Sunday, July 31st, 2011

Replay: Migration Matters

Wendell Cox provoked a bit of a predictable tempest with his recent piece on migration, “The Decade of the South.” I suspect Cox relishes his role as provocateur-in-chief. So I’ll let him provoke me into summarizing some of the thoughts I’ve had on migration.

There’s a school of thought at that net domestic migration is the primary statistic of urban health. I believe I’ve even said something of the sort myself. In this logic, people are “voting with their feet” about which cities and states are implementing the best policies.

Of course, this creates a challenge for urbanists because the migration data is all away from major dense urban cores towards the South and various Sprawlville, USA’s. Hence when these numbers are thrown in their faces, they dismiss or discount them.

But people who don’t believe this tells the whole story are quick to understand the power of migration logic in other situations. Cities that are attractive to international immigrants are feted, as are those that hoovering up the “creative class”. Becoming a preferred migration destination for “talent” is a standard paradigm in urbanist development circles. And of course the slowing of migration to the Sun Belt and a nascent back to the city movement have been heralded by some as the end or reversal of previous trends.

So I think at some level, we all realize that migration matters. The question is, what is it telling us? I won’t pretend to have all the answers, but wanted to share a few thoughts around various aspects of this.

Vote With Your Feet

This seems to be the dominant paradigm, and there is clearly something to it. In this view, people decide to relocate based on various factors like taxes and regulation, cost of living, amenities, type of environment, economic opportunity, and more.

If you look at the Midwest, the cities that are doing well on most measures – Columbus, Indianapolis, Kansas City, and Minneapolis – also have net in-migration or are nearly neutral, while places like Detroit and Cleveland are losing people. Chicago is an interesting counter-example of a place that is conventionally viewed to be successful but has net out-migration. I’ll return to that in a bit.

We also see that talent hubs like Seattle, Austin, Portland, and Raleigh are seeing people move in. They are certainly out performing many other less sexy locales.

And Austin of course is in Texas, where huge numbers of people have moved. As much as we urbanists might not like it, the Texas story is real. The state has generated enormous numbers of new jobs, a total unmatched by other states. The same is true of many other southern places.

So clearly there is something to this notion of voting with your feet.

The Limits of Net Migration

But when we look behind the numbers, we start to see interesting patterns emerge. First off, migration isn’t uniform within states. Cox notes that Ohio had huge out-migration and categorized it as an “economic basket case”, yet Columbus has in-migration. Indiana is losing people, but Indianapolis is not. Nor are suburban areas near Louisville and Cincinnati. Given that different regions within the same state perform very differently, state level policies like tax rates and right to work laws can’t be the only answer.

So although “vote with your feet” has validity, perhaps there are other dynamics at work.

Greenfields vs. Brownfields

The South is frequently touted as a place whose favorable tax and labor climate is attracting business. But that is far from uniform. Places like Atlanta exploded into major business centers, other cities did not. In Tennessee, Nashville is thriving while Memphis is not. The south even has outright Rust Belt cities like Birmingham. Why are there so many struggling places in this nominally favorable environment?

Jim Russell suggests another key driver of the move to the south is the enormous advantages of greenfield development:

Bankrupt Birmingham is located in Alabama, a right-to-work state. Tennessee also has this distinction and is home to recovering industrial-dependent Chattanooga. Both the South and the North experienced an implosion of the manufacturing sector. Supposedly better state policies in the Sun Belt didn’t save any of those places from this fate. Why would it work magic in Pennsylvania for Erie?

In the face of globalization, states seem increasingly impotent. I suspect that most people, including the politicians, don’t understand the forces at work. I see a number of parallels with the brain drain issue. There is a poor accounting of the current situation. Ideological thinking dominates political discourse and policy suffers as a result.

During the 1980s, deregulation was the economic dogma of choice. The states with the least rules and lowest taxes should win. That’s not how it played out. If your city was saddled with high legacy costs, then state policy didn’t matter. Just so happens that the Rust Belt had (still has) a lot more of those cities than the Sun Belt did.

You can find the same economic geography within regions concerning the urban core and suburban periphery. Essentially, Americans have moved from brownfields to greenfields. When the unit of analysis is state, that trend gets lost in the data noise.

This is not to absolve bad decision making. High legacy costs in part derive from bad policy choices. But there is an independent logic to greenfield development. Because the world constantly changes but things like your physical location don’t, and your built environment and institutions are difficult to change, this means any city will eventually find itself facing the call of reinvention. Things like infrastructure and buildings constructed to outdated standards, institutional cruft, and deferred liabilities are almost inevitable, no matter how well run your city or state.

For more on this topic, I’d suggest reading my “What’s Killing California?” or Ryan Avent’s “Disruptive Technologies“.

Network Migration

There’s also the fact that migration, particularly international migration, involves networks. Why do people move? What factors into their decisions? A lot of times, factors other than purely economic considerations come into play, such as a personal connection to a place, proximity to family, or proximity of ethnic kinfolk.

One Louisvillian told me that “Our most important export is our women.” By that, he meant that college educated women who moved away from Louisville often returned with husband and family in tow when they had children. In effect, a significant source of new blood in Louisville is people who marry Louisville women and move back to be close to family. This illustrates both “boomerang migration” (return to a place you left) and family networks.

