Migration Stories are Full of Myths
With New Census Data Comes New Explainers and Migration Just-So Stories
The Census Bureau has released its annual estimates of population and components of population change. Reflecting estimates of population change in 2014, these numbers always generate excitement (among massive nerds, that is) because they are supposed to be “intercensal” estimates, in the sense that they should be reasonable indicators of how the next census will pan out. They are also the most up-to-date information on local area population growth available, being released within a few months of the study period’s end.
I’ll avoid going on a tangent about why I prefer the American Community Survey over the Population Estimates series and concede to the zeitgeist that considers these estimates truly newsworthy and authoritative, and reliable best estimates for the areas in question. Rather, I want to take a look at some coverage of these reports and note how “Migration Mythology” gets perpetuated.
A caveat: while I’m going to criticize some writing on migration, I want to be clear, I don’t have anything against the authors. I want to make constructive comments that can help inform future writing on migration, not just be a downer, hammering whenever somebody writes something on migration that isn’t exactly correct. I appreciate all the work journalists do, and actually quite like the writing of all the folks I critique.
The Myth of Suburban Recovery
Did Urban Cores Ever Really Stand a Chance?
The Washington Post covered the new Census numbers by saying that “Americans are returning to the far-flung suburbs”. Their article suggested that “exurban” living seemed to go out of vogue for a few years, and that urban cores became more popular, while now that trend has reversed. This is part of a big Migration Myth I’ve discussed before: the Myth of an Urban Revival. Allegedly, Millennials and “Creatives” were driving the long-awaited American return to city centers, where we could finally live the high-density lives we’ve always been eventually destined to live, like those wonderful Europeans who figured everything out before us.
I’m being sarcastic here for a reason: when we talk about migration, people (including me!) have a mental model, an idea of where things are headed. For many writers over the last decade, some variation of the above was in the back of their mind. Americans, led by a more progressive generation, were abandoning suburbia at last.
Unfortunately, this was a bogus interpretation. The Washington Post article, to its credit, phrases things carefully, saying exurbs “seemed” to do worse, and many experts “believed” urban cores might prevail. But these guarded terms aside, the article catalogues a track record of systematic error by many migration experts.
The above chart makes the argument: look how exurbs declined around the recession, and look how urban cores did better! Americans were obviously preferring cities!
That interpretation is fraught with problems, however. First of all, the convergence begins around 2005: the exact time when many evangelists of the new urbanization claimed the US was so terribly suburban. During the bottom of the recession, from 2009 to 2010, exurban areas actually grew faster. Third, its notable that other suburban categorizations consistently did better than urban cores: there was never a time when the urban core grew faster than suburbia on the whole. Only by dissecting the metropolis into selectively curated subdivisions do we arrive at this “convergence.”
To their credit, the Washington Post also discussed migration. This is the real problem with the narrative: urban cores didn’t grow because Americans preferred them, they grew because fewer people died and more people were born. This matters a very great deal. Births and deaths are largely just reflections of demographics relating to age, as well as local social and cultural norms. Urban areas attract high-fertility immigrants and have more young people, so more babies, less funerals. Suburbs and exurbs have more older folk and indigenous Americans, so less babies, more funerals. Americans don’t prefer urban cores, residence there just happens to associate with fertility and youth.
When we look at migration, the story is clear. Sure, the net migration of urban cores improved, but urban cores haven’t drawn net migration in 15 years, while the suburbs have drawn net migration every year for 15 years.
The record is clear: during the recession, cities were slightly less distasteful. There was no great urban renewal. There was no great demographic shift. There is no suburban comeback. There’s just a trend that’s about where it’s always been. That makes a boring headline, but it’s nearer the truth than some exciting narrative of urban cores seizing the lead, then falling behind to resurgent exurbs.
The Myth of Big Explanations
Are Migrants Really Looking for Cheap Homes and Warm Winters?
The New York Times’ Upshot blog covered the Census data from a different perspective. Instead of focusing on suburbs versus cities, they talked about “The Giant Retirement Community That Explains Where Americans Are Moving.” This article is simultaneously less incorrect than the Washington Post and vastly more ridiculous.
By venturing generalized explanations for why people move, Upshot was venturing into dangerous territory. I’ve written about explicit reasons why people move before, but have really tried to hammer home many times that there is no central reason why. Migration is deeply impacted by personal networks, personal migratory history, job opportunities, culture and language, religion, news and information, policies, and, of course, weather and housing: big, over-arching explanations are very likely to be wrong, wrong, wrong.
The Upshot focuses on two key variables of interest: weather and home prices.
