Tesla Motors, which recently wrangled more than $1 billion in tax breaks from Nevada, proved itself unusually adept at obtaining government support for its private enterprise. But it’s hardly alone in tapping state and local governments for incentives, such as tax credits for jobs created and money invested. All 50 states offer tax breaks designed to attract new businesses or prevent existing ones from moving, costing more than $80 billion annually in lost revenue.

Unfortunately, few states actually go back and check to see whether the incentives they awarded boost the economy, rather than just the companies’ fortunes. While more states are working to improve accountability by making sure businesses do what they pledged to do when they were awarded the incentives, there’s still relatively little work done to measure how much better off an incentive program has made the state as a whole, if at all.

Governments offer incentives on the theory that giving a tax break or a handout will lead to more jobs and investment over the long run. Numerous studies, however, have raised doubts about that theory, suggesting that many incentive programs simply shift jobs around rather than creating new ones, or reward companies for steps they would have taken anyway.

Of course, a state may be perfectly happy with a program that brings in jobs and investment at the expense of other states. Although winning the Tesla “gigafactory” cost Nevada up to $200,000 per job at the plant, it’s still 6,500 jobs that might otherwise have gone somewhere else. Nevertheless, it’s important for states to find out whether the breaks and subsidies they provide increase the state’s economic output and wealth, rather than just assuming they do.

According to the Pew Center on the States, only about 10 states do that kind of follow-up. The ones that do have found some eye-opening results. For example, when Louisiana checked on the 9,379 jobs that were supposed to be created with the help of its enterprise zone incentives in 2009, it found a likely net gain of only 3,000 jobs. That’s because many of the projects would have occurred without the incentives, and because some of the new jobs came at the expense of nearby competing businesses that scaled back or shut down.

It’s impossible to say how much is too much to pay in incentives, considering that tax breaks are just one of many factors businesses consider in deciding where to invest. Some states, including California, take the partial step of demanding measurable results from their incentive programs, such as the number of employees hired and the amount of capital invested. But they need to go further, examining how those results ripple through the rest of the state or local economy. Only then can they tell whether an incentive delivers on its promise.