Bebeto Matthews/AP Photo
Protesters carry anti-Amazon signs during a rally and press conference on November 14, 2018, opposing Amazon getting subsidies to locate its second headquarters in Long Island City, New York. Amazon would later pull out of the deal.
I lived in San Francisco during the first dot-com boom, and I can pinpoint the moment when it was clear that the tech industry had fully given in to irrational exuberance: I left my office to see a Ferris wheel set up in the street for a website launch party, and then a day later saw a different Ferris wheel set up for a different launch party in the same location. It just didn’t feel sustainable, and sure enough, it wasn’t.
For the second dot-com boom, the one associated with the platform giants, Amazon’s sweepstakes for a second headquarters had that feel to it. Over 230 cities groveled at the feet of Jeff Bezos in 2018, offering up billions of dollars in subsidies for the privilege of hosting an executive office tower. The display was rather nauseating, from then-Boston mayor and future labor secretary Marty Walsh’s “love letter” to Amazon, to then-Chicago mayor and future U.S. ambassador to Japan Rahm Emanuel promising to give Amazon income tax receipts from the company’s own staff, to the city of Stonecrest, Georgia, promising to rename itself “Amazon” if it won the bid.
The frenzy also made no sense. One of the most powerful and wealthy companies in the world didn’t need to ask for giveaways to build an office park. Large companies routinely have auxiliary locations in important cities, and nobody runs a tournament to see how much money they can get for locating it. What’s more, Amazon was always going to pick the cities where it made business sense—big cities key to the global economy like New York and Washington—and not some town in the middle of Georgia. (Indeed, Amazon would end up picking New York and a nearby suburb of Washington, though it pulled out of New York after pushback from local lawmakers.) Plus, all the promises of jobs and opportunities for the winning bidder would obviously fall away if Amazon could no longer justify its growth targets.
Funny thing: Last Friday, Amazon gave word that it would pause construction on “HQ2,” as it was known, set for Arlington, Virginia. One phase of the sprawling campus is still set to open, with accommodations for up to 14,000 Amazon executives. But the second phase, which includes public green space, commercial restaurant and retail outlets, and an amphitheater—basically all of the community and broader economic benefits that a city would expect in return for a truckload of subsidies—would be pushed off.
If it seems like a bait and switch, well, it probably is. But there’s another important piece to consider. What this move confirms is that Amazon never cared about the HQ2 extravaganza as a way to find office space for its executives. The main motive was as an intelligence-gathering operation for its logistics business.
Now, there are other reasons why Amazon decided to slow down its Arlington project. Most obviously, more tech employees and executives are working from home, and as I wrote last month, Amazon is planning to lay off 18,000 workers, including 5 percent of its corporate staff, thus reducing the need for new office locations for the time being. The promises of five years ago mean nothing in the face of the new economic reality today, with the stock price plunging and a leaner workforce imagined in the future.
Matthew Barakat/AP Photo
Construction is halted due to rain, April 30, 2020, in the Arlington, Virginia, neighborhood where Amazon is building a new headquarters.
Of course, that’s part of the point: Amazon’s promises to cities were always contingent on its own financial picture. Every city should have known that, and shouldn’t have rested its hopes on Amazon growing forever, like the “National Landing” area of Arlington appears to have done. Long Island City, which Amazon initially chose as one of its two HQ2 winners before a coalition led by New York state Sen. Michael Gianaris and Rep. Alexandria Ocasio-Cortez forced the company to tuck tail and run, should be breathing a big sigh of relief right now.
But Amazon got something much bigger out of the bidding war. It got proprietary information from cities in virtually every corner of the United States: information about public property, access to infrastructure, and the makeup of the workforce—a treasure trove of documents for a company that wants to put in a facility somewhere. It also got a specific bid for how much that city would be willing to give to Amazon in economic development subsidies.
This is incredibly valuable information for a company that is building out an enormous footprint across the country. Amazon acquired enough information to go to a city and say, hypothetically, “We cannot give you a headquarters, but we would like to put in a warehouse/data center/technology incubator/air hub in your community. Maybe we can talk about that.”
Amazon announced the HQ2 lottery in 2018. It nearly doubled its facility capacity from the end of 2019 to the end of 2021. In fact, it overbuilt relative to current demand, and is now closing some of those facilities. But the subsidies kept rolling in.
Good Jobs First, which tracks economic development handouts, currently lists $5.1 billion in Amazon subsidies since 2000. The bulk of that money—about $3.3 billion—came in since 2018. Cross-referencing the places since 2018 that gave Amazon cash and the 238 cities that bid for the Amazon headquarters yields a significant overlap. Boston, Charlotte, Nashville, Denver, Tulsa, Raleigh, Portland (Oregon), Wilmington (Delaware), Worcester (Massachusetts), and numerous others both submitted bids for HQ2 and later granted subsidies for Amazon distribution centers or other facilities.
Even smaller cities that got distribution centers were done strategically to reach those larger metros, nearly all of which submitted bids and gave Amazon insight into where to build. Knowing that a larger city will be building a road in a particular place makes it easier to know what town with access to that road would be attractive for the next node in the logistics network.
Amazon got all of this city planning data on the future construction of America for free, as well as bidding information that offered insight as to who might pony up fat subsidies. That was always the goal of the HQ2 project; where corporate office executives would live and work was completely secondary. The pausing of the Arlington project helps reinforce that.
It was not hard to figure this out at the time (shameless self-promotion here, and here), and now it’s really obvious. Amazon got exactly what it wanted out of the HQ2 nonsense; the cities vying for its headquarters, not so much.
