A new headquarters from Amazon (whose chief executive, Jeffrey P. Bezos, owns The Washington Post) would pack enough of an employment and economic punch that it could have a measurable effect on presidential elections. To estimate that effect, we need to understand who Amazon’s workers will be, and how they’ll vote.
The company's solicitation for bids indicates that the retailer is looking for a workforce that is “highly educated” in software development and other disciplines and steadily renewed by a “strong university system” and outside workers drawn to the area. These and other requirements strongly indicate that the company is looking to attract recent college graduates and other young, educated workers.
Amazon’s HQ2 promotional materials focus on hiring local workers (as you might expect from a colossal corporation trying to win incentives from host cities), but their requirements also emphasize a city’s “ability to recruit talent to the area.”
Which one predominates in practice? We can’t venture a guess, but the answer would significantly affect the political impact of Amazon’s decision. Outside workers are presumably more likely to change a city’s political mix than residents, but the arrival of a major new employer would help a city retain the sorts of workers that, in a previous era, might have moved away.
The group Amazon seems to regard as its hiring pool, college graduates younger than 40, leans heavily Democratic. In the 2016 election, the national network exit poll found 56 percent of them voted for Democrat Hillary Clinton vs. 35 percent for Trump.
But will there be enough of them working at Amazon's new headquarters to swing any states? Let’s try running some numbers.
It’s easy enough to count the workers. Amazon says it will hire as many as 50,000 people at its new hub over the next 10 to 15 years, as well as create “tens of thousands” of additional jobs through direct investment. In Seattle, it pegs that additional-job number at 53,000. We’ll use that figure in our back-of-the-envelope math,to obtain a high-side estimate.
That starts us out at 103,000 theoretical Amazon and Amazon-adjacent workers.
To estimate how many voting family members those workers might bring along, we can use a 2015 Pew Research Center analysis of Census Bureau data that finds a marriage rate of about 65 percent for adults age 25 or older.
That brings us up to 169,950 workers and spouses. Caveat: We have no idea how many Amazonians married each other. The government doesn’t release that info.
Not all of those are eligible voters. Some may be felons, while many more won’t hold citizenship in the United States. For our estimate, then, we’ll count only 92 percent of the total — roughly the percentage of adult U.S. residents who, according to the Census Bureau, held citizenship in 2015. That drops it to 155,844.
And not all those who are eligible to vote go to the polls. According to the census, voter turnout was 64.1 percent for bachelor’s degree holders ages 25 to 44 in 2016. That drops our number to about 99,896.
Based on our assumptions, the Democrats would capture about 55,942 of those voters and Republicans would get 34,964, making for a total Democratic margin of 20,974.
That’s more than Trump’s 10,704 margin in Michigan and close to his 22,748 margin in Wisconsin, but it’s not big enough to flip any of the states on the Amazon shortlist.
Ours is already a high-end estimate of Amazon’s impact — it accounts for both Amazon’s hiring and the jobs it claims to directly create, and assumes that all such jobs will be new to the region. A more conservative set of assumptions would result in a margin of less than half that size. But neither model captures everything.
Specifically, the cities clamoring for the online retail giant’s attention are counting on a salubrious knock-on effect. They hope that Amazon’s economic activity and stamp of approval will draw in other tech companies and help their city become the next Seattle or San Jose.
If that happens, all of the political effects here will be multiplied, perhaps to the point where the state hosting the chosen city will tip from one party to the other — unless Toronto wins, in which case all our assumptions are bunk because it's in Canada.
These cities are begging to have their economies transformed by Amazon, but not all of them, particularly those in red and purple states, may realize that it could transform them politically as well. After all, U.S. tech hubs end up following a similar political pattern.
Even if the city’s presidential pick doesn’t change, their new Amazonian voting bloc is likely to upend local and congressional elections.
Scott Clement contributed to this report.