A city celebrates landing a major factory with 2,000 promised jobs, offering $200 million in tax breaks to seal the deal. Five years later, the factory employs 400 people, most transferred from a nearby state, and the company has quietly stopped filing its required progress reports. The mayor who cut the ribbon is now a lobbyist. Nobody gets fired. Nobody pays back the money.

This story repeats itself in nearly every state, every year. Understanding why requires looking past the ribbon-cutting photos into the messy machinery of how tax incentives actually work—and often don't.

When a big company announces it's looking for a new headquarters or factory location, what follows isn't really a search. It's an auction. Site selection consultants—an entire industry that exists just for this—quietly signal to states and cities that they should compete. Amazon's HQ2 competition was just the loudest version of a game played constantly behind closed doors.

The incentive packages get bigger because officials fear losing to the neighboring state more than they fear overpaying. A governor can point to jobs 'won' at a press conference. Nobody holds a press conference about the tax revenue quietly forfeited over the next twenty years. The math is asymmetric, and companies know it.

Studies consistently find that in roughly 75 percent of subsidy deals, the company had already decided where to locate before negotiations began. The incentive didn't change the decision—it just transferred public money to private shareholders. The auction wasn't about attracting investment. It was about extracting a discount for investment that was coming anyway.

Takeaway

When the same decision would have been made without your money, you didn't buy an outcome—you bought a story to tell voters.

Job promises in incentive deals are less like commitments and more like weather forecasts—directional guesses everyone treats as facts. A company might promise 1,500 jobs, but the fine print often counts contractors, part-time workers, and positions transferred from existing facilities. The 'new' jobs frequently aren't new to the region, just new to that address.

Then there's the ripple math. Officials multiply direct jobs by generous factors to project 'induced' employment—the barista who serves coffee to the new workers, the accountant who does their taxes. These calculations, produced by consultants paid by the companies seeking incentives, tend toward optimism. Independent audits routinely find actual job impacts at a fraction of projected levels.

Meanwhile, the jobs that do appear often pay less than promised, offer fewer benefits, or evaporate when the tax credits expire. A warehouse that qualifies for a 'high-wage jobs' credit might use creative averaging that includes executive salaries to hit thresholds while most workers earn near minimum wage. The number on the press release and the reality in the parking lot rarely match.

Takeaway

In implementation, definitions do more work than dollars. Whoever defines what counts as a 'job' controls whether the program succeeds on paper.

Most incentive deals include 'clawback' provisions—legal language saying companies must return the money if they don't deliver promised jobs or investment. On paper, this sounds like accountability. In practice, clawbacks are among the most consistently unenforced tools in public policy.

The reasons are structurally boring, which is exactly why they persist. Enforcement requires staff who monitor compliance, lawyers willing to sue major employers, and political leaders willing to demand repayment from companies that might threaten to leave entirely if pressured. Economic development agencies are usually staffed to recruit companies, not to prosecute them. The recruiter and the enforcer are often the same person, wearing incompatible hats.

Even when violations are documented, settlements typically recover pennies on the dollar. Companies negotiate 'modified' agreements that reset the clock or forgive shortfalls. Sometimes the original officials who signed the deal are gone, and their successors have no political stake in enforcement. The clawback exists in the contract, but the muscle to invoke it exists nowhere in the org chart.

Takeaway

A penalty that nobody is assigned, funded, or incentivized to enforce isn't a penalty—it's decoration.

Tax incentives aren't inherently corrupt or foolish. But they operate in a system where the people offering them get credit immediately and the people paying for them—future taxpayers—have no seat at the table. The result is predictable.

The fix isn't ending incentives entirely. It's asking harder questions upfront, defining outcomes precisely, and funding enforcement as seriously as recruitment. Accountability without staffing is just optimism in legal clothing.