Inequality makes it harder than ever for cities to say no to Amazon

New York and Virginia have given Amazon somewhere between $2 billion and $3 billion in tax breaks to convince it to do what it was probably already going to: open its two new headquarters in their states.

But as easy as it is to point out how wasteful this is - literally anything would be a better use of this money, but especially if it went toward needed investments in schools and infrastructure - it's hard to get governments to stop handing out these kind of goodies. Why is that? Well, part of it is a simple matter of misaligned incentives: politicians get to bask in the glow of positive headlines about all the jobs they're supposedly bringing to their communities, but aren't around for the negative consequences when, as is often the case, things end up costing more and delivering less than was promised.

There's more to it than just that, though. The same reason that it doesn't make any sense for New York to subsidize Amazon - agglomeration - is also why politicians might feel pressured to do so nonetheless. Which is to say that America's cities are stuck in a prisoner's dilemma where no matter how obvious it seems that they not engage in corporate welfare, they still do, because even losing once raises the specter of losing over and over again. (Amazon's chief executive, Jeff Bezos, owns The Washington Post.)

The important thing to understand here is that, at least within the United States, taxes don't have a lot to do with where companies set up shop. If they did, Silicon Valley would be in Texas instead of California. Or, more to the point, Amazon's second headquarters would be in Atlanta instead of northern Virginia. Indeed, Georgia put together a much more generous incentive package that not only included $2 billion in direct subsidies, but also, as if we needed further confirmation that we live in a post-parody age, the kind of private subway car just for Amazon employees that The Onion had joked about cities desperately offering.

So what does matter, then? Simple: workers, workers, workers. Or, as former Microsoft CEO Steve Ballmer memorably put it: developers, developers, developers. Now, being near a top-tier research university can help create a pipeline of future employees, but what's even more important are the companies that are already in a place. Having a lot of other tech companies around, for example, means that there's a big pool of tech workers for Amazon to hire from down the line. Which it freely admitted was its biggest consideration. "Economic incentives were one factor in our decision," Amazon wrote on its blog, "but attracting top talent was the leading driver."

There are bigger benefits, in other words, to clustering around companies like your own than there are to just getting a tax write-off. And it's not just about hiring, either. It's also about the kind of cross-pollination of ideas that happens when you're around like-minded companies. You're quicker to adopt new and better ways of doing things, and to come up with even newer and even better ways yourselves. It's what economists call the benefits of agglomeration, and it's the opposite of Yogi Berra's famous saying: everyone wants to go there because it's too crowded.

Which is why New York and Virginia shouldn't have had to use the lure of tax breaks to get Amazon to come. It probably would have wanted to anyways. These places, after all, already have pretty big tech footprints, are geographically close to other important industries, and have the infrastructure in place to handle lots of new workers. If anything, they should have been using the money they spent subsidizing Amazon to upgrading their existing - and far too often faltering - public transit to make themselves more attractive to businesses in general.

But it doesn't take much imagination to see how they could feel blackmailed into doing otherwise. Just because you're a cluster today, doesn't mean you will be tomorrow. Letting Amazon go, as we said before, to Atlanta might make it more likely that the next one would, too. And before you know it, then you're the one having to offer billions and billions more than other cities to try to attract tech companies. Better to grease the wheels with a slightly smaller, but still significant, subsidy - or so they tell themselves. The reality, of course, is that New York and the Washington, D.C., metro area really don't have anything to worry about, but corporations still know how to prey on their insecurities. As The Atlantic's Derek Thompson points out, state and local governments spend upward of $90 billion a year trying to poach jobs from one another this way. This is macroeconomically useless behavior that, studies show, doesn't improve overall job creation. All it does is increase short-term stock prices at the cost of not increasing our long-term ability to grow.

This arms race between cities is a lot like the one over college admissions. In both cases, the results probably don't matter as much as people think, but, because we live in such an unequal society, they don't want to take a chance on it. So rich cities, like well-off families, do everything they can to stack the deck in their favor, even when they really don't have to. The rewards for success are so high now that anything else seems irresponsible.

Or at least that's what Corporate America wants you to think.