New York City and the state sent loads of data to Amazon during its search for a new headquarters, offering a peek into the valuable information the company collected during the HQ2 process.
NEW YORK — An avocado at Whole Foods costs $1.25. Columbia University handed out 724 graduate degrees in computer science over the past three years. And 10 potential land parcels in Long Island City are zoned M1-4, for light manufacturing.
New York provided all of these data points, and thousands more, to Amazon as part of its successful bid to woo the tech giant to town.
On Monday, New York City posted online the 253-page proposal it submitted, along with New York state, to Amazon in March. The city quickly took the file down, saying it should have checked with its partners before posting it, because the document included proprietary information. But The New York Times downloaded the document before it was taken off the public website.
The proposal shows the types of data, some rarely available publicly, that the company amassed from cities across the country as part of its search for a second headquarters.
Most Read Nation & World Stories
- Here's what we know about the 'delta-plus' variant, already detected in U.S. and elsewhere
- Should I mask? Can I travel? What about hugs? How delta is changing advice for the vaccinated
- Passenger arrives taped to his seat after assault of 3 flight attendants
- Bill Gates says Epstein relationship was ‘a huge mistake’
- Virtually all emperor penguins may be pushed to brink of extinction, study finds
In the end, Amazon chose Long Island City in Queens and Arlington, Virginia, for two new headquarters. But more than 200 cities put in initial bids for the project, and 20 of them provided Amazon with more information when they made Amazon’s shortlist. Each submitted valuable research that the company can mine in the future for picking satellite offices, warehouse locations or research hubs.
When Joseph Parilla, a fellow at the Metropolitan Policy Program at the Brookings Institution, first scrolled through New York’s submission, which The Times asked him to review, one word kept flying out his mouth: “Wow.”
Parilla, who researches economic development, was amazed at the level of detail and hard-to-find data the city had provided. “We are doing a project right now that, if we had access to Amazon’s database, it would make our life so much easier,” Parilla said.
An Amazon spokesman, Adam Sedo, said most of the information the company received was publicly available, and economic development offices regularly provided such data to companies looking to locate in an area.
Here is a look at what Amazon obtained, and how the company did it.
What did Amazon ask for?
In September 2017, Amazon put out an eight-page request for proposals from cities and states. The first round of responses from 238 locations included glossy marketing pitches, with slick graphics and broad proposals for why Amazon should come to their regions. New York’s first proposal included a rendering of a building with Amazon’s logo shining in the skyline.
After Amazon announced its shortlist in January, it gave cities a 29-page request for information that required far more precision and was more about practicalities than flash. It asked cities to respond by early March with a huge text document punctuated only with a few maps.
What information did New York initially provide?
In its first proposal, New York offered to use eminent domain to help Amazon get necessary land, and highlighted four city neighborhoods for a new Amazon headquarters, including in the World Trade Center complex and the redeveloped Farley Post Office in Midtown West. Amazon ended up selecting Long Island City and there are no plans to use eminent domain.
The main criteria Amazon laid out in the initial request, like airport connectivity and education levels, were relatively easy to come by, Parilla said. He could pull it together in a matter of a few hours on his own.
The unique information in the first round of submissions was the details on available real estate sites and incentive programs, said Richard Florida, a professor at the University of Toronto.
What new information was in the second proposal?
New York’s second response included dozens of pages of detailed information on outcomes from the city’s educational institutions that, the document says, is not all publicly available. “Some of this info only exists within organizations,” Parilla said. “It would be hard to coordinate all of that gathering.”
Amazon asked for detailed information on the availability of machine-learning specialists, user-experience designers and hardware engineers — three jobs critical to its growth. The proposal says New York University awarded 64 undergraduate and 63 graduate degrees in integrated digital media, which includes design, in the past three years. Amazon also learned that Columbia University has outreach programs for STEM programs for K-12 schools in the Bronx and Upper Manhattan that reached 3,682 students last year, and that Columbia planned to expand the offering.
The data, while not flashy, is core to the ability to understand where the best pool of potential employees live and could be developed, which drove the search for the new headquarters.
“During the process it became clear to us that the overriding criteria was going to be the ability to find and attract talent,” Jay Carney, a senior vice president at Amazon, said in an interview immediately after the company announced its plans for New York and Virginia.
Was any of this information public?
Most of it — but the company got the city and state to pull it all together.
New York’s response includes at least 140 footnotes documenting data sources that show how much information was publicly available. Some of it was very basic. New York had to detail, for example, that the state government has a bicameral legislature — the Senate and Assembly — whose officials are elected to two-year terms, and that 70,000 4-year-olds attend the city’s free, full-day pre-K programs. And it laid out the zoning designations for dozens of individual parcels.
While that information is public, it is spread across dozens of sources, and can be time-consuming to compile. The New York response has tables from federal census reports, published studies, city databases and news articles.
Amazon also asked cities to provide the costs of a “gallon of 2% milk, loaf of whole wheat bread, and an avocado” at a local Whole Foods, the organic grocer that Amazon owns.
As Parilla said, “It’s like why do the work when somebody else can do it?”
How is that information valuable?
Having that information, compiled in one place, can let Amazon move quickly in the future, Florida said. The company can go back to cities and say, “We didn’t put HQ2 in your neighborhood but we have this really good idea for an R&D hub,” Florida said. “You have this great site here, and you told us what incentives you would have,” and push them to act.
Sedo, the Amazon spokesman, said the company had already used information gleaned from the first round as it looks to future growth. For example, Amazon learned about a program in New Orleans that lets companies inform classes taught at community colleges, which made Amazon think it could be a good place to build a fulfillment center with robotics that require specialized skills to maintain and operate. He said real estate information, while valuable, can change quickly as cities develop.
“You now have this database that you can go back to,” Parilla said. “You have so much more information.”