A new online study by BuildZoom chief economist Issi Romem takes a look at the characteristics of people moving in and out of metro areas and draws some striking if unsurprising conclusions.
Per Romem, coastal elite cities are trending towards becoming more elite because of their migration patterns. In these cities in migrants are more educated, have higher incomes, and more workers per household than out-migrants. The opposite is true in low cost interior cities where in-migrants tend to have less education, lower incomes, and fewer workers per household than out-migrants. Here’s his chart on income and migration.
Another interesting finding is that the coastal city in-migrants are less likely to be homeowners, suggesting in his view that they are feasting on transients.
Here are some excerpts:
In addition to the San Francisco metro area, other expensive coastal metros such as Los Angeles and New York also exhibit positive income sorting. In- and out-migrants in lower-priced metros like Atlanta and Dallas appear to have similar incomes. Rust Belt metros like Cleveland and Detroit exhibit negative income sorting, whereby the median household moving in earns less than the one moving out.
Why do the expensive coastal metros exhibit positive income sorting? These metros are expensive because they have restricted their supply of new housing even as they continue to generate strong demand for it. As a result, homebuyers and renters in these metros compete with each other over a smaller housing stock, bidding up prices, and pushing out those who are the least financially able. The high cost of housing also deters potential in-migrants, especially if their incomes are low, which drives up the income levels of those who are ultimately observed moving in. Of course, homeowners and certain fortunate renters can stay put in their homes for a long time, even when they can no longer afford them at current prices, so positive income sorting can be a slow process.
…
The pattern observed with respect to migrants’ age is consistent with the notion of a transient class, whereby individuals arrive in young adulthood to experience the life and career opportunities that expensive metros have to offer, and leave once they confront the realities of raising children in this environment. The notion of a transient class is also consistent with the pattern observed earlier with respect to the number of earners per household, especially inasmuch as it is driven by expensive metro in-migrants boarding with roommates, and families with children seeking cheaper housing elsewhere.
…
Positive income sorting in the nation’s most expensive metros helps sustain housing price appreciation. Of course, if current residents’ income rises that helps sustain price appreciation, but sustained appreciation does not require a proportional increase in current residents’ income. Instead, positive income sorting helps the population afford housing at higher prices by substituting out-migrants with more financially-able in-migrants. If those leaving would have stayed had housing prices not risen as much, then the process of positive income sorting can rightfully be called displacement.
…
Sustained housing price appreciation could also exert increasingly selective pressure on the kinds of jobs that are feasible to maintain in the expensive metros. Employers who wish to stay in business cannot afford to pay workers a wage that exceeds their contribution to revenue. As a result, employers in the expensive coastal metros would be compelled to maintain only the most productive roles–those which justify wages high enough to compensate for local housing costs–and other types of work would either be foregone, automated, or relocated.
Click through to read the whole thing.
A spreadsheet with details for every metro area can also be downloaded.
One thing I’ve been curious about, and maybe this chart shows it a bit and I’m just not realizing it, is how many people born in these elite metros manage to gain the skills to stay there long term. Can some average kid growing up in Queens at a middling public school expect to afford to stay in New York City long-term, or will he be pushed out into the suburbs or a different metro entirely? Does he even have a chance of joining the elite Manhattan economy, or is all of that talent being imported? I think there’s some interesting dynamics surrounding which areas of the country are creating talent and which ones are importing it instead of using their own citizens, but I can’t recall if I’ve ever seen a comprehensive study on it. I’m sure one is out there somewhere.
I think it was an Atlantic article that wrote up a study on where the best places are to enter, or remain in, the top quintile of earners.
And those places were the elite coastal metros, and towns surrounding the most elite universities. NY and SF are far and away the best places to be born poor, working class or lower middle class. It wasn’t even close.
That kid in Queens has far, far, better chances than anyone between the Hudson and the SF bay.
