Superstar cities versus everyone else is a fake problem
“Going to DC on a field trip in middle school was my personal Vietnam” — Midwest Boomers
No event symbolized the prevailing narrative of superstar city dominance better than Amazon’s HQ2 search. Over the course of 14 months, the e-commerce giant made a huge show out of its decision on where to place their second major headquarters, complimenting the original in Seattle. It was exactly the type of event that Bloomberg and Forbes writers dream of: a nationwide, winner-take-all competition with thousands of jobs on the line, featuring one of the most prestigious companies in the world.
Amazon did its best to hype up the process, releasing a list of 20 finalists in January 2018:
Amazon’s finalist list was fairly diverse. It contained the 4 major Acela corridor metros, 6 Sun Belt cities, and 4 Midwest cities, plus Denver, LA and even Toronto.
After 10 months of driving up the tax incentive bidding war between finalists, Amazon made its decision. In a twist, they chose not one, but two locations for HQ2. But there was nothing surprising about the two metro areas they chose: New York City and Washington DC.
NYC and DC, along with Boston, are the east coast’s three superstar metros. The west coast has three of its own: Seattle, San Francisco and Los Angeles. Together, these two groups form what Pete Saunders calls the “Superstar Six”. These six regions have several things in common: high economic output, high housing prices, high foreign migration rates — and high rates of people who write for national media outlets.
National journalists love to write about the cities they’re overwhelmingly based in, and how they’re so important and filled with “talent”. Shortly after the HQ2 decision was leaked, the New York Times published an article titled In Superstar Cities, the Rich Get Richer, and They Get Amazon. The core thesis of the article is that it was inevitable that HQ2 would go to some combination of {NYC, DC, BOS, SF, LA}, and there was never a realistic chance that it would land in Indianapolis, Columbus or Pittsburgh.
That thesis is correct. The most prestigious companies in tech have largely steered clear of making huge investments outside of the “Superstar Six” regions. And when they have done so, it’s mostly been in rapidly-growing Sun Belt metros like Austin, not in the Midwest. The article also correctly diagnoses the reason for this: the Superstar Six have a ton of highly-specialized workers with the skillsets they want. Amazon simply has far more hiring options in NYC or the Bay Area than any Midwest metro.
What I don’t like about this article, though, is the framing: the implication that those who live in the superstar metros are doing awesome and will continue to live more and more prosperously, while those who live in the less flashy regions that got passed over are being left behind by an increasingly globalized economy. Some representative quotes:
“New York and Washington are leaving the rest of the country behind. Companies like Amazon explain why.”
“A small number of rich and internationally connected cities keep increasing their economic advantages — and as a result, the inequality widens between them and everywhere else.”
“In one sense, it’s good for the country that companies like Amazon are as productive as they are, where they are, thanks to these forces. But higher national productivity is no comfort to communities that aren’t gaining jobs. Reconciling the two sets of places is one of the country’s deepest economic challenges.”
“We look naïve in even raising the question that this could have gone to a different kind of Midwestern, heartland place,” Mr. Muro said. “There wasn’t really an alternative.”
Midwest Decline is Overrated
Based off the above quotes, you might think that most of the Midwest is economically depressed, with substantially below-average incomes and GDPs, as all of their college-educated talent flees for the coasts. But you would be wrong. Indianapolis (68k) has a higher per-capita GDP than Portland (66k). Cleveland (65k) is higher than Sacramento (62k). Minneapolis (73k) is higher than Houston (69k). The Midwest is still wealthier than the South, and about as wealthy as the nation as a whole, despite the explosion of Silicon Valley, the decline of manufacturing, and widespread migration to the Sun Belt.
And the Midwest has a *positive* net-migration rate with the Northeast, California and Washington State. That’s right, more people are moving from California to Indiana, and from New York to Ohio, than vice-versa.
The South used to be extremely poor. Now, it’s less poor.
While Midwest decline is overrated, it does exist to some extent. Southern decline, on the other hand, has been largely non-existant over the past few decades. On the contrary, most southern states have started to close the economic gap with the Northeast. From 1997 to 2016, the GDP’s of Texas, Florida and North Carolina grew by 163%, 130% and 123% respectively. Meanwhile, the GDP’s of Massachussetts, New York and New Jersey grew by 114%, 109% and 89% respectively.
In addition to gaining economic share, the Sun Belt has rapidly siphoned population from the Northeast. From mid-2018 to mid-2019, the South gained a net of 222k people from the Northeast, with over half of that coming just from New York and New Jersey.