It also shows importantly that where you live is also a function of where you are in your life. I don’t think it’s entirely reasonable to expect or desire people to live in the same place from cradle to grave. What makes sense for a young single might not make sense for a family with kids or for a retired couple.

International migration works on family and ethnic networks as well and is often startlingly specific. For example, many Mexican towns have developed almost sibling relationships with US cities because migrants from that town clustered in a particular place. One example is the centralization of immigrants from the town of Tala, Mexico in Indianapolis. This was the subject of a Nuvo cover story “Bienvenidos a Talapolis“:

[Tala] Mayor Cipriano Aguayo, one of the immigrant pioneers to Indianapolis in the 1970s, returned to Mexico when he was able to save enough money to support his family. Aguayo still has two brothers in Indianapolis. According to residents of Tala, including the mayor, nearly every home in the small town has a family member in the United States and the great majority of those are in Indianapolis. The money they send back is extremely important for Tala’s economic sustainability.

“More money circulates in this region due to remittances than anything else,” Aguayo says, “so we depend a good deal on the cash that comes from Indianapolis.”

Some of it supports civic projects, an increasingly common trend in Mexico. Lucila Madrigal, a Tala municipal official whose sons live in Indianapolis, reports that emigrant savings helped finance a bridge, renovate the cemetery and pave streets in the county. Local families depend even more on money sent home by their “absent sons.”

While Tala’s immigrants to Indianapolis maintain strong familial, cultural and financial ties to Mexico, they have also helped build the sort of cultural foundations here in Indianapolis that both facilitate assimilation and maintain the ties that bind this transnational community together.

“We Mexicans aren’t used to being without some family nearby,” says Fabian Alonso, who divides his time between Tala and Indianapolis. “So we all try to get someone from here to move up there and that’s how we created our neighborhoods that replicate those in Tala.”

Maybe economic opportunity brought original pioneers to Indianapolis, but much of the rest of the migration was driven as much by network effects as economic logic. Otherwise why not just stay in Texas rather than keep going north? And from what I’ve read, many of Indy’s original international immigrants arrived more or less by accident or mistake. The roots of the Tala diaspora lay in the 1970’s, when Indianapolis was a backwater town trying to shed it’s “India-no-place” label.

There are plenty of other examples around. Fort Wayne has 3,000 Burmese, not because it is an economic hotbed, but because of an active local policy to bring in refugees.

The large, traditional American port of entry cities have both the most and easiest physical connections to other countries, as well as long established and robust migration networks. So it is no surprise they remain dominant in international immigration. This may not say much about whether that is “logically” the places immigrants should go.


Another source of migration is people retiring and moving someplace to enjoy it. As this often involves leaving places with lousy climates for more sunny locales, it is a built in source out migration from the Frost Belt. These people may be leaving for reasons that have nothing to do with the amenities, tax structures, or regulatory schemes that are favorable to business and labor. Rather, they want places that are good to enjoy their life and don’t place a financial burden on their particular financial profile.

Huge numbers of people moved to Phoenix, but does that mean Phoenix is a great business center, or just a great place to golf?

Network effects are also in play here. A retired couple I know from Pittsburgh just moved to North Carolina to be closer to their children who left years ago. This might have said something about what Pittsburgh’s economy used to look like, but it doesn’t have much bearing on what it is today. Economics might explain the children, but doesn’t explain the parents. Places that previously experienced an out-migration wave driven by a bad economy or policy might, for this reason, suffer a hangover for many years even if the local economy hits an inflection point.

Again, this is a result of stage of life, not per se local policy. Many people may choose to retire in place, but the reality is, many others won’t.

Moving Out, Or Not Moving In?

There’s also the question as to whether or not net out migration is more a reflection of people leaving or not coming. People are constantly migrating away from everywhere. The question is, are they being replenished? A place could have a below normal out migration rate, but if no one is moving there, it still shows net out migration.

This could particularly be the case in places that are near an inflection point. Bad conditions have stopped driving people out, but nothing has changed to create an inbound dynamic. Perhaps because there have been so few people moving in, there are few established inbound migration networks. Also, the reputation of the place (or lack thereof) may cause many to not even put it on the list. Portland gets touted as a great place in the media every day, so people who might be interested in what it has to offer will naturally have that city come to mind. That’s less likely to happen for Pittsburgh.

Jim Russell again argues that we should look mostly at in-migration, not net migration as the figure. I’m not sure to what extent I totally agree with this, but there seems to be something to it, particularly for cities at an inflection point.

New York City

Then there’s the case of America’s largest tier one cities, and especially New York. These cities have very high net domestic out-migration. Cox notes that the majority of New York state’s 1.65 million out migrants left from New York City.

But it isn’t hard to see that a place like New York is a structural exporter of people. First, it is a huge magnet for international immigrants. After a time in NYC, if some of them move on then, bang, they are domestic out migrants. On the domestic ledger that’s 0 in, 1 out, a clear net loss, but not the whole story.

Also, it’s noted that NYC takes in lots of young people, but many leave when they get marrie and have kids. Again, let’s play this scenario out. Two young singles move to NYC, marry, have two kids. When those kids reach school age, they move to the suburbs or back to the wife’s hometown in Kansas City. That’s 2 in, 4 out, a net loss of 2 people.

Is this a bad thing? I won’t suggest that we shouldn’t try to improve our urban