I’m surprised to see Columbus below the regression line. I think this shows that Columbus produces high income people locally instead of importing then like sunbelt metros like Charlotte, Atlanta, and Tampa. That’s an important difference. Columbus isn’t competing with sunbelt towns, it’s a ‘farm team’ for other, presumably larger and wealthier, metros.
Thank THE OSU.
As many grads leave OSU each year as enter as first year students and Columbus has had net gains of about 30,000 people per year for a while now. I don’t think OSU is the primary factor.
It is hard to imagine that as many back office software engineers would be located in Columbus if it weren’t for Ohio State. That is to say, I suspect that job growth in the region has been greatly spurred by companies locating offices in Columbus in order to take advantage of abundant Ohio State grads.
Exactly. They aren’t students, they’re employees.
…employees who first came from other parts of Ohio and Northern Appalachia to school at TOSU, and stayed.
Why is it important where they came from?
Emptying out the small cities of Ohio (and Indiana) just moves “urban abandonment” to different places.
For every Short North or German Village, Fountain Square or Broad Ripple, there’s a dying downtown outstate to deal with.
That’s too conspiratorial. Columbus offers an example of how other towns in Ohio can get a piece of the new economy. Preventing Columbus’ growth won’t help Lancaster, Lima, or Youngstown.
No one is suggesting that Columbus’ growth be prevented. Only that the long-term effect of its growth is shrinkage in those smaller cities as their younger folks gravitate to TOSU and jobs in Columbus afterward.
—
Also…with regard to Lancaster, read “Glass House” by Brian Alexander. It illustrates the decades-long “conspiracy” to bleed small industrial cities of their accumulated capital (and jobs) that has nothing to do with Columbus, but everything to do with Wall Street. The conspirators are called “hedge funds”. Their conspiracy is based on a measure they call “EBITDA”. It treats capital-intense domestic manufacturing facilities as cash-generators only, and workers as costs to be minimized, until the machinery can no longer be operated because it’s broken and the workers who knew how to keep it running are all driven off.
Still too conspiratorial. Those same hedge funds have driven growth in other places. Lancaster has no right to exist. The country and world owe it nothing. If Columbus, Indiana and Lexington, Kentucky can grow, Lancaster, Ohio can grow.
It’s not “conspiracy”, it’s fact. Donald Trump pretends to understand the economic disease and loss for the benefit of “his” voters, but it’s his NY pals who do the deals to rape steady old American manufacturing companies. People in Lancaster Ohio get no say in who manipulates their community’s biggest asset base this year.
My argument isn’t necessarily about growth, but about keeping what they have and not losing it to financial engineering. Again, read “Glass House” to understand the impact financial engineering using EBITDA has on a viable factory. Private equity floats massive debt against an income stream that diminishes because the scheme doesn’t permit reinvestment or decent wages and benefits. It essentially converts depreciation into wealth for the financial engineers, who then don’t care what happens to the hard assets.
Smaller cities (at least those not named Rochester, Minnesota) don’t have the chance to do “eds and meds” centers around giant public universities, nor can they become state capitals.
Lexington is a university town, much like Bloomington or Lafayette Indiana, or Athens Ohio. UK is by far the largest employer.
Columbus, Indiana is a singular accident of history that I happen to have good personal knowledge of. Two Fortune 500 companies grew up there, both related to the vehicle industry. (One is gone, merged away.) On that manufacturing base, they doubled down on manufacturing in the 80s when Cummins and Arvin went through some hard times. Columbus attracted a number of big “transplant” factories including Toyota Material Handling, Enkei, and NTN Driveshaft. All are broadly “vehicle” related (Toyota makes its lift trucks there) and so the city is still not recession-proof, but it is more diversified.
Why did they “rape” Lancaster and not Columbus, Indiana? Did the people in Columbus, Indiana have a say in their economy that the people of Lancaster didn’t get? Does the world owe Lancaster a living? Is Lancaster more deserving than anywhere else?
