What the Amazon tax breaks really mean

I think the easiest way to understand Amazon’s HQ2 and the specific question of cities handing out tax subsidies is to abstract away from the particular details of Amazon’s package.

A normal way for governments to raise revenue in the United States is with property taxes. If you build a gigantic new office tower somewhere, then you are going to owe a lot of property taxes to the local government.

So if you’re smart and have good lobbyists, you’ll say, “Hey, we were thinking of building a huge office tower in your city, but your taxes are so high. Maybe instead of paying the 8 percent property tax rate that’s on the books, we could pay 2 percent instead for the first 20 years.”

The city may well agree to this proposal because 2 percent of a giant office tower is still a lot more than 8 percent of a parking lot. What’s more, bringing your office tower to town is going to generate a lot of sales tax and income tax revenue while probably helping to bolster property values across the board. So while reporters are going to write, accurately, that your company got millions of dollars in tax breaks in order to build your office tower, the city still ends up with more tax revenue than it would have had if you hadn’t come to town.

So you can’t really say things like, “Instead of handing out millions in subsidies to Jeff Bezos, we should invest in the subway.” The subway doesn’t need a tax break; the subway needs actual money. And the tax breaks don’t represent money in hand that could be spent.

Which doesn’t mean that the tax breaks are good!

One basic issue here is that competition between cities has regressive impacts. Opening a big office tower generates a lot of economic surplus. It’s good for the company, but it’s also good for the host city. In these Amazon-style competitions, the company pits cities against each other to see which one will cede the largest possible share of the surplus back to the company. Those subsidies aren’t an actual cost to the city that wins the competition, but they do represent a lost opportunity to share the economic surplus more broadly.

I don’t have an elegant solution to the problem of municipal competition, except to say — as I have said before — that the federal government ought to start playing a larger role in these situations to try to prevent cities from racing to the bottom.

What we want is a healthy form of competition where companies try to locate in cities that provide a high quality of public services relative to their tax revenue, so that cities have an incentive to try to govern themselves well. The current dynamic not only allocates too much surplus to rich companies, it undermines that healthy form of competition. Amazon doesn’t need to care whether New York City has cost-effective government or not; it just opts out of paying the costs.

But the very fact that mayors are inclined to hand out these subsidies should teach them a lesson about tax policy.

The specific logic of Amazon subsidies is, when you think about it, exactly the general logic of Henry George’s old idea that we should tax land value rather than property value.

Under Georgeism, in essence, everyone gets a “tax break” for investing in new physical structures because there is no taxation of investment in new physical structures. It’s only passive ownership of the land that gets taxed.

This obviously wouldn’t solve all the problems associated with cities’ competitive bidding for business favors. But the fact that cities are inclined to make these kinds of bids and aren’t necessarily wrong to do so ought to prompt some deeper thinking on their part about the extent to which their tax codes are or are not aligned with their actual policy goals.