High-skilled workers in the Superstar Six make a ton of money — and then give a ton of it to their landlords
There’s no doubt that raw economic output is highest in the Superstar Six. Per-capita GDP in the Bay Area is 130k, nearly double that of anywhere in the Midwest or Southeast. But who is capturing all of that extra economic output? I’d argue it’s not the young tech workers, who pay much higher rents (and income taxes) than their counterparts elsewhere. Instead, it’s largely those who have owned land in the Bay Area for years, and watched their property values skyrocket year after year, while their property taxes barely rise due to Prop 13.
To be clear, you shouldn’t feel sorry for techies in NYC or San Francisco. They live comfortable lives in some of the most interesting cities in the world. But there isn’t much reason to feel sorry for the white-collar professionals who live in Carmel, either, even though their region doesn’t get discussed much on Twitter, and never had a legitimate chance to land Amazon.
Every major metropolitan area has a favored quarter. In Indianapolis, it’s Hamilton County. In Kansas City, it’s Johnson County. These are places where a majority of adults have a Bachelor’s degree (albeit from Big 10 / Big 12 schools, not the Ivy League), and a good paying job (albeit with the likes of Eli Lilly and Cerner, not Google and Amazon), and own a spacious, affordable home.
They’re doing just as well, perhaps even better after taxes and housing costs, as their counterparts in San Mateo and Nassau Counties, or in Brooklyn Heights and Northeast San Francisco. And most of them don’t really care that their regions don’t have good public transit or aren’t internationally renowned. My uncle, who’s lived on a lake in exurban Hamilton County for 30 years, told me “you couldn’t pay me enough to live in New York”.
Instead of hand-wringing over the fact that the most prestigious companies like to congregate in ~6 metropolitan areas, leaving the white collar professionals of other major metros to make solid middle-class salaries working for successful-but-not-as-flashy companies like Cerner and Eli Lilly, journalists should focus on a much larger divide: the canyon between these favored quarters, and the poor inner city neighborhoods in the same metropolitan areas.
The primary economic divide is still inner cities versus their suburbs
17.9% of New York City residents live in poverty. That’s only a tick lower than Indianapolis (18.0%), and it’s higher than Kansas City (16.1%). Of course, all of those figures are much too high, as is the rate in San Francisco (10.3%), the richest major city in America. Meanwhile, inner-ring suburbs in major metropolitan areas almost invariably have high incomes and low poverty rates. And while outer-ring suburbs and exurban areas tend to be poorer than inner-ring suburbs, they rarely reach the same poverty rates as their corresponding inner cities. These are the fault lines that economic and racial segregation fall along in today’s America: Carmel versus Center Township, not NYC versus Indianapolis.
In small metros, too
The pattern of “poor cities, rich suburbs” is not only present in metros with millions of people.
South Bend, Indianapolis and New York City are all very different cities. But they all have double-digit poverty rates and median incomes below the national average, while being bordered by suburbs with six-figure median incomes and well-regarded school districts.
For more on the history South Bend and its suburbs, read Joseph Molnar’s series More People.
Poor urban neighborhoods, not rural areas, have been most left behind
The second half of the “superstar cities vs. everyone else” narrative is that, in additon to hurting smaller metropolitan areas, the dominance of superstar cities has resulted in rural America getting left behind, which helped fuel Trump’s win in 2016. This is largely false.
There’s a huge variance in how well-off America’s rural counties are. Many North Dakota counties are near the top in terms of per-capita GDP, while parts of Appalachia are among the poorest counties in America. But as a general rule, the vast majority of rural America is better off than the poorest urban zip codes. Center Township in Indianapolis (population: 151k) has a poverty rate of 26%, higher than any Indiana county, and a median income of 37k, lower than any Indiana county. And the Indiana counties with the highest poverty rates are not pure-rural counties, but counties with small cities, such as Muncie’s Delaware county.
Furthermore, there is essentially no correlation between the income of a county and how it swung politically from 2012 to 2016, once you control for education. Almost everywhere that has a large population of white people without Bachelor’s degrees swung heavily to Trump, whether they were doing fairly well economically (eastern Iowa) or poorly (northern Indiana). Rural America getting ecnomically “left behind” was not the primary cause of Trump’s win. The areas that have been most left behind are the most Democratic zip codes of all: majority-minority precincts in inner cities (Trump did, however, make gains in some of those in 2020).
The Real Problem with Superstar Cities
If there is one problem with the Superstar Six, it’s NIMBYism. NIMBYism prevents them from growing further and becoming even more productive. Because the people who are actually driving the excess economic output in these regions are largely losing it to their landlords, domestic migration is overwhelmingly flowing away from them, and towards the less-productive Southeast. This also means less people are living in dense cities with high rates of public transit use, and more are living in Sun Belt sprawlvilles like Atlanta and Austin. That’s a disaster for the climate.
But the “everyone else has been left behind” narrative is mostly false. Economic inequality in America is overwhelmingly about differences within metro areas, not differences between metro areas.