But note that Indianapolis is just above the line…because IU and Purdue are not in Indy, but major IU grad schools are.
It’s not a surprise at all. The fortunes of Columbus, Indianapolis, Charlotte and other interior ‘boom towns’ are vastly, vastly over-stated. Ohio is mired in decline, and has a long way to go before it hits bottom.
What’s wrong with the boomtowns? They certainly aren’t perfect places, but it doesn’t seem like a bad thing to have metros with both affordable housing and growing economies.
Will has resentment issues. He doesn’t have an argument.
For one thing, the boomtowns are blue or purple dots in red states, and they operate with state-imposed limitations as a result.
Are there state imposed limitations for the boom towns that are significant problems for those metros? I would say that arguable the lack of Medicaid expansion is a missed opportunity, but outside of that I am not sure I can think of any that are real dollars and cents issues. My impression is that both red and blue states have been reducing their support of public universities.
What are these limits? I have no idea what you might mean.
Limitations on local anti-discrimination laws (heard of RFRA laws?). Limitations on local taxation. Limitations on local transit (Indy is prohibited from building light rails by state law). Limitations on local zoning and development law. Limitations on local minimum wage setting. Even limitations on local plastic-bag bans
So, you’re arguing that states specifically seek to thwart the development of their most successful cities through state laws? Why? Don’t those same laws apply to other cities in those states? How have they limited Columbus?
@Matt- It depends on the state, but Dillon law states allow for a lot of meddling in local affairs by state legislature. In the state where I live, the state legislature attempted to change the zoning laws of specific cities in the state, which would have been perfectly constitutional and legal had it passed. There are certainly states that limit local anti-discrimination or minimum wage ordinances, although I am not sure how much that impacts economic development (probably somewhat, but I don’t have good information on the magnitude). The limitation on Indy building light rail is a good example, although I happen to think that it is actually prudent and good legislation, though I could be convinced that the social benefit of money spent on passenger rail is higher than spending that same amount of money in other ways (or just in having lower taxes).
I still don’t understand how this affects cities as “income sorting machines.”
Cities that are less free to chart their own course are less free to do the things necessary to attract (1) progressive companies with jobs, and (2) young/progressive workers who want jobs in those companies.
How has Columbus attracted the progressive jobs and workers it has if these limitations have been at work? With growing economic power comes growing political power to structure state laws.. how about ending non-compete contracts in Ohio?
Columbus has three very large legacy employers (besides the state university and state government): Nationwide Insurance, Chase (former Bank One), and The Limited/L Brands, plus a couple of big fast food entities.
It embodies Aaron’s suggestion that second-rank cities become “back offices” for more expensive locations. A fair chunk of the growth there is not new companies coming in, but consolidation of functions from elsewhere. And the state entities.
…and state laws in Ohio made this happen and prevented other things happening in Columbus? I’m very skeptical of such conspiratorial theories. The world just can’t be manipulated that easily.
This is interesting, and I understand it personally. I was an new aerospace engineer who had moved out to SoCal in the waning days of the Cold War, and quickly determined that it was far too expensive a place for live (I think an engineer should be able to afford a house close to his place of work!), and moved to a place with a normal cost of living, even at a lower nominal salary. Since then, it has gotten far worse, and engineers, being the practical folks that we are, are hard to keep around these high-cost areas without accompanying high salaries (which Google et al can do, but not Boeing).
So is Boeing shifting programs to St. Louis from Seattle and Long Beach?
Having lived in four of the “elite” coastal areas- Vancouver, Seattle, Portland and SF, I can attest that one of the reasons lower income can’t or don’t move here is that these places aren’t sprawl machines. Unlike Houston, they don’t have a major real estate sector dedicated to building roads, homes and shopping centers. Sacramento was that kind of place in the mid 2000s, and there were entire exurban communities 20 miles out of town that were almost entirely Hispanic sheet rockers and other construction workers, living good lives while the construction lasted. After 2008, those communities were blown away.
