Three years ago, I wrote in The Federalist about Amazon's decision not to establish a New York City headquarters in the face of strong opposition from Alexandria Ocasio-Cortez and many liberal groups. Opposition mainly stemmed from the $3 billion tax incentive package that was prepared for the company.

My conclusion was simple: AOC was right to oppose the tax giveaway.

The question of whether, or to what extent, incentives are necessary isn't just an issue in the case of Amazon, either, and research into the incentives that include or imply "but for" language—"but for the incentive, the project won't happen"—[is] helpful here. For example, a study by the W.E. Upjohn Institute published last year reveals that the vast majority of businesses that receive tax incentives under a "but-for" rubric likely would have pursued their projects even without the incentive. . . .

This failure of stewardship from governments across the country costs state and local taxpayers billions of dollars annually. That affects not only government services, including roads and education, but also a government's ability to reduce taxes for everyone, if it so desired. The city of Kansas City, Missouri, where I'm from, redirects $90 million annually from its budget through tax incentives, but that doesn't include the additional $42.5 million those decisions redirect from the city's public schools and other taxing districts, who rely on these tax streams but have relatively little say in their diversions.

When a government incentive package is worth billions of dollars, it has truly reached the outer frontier of questionable economic interventions. At $3 billion, New York's Amazon package was prodigious, but what if I told you last year Kansas City put together an incentive package worth up to $8.2 billion for a nondescript data center development? What if I told you the state of Missouri was also kicking in potentially millions of dollars?

And what if I told you that last week, Facebook—I mean, "Meta"—accepted their offers?

The data center, set to open in 2024, will support up to 100 permanent jobs and is the first of its kind in Missouri. During construction, the job will bring in an additional 1,300 jobs. . . .

The $800 [million] facility will come to Golden Plains Technology Park, a 5.5-million-square-foot land development in Kansas City's Northland.

Tax incentives are always a dicey proposition because they substitute government decision making for the market. But tax incentives for data centers are especially dicey because the number of permanent jobs they generate is almost always paltry, with the "cost per job" of these packages often landing over $1 million. Although the final accepted package remains to be published, the KC Meta data center project seems likely to far exceed that already absurd baseline of cost per job.

Economic development policy that is determined by the number of ribbons that get to be cut and hard hats that get to be donned by politicians is inherently fraught with perverse incentives that force every other taxpayer to carry the weight of government on behalf of an increasingly select group of developers and corporations. Local government should be reducing taxes on everyone rather than just for some of the wealthiest companies in the world, especially in an era of historic inflation.

That Facebook gets priority over every other individual and company in Missouri is infuriating—but unfortunately, it is not surprising.