting a BMW plant back in 1994 turned out be a smart use of incentives in part because complementary investments were made in community colleges for training. The state has become a leading auto production center over the past quarter-century, with some 72,000 jobs now in the sector.

Too often, states and cities tend to jump on the same bandwagons at the same time. But no matter how much money they’ve lined up to give away to movie producers, for example, there’s only so much film, television and video production to go around.

It’s better to target incentives to companies that can build on already existing local strengths. The Oklahoma Department of Commerce has a program devoted to building the aerospace industry, using incentives to grow an existing industry known for generating good-paying jobs. In Utah, incentive dollars are largely targeted to companies in six clusters, including information technology, life sciences and outdoor recreation.

“You really do need a strategy that’s multifaceted,” Bauer says. “You do need to structure incentives so they’re not just wide open.”

Taking Care with Contracts

Utah has an incentive program based on rebates. Once a company meets the criteria for receiving incentives — creating a given number of jobs that pay a minimum of 110 percent of the county average in relevant job categories, for instance — it gets a rebate. The average over the past year has been 20 percent of a company’s overall tax bill dating back seven years. About two-thirds of the rebates have gone to companies that were already based in the state but have expanded.

Offering rebates is a good way to ensure that promises are kept. States and localities are often left without recourse if they devote lots of money to companies that don’t deliver. Too often, they either forget to bake protections in, or feel they can’t make too many demands at the negotiating table.

“So many of the problems and the bad stories come about because internal administrative processes are not where they’re supposed to be,” Bauer says. “Legislators have good intentions on a lot of this stuff, but they often think of incentives as an event, something they’re going to accomplish but then not going to think about that much.”

Bauer notes that many places have no choice but to play the incentives game. Economic activity is becoming increasingly concentrated in a small number of major metropolitan areas. Last year, 78 percent of venture capital funding went to just three states, California, Massachusetts and New York. The rest are left scrambling for scraps. Forty states each captured less than 1 percent of the national share of venture capital dollars.

“In many parts of the country, if you’re going to get transformative investments in many of our midmarket cities, there’s going to need to be government investment,” says Bruce Katz, director of the Nowak Metro Finance Lab at Drexel University.

Incentives are often given out when places are desperate, hungry to attract jobs at essentially any price. It’s better to invest in the future when you’re not desperate, Lettieri says. “In the good times, you have to think about what happens when the good times end,” he says.

It’s completely understandable that places that need to stimulate growth will do so in what appears to be the most expedient way possible. Guaranteeing that a company will come to town through incentives is a lot simpler than patiently fostering growth by building on local strengths.

But the places that are hungriest may find themselves priced out of the game. Slattery, the Columbia economist, notes that the places with lower average wages end up paying higher subsidies per promised job. Because they’re less attractive, they have to pay firms more to locate there.

That is, if they can even make that sale.

“What we see in the data is that the poorest places that think they’re going to benefit the most are never winning subsidy deals,” Slattery says. “They’re not the winners, and they’re not runners-up.”