I’m not sure that the new tech workers here in Seattle are all that transient. It is true that many have moved from the Bay Area, where housing is almost twice as expensive as here, but many seem to be slimming down their housing expectations, buying older, 1950’s housing on the north side of town about 20 minutes from South Lake Union, and are putting their kids in public schools. The wave of those new tech parents is not-so-subtly pushing Seattle Public Schools, for years notoriously bad or at best mediocre, into much better instruction. The first ring suburbs here, like Bellevue and Redmond, are also awash in new multi-family construction, much of that leased to visa workers from other countries, many of whom are used to living in higher-density housing. Last week I stopped in at a vertical shopping center, in which there was a Trader Joe’s downstairs and an REI store upstairs. That project would not have been built here ten years ago, but now developers are figuring out how to add the necessary parking and escalators, and customers are finding their way there. All in all, things are becoming a lot more like Vancouver! Yes, there is gentrification here, but it is nowhere near the issue that is in the Bay Area, and meanwhile I am finding young tech workers born in Sri Lanka and India on the far outer beaches of the Olympic Peninsula, trying out their new hiking and outdoor gear. I’m not so sure they’re leaving. After all, there are now dozens of Indian supermarkets scattered throughout this region. A lot of other places in the world have challenging density and housing pricing issues. I’m not so sure that, given an interesting place to live, complete ethnic communities around them, and good job prospects, many of these people are going to move elsewhere. 20 years ago Bellevue was known as a very vanilla place. Today one third or more of its households speak a language other than English at home.
P.S. Chris: Boeing is shifting work from Seattle to Charleston, not St. Louis. The issue is labor training, and the Carolina’s have better programs.
SC is a ‘right to work’ state.
The union is an issue in Seattle, and the costs are lower in South Carolina, but there is a very highly trained workforce here, and over the last 20 years management seems to have largely given up on working with labor, the same way that Detroit gave up on working with the U.A.W. Yes, “divorce” is a two-party affair, but that means both parties are responsible for the failure of the partnership. Other countries like Germany have very good relations between labor and management, to the extent that their partnerships are the basis for much industry training. If we as a country are ever going to have a well-trained workforce, we are going to have to get past our 100-year-old history of companies relocating for ever-cheaper wages. After all, the logical conclusion of that is operating in places like Haiti and Afghanistan. Even China is now realizing that it needs a smarter, better trained workforce skilled in programming and problem-solving, people who augment and control robots
Thanks for the update Rod. And I fully agree about pursuing the race to the bottom much further. Apparel is one thing; flying cigar tubes are another entirely.
“Other countries like Germany have very good relations between labor and management, to the extent that their partnerships are the basis for much industry training. If we as a country are ever going to have a well-trained workforce, we are going to have to get past our 100-year-old history of companies relocating for ever-cheaper wages.”
Daimler, BMW, and Volkswagen factories in RTW Alabama, Tennessee, and North Carolina indicate that they aren’t immune from relocating to reduce fully-burdened labor and regulatory costs.
Another equally compelling set of arguments for auto-plant locations has nothing to do with labor: even if labor costs are the same, the SUV imported from Germany costs more than the one built in the US strictly on transport/logistics/import costs, as well as the capital cost of carrying a longer pipeline of completed vehicles. The US-built vehicle also provides a hedge against nationalism and protectionism.
These factors help to explain Toyota, Honda, Subaru, and Mitsubishi building US plants in Ohio, Indiana, Kentucky and Illinois during the last auto trade war. (Mazda was in Michigan for 20 years.) Note that all those are states with significant historic Big Three production. Only the newer Kia and Hyundai plants went to the Deep South.
In fact, Subaru builds 40% of its worldwide vehicle total (and the majority of its US sales) in Indiana. Honda and Toyota also assemble vehicles in Indiana; all three built their operations here before Indiana became a RTW state in recent years. The state’s UAW history didn’t scare any of them away.