This graph shows population growth by temperature. Now it’s true that there’s some evidence that migrants do care about the weather. But it’s inconsistent: people move both to and away from Florida and Michigan for the weather. Depending on the individual inclinations, seasonal preferences, and personal schedule, different states have “good” weather. Less than 2% of migrants cite weather as the reason for migration. Furthermore, as I’ll show further below, the overall trend of migration in the US has essentially zero correlation with weather.
But of course, this graph doesn’t track migration. It looks at population growth. Just like the Washington Post article, the New York Times talks about population without thinking about the theory. Warm states have higher population growth partly because of large immigrant populations that tend to be higher fertility and younger, and also due to proximity to coasts and the Mexican border, leading to higher international migration. Migrants from Central America or Mexico are actually moving into colder climates. Meanwhile, northern states may be colder, but they’re also older, more land-locked, and have fewer immigrants: so lower fertility.
Unless the New York Times is arguing that babies are created by warm weather, the graph above must be seen as a fun but meaningless correlation. Now, it’s true that migration does have some casual association with weather, and may matter for retirees. But retirees are a small share of migrants overall.
Regarding home prices, the Upshot gets the broad points right: high prices bad, low prices good, unless low prices mean depressed economy. Although very low prices can be stimulative if they are partly based on stigma unassociated with productive capacity. But that’s arcane and boring.
Where the Upshot errs, like most migration commentary, is in seeking a general theory of migration. The Upshot wants to explain migration to you, so that you can have The Explanation. But that’s not how migration works. An infinite variety of factors influence migrants, and ultimately simple push-pull factors are less valuable indicators than network dynamics. You can’t talk about migration as if there’s some cohesive pool of migrants weighing a fixed set of variables. Migrants have bounded information and make decisions within a given network: they’re as likely to be influenced by family and alumni ties as taxes or home prices.
Trust the Census Bureau
Where are Americans Going and Has It Changed?
I’m a bit of a nerd, and so my affection for all things Census will be well known to readers. Unsurprisingly, the Census has done a very good job summarizing its own work. They discuss how migration has changed in the excitingly titled “Moving in the USA: Domestic Migration Before and After the Recession.” That’s a headline, people.
They produce two very interesting maps enabling us to see how migration has really changed. The first focuses on counties that have gained through migration.
Dark blue areas were “losers” in 2006 and “gainers” in 2014, in terms of migration. As we can see, there are some places in the country with real “turnaround” stories. North Dakota, Montana, and West Texas have all benefited from the oil boom. Eastern Texas as vibrant cities experiencing growth, South Florida has apparently something good happening, California’s record is improving, Maine and Massachusetts are seeing new pockets of strength, and throughout the plains, midwest, and midsouth, there are individual counties experiencing gains. There’s no clear “trend” to the data, but we can see very clearly how things have changed.
This map looks at areas that lost people. Dark blue are counties that gained in 2006 but lost in 2014. Again, there’s not a clear nationwide trend to be picked out, but a few regions stand out. South Georgia, the arc from southern Missouri to western Arkansas to central Louisiana, rural mountain regions out west, virtually all of Illinois, and mid- or low-density partsNew England have all seen worsening migration records. These regions all have different stories and all face different challenges: none of which got highlighted in commentary seeking a single, generalized explanation.
New migration data releases inevitably provoke a new wave of explanations for that migration. Everybody wants a trend, a story, an explanation. But the truth is, there are huge variations by region. America doesn’t have a migration story: it has innumerable migration stories. And the release of one additional year of data does not radically change the picture. For the most part, “trends” are trends because they go on for a long time. When we look at this year’s data, the most valuable conclusions won’t come from the big explanations, but rather from the individual stories of communities that have experienced major shocks.
PS- Astute readers will note I didn’t toss any stones at the Wall Street Journal. That’s because their coverage was pretty “just the facts” oriented. And, being perfectly honest, I’m heavily biased by the fact that they cited some of my work, so it’s hard for me to make a neutral criticism.
Start the series from the beginning!
If you like this post and want to see more research like it, I’d love for you to share it on Twitter or Facebook. But what’s just as valuable for me is if you click the recommend button at the bottom of the page. Thanks!
Follow me on Twitter if you want to see my daily reading. Follow my Medium Collection at In a State of Migration if you want updates when I write new posts. And if you’re writing about migration too, feel free to submit a post to the collection!
I’m a grad student in International Trade and Investment Policy at the George Washington University’s Elliott School, and an economist at USDA’s Foreign Agricultural Service. I like to learn about migration, the cotton industry, airplanes, trade policy, space, and faith.
My posts are not endorsed by and do not in any way represent the opinions of the George Washington University nor the United States government or any branch, department, agency, or division of it. My writing represents exclusively my own opinions. I do not receive any financial support or remuneration from any party for this research.
