Why Amazon and Google Chose New York City Instead of a Place That Needs Them More

Photo: Christian Heeb/laif/Redux

Vox’s Matt Yglesias laments Amazon’s apparent decision to split its planned, exciting “second headquarters” into two boring, large branch offices in two obvious, high-cost markets:

Simply put, while locating large pools of high-salary white collar positions in the New York and DC metro areas makes a ton of sense for Amazon, it doesn’t actually make that much sense for either greater New York City or greater Washington. Amazon’s presence will tend to exacerbate those cities’ crises of housing affordability and overburdened transportation infrastructure. And it makes no sense at all for the United States of America, which urgently needs more economic opportunity in dozens of other metro areas that have a different set of problems.

It’s not just Amazon; shortly after I wrote that nobody calls Google’s huge New York office a “second headquarters,” the Wall Street Journal reported that Google intends to expand its New York presence to 20,000 workers, which means it will rival the size of Amazon’s planned “second headquarters” here.

Certainly, it would be nice for Cleveland if Amazon (or Google) had chosen to locate its second headquarters there. But I disagree that, as Yglesias goes on to vaguely suggest, the federal government ought to explicitly try to push superstar tech companies away from high-cost, high-desirability metros, and into cities with lots of available office space and housing to revitalize.

There are clearly things Amazon and Google think they can do best by doing them in New York and Washington and Seattle and the San Francisco Bay Area. If we force or induce them to go elsewhere, what productivity and innovation will be lost in the economy? Probably at least some; otherwise, they wouldn’t choose to locate in places where real estate and labor are so expensive.

There are good reasons for industries to agglomerate — for example, they want to be where their suppliers and customers and competitors are, and where qualified workers are available to hire. In the case of companies like Amazon that are seeking to hire in-demand, highly paid workers, they locate where their desired employees are willing to live. And a lot of Amazon’s target hires wouldn’t be willing to live in Cleveland. (Sorry, Cleveland.)

The geographical distribution problem Yglesias identifies is real. I just think we should probably look away from the tech industry to address it.

I am reminded of a theme Bloomberg’s Matt Levine repeatedly addresses in his Money Stuff newsletter: For an elite slice of the labor market, tech is the new banking. Young people used to come out of Harvard and Yale and go to work at a bank like Goldman Sachs for a few years, because that was just the thing one did. Now, those people go work at Google or Facebook or Amazon. And while many of them are eager to move to Silicon Valley or Seattle, a lot of them would like to live in New York. So, those companies open big offices here.

But what about the banks? As banking increasingly becomes, as Levine puts it, “no longer a standard post-graduate form of finishing school” but “just a job for the sorts of people who might want that sort of job,” does so much of the financial industry need to be in the most expensive metro areas anymore?

Of course, a lot of banking is already outside the very biggest cities. Bank of America has a big tower in midtown Manhattan, but it’s headquartered in Charlotte, North Carolina. San Francisco–based Wells Fargo (my former employer) also has a huge office in Charlotte. Maybe one way to free up more office space and luxury apartments to support all the tech jobs the giants are creating in New York and San Francisco is to move more and more of the bankers currently located in those cities to North Carolina.

Banking is not the only industry that might be geographically rebalanced. There can be more filmmaking in Atlanta, more aerospace in Wichita, more biotech in the Research Triangle, more advertising in Chicago.

And of course, as Yglesias also notes, there is no reason besides political opposition that New York and Washington can’t build more housing to keep up with the jobs they create. But given that they are likely to build less housing than they should, these markets are sure to go to the industries that can bid highest for office space.

Tech is always likely to be a high bidder, and won’t be drawn to Cleveland. It’s worth thinking about what other industries might be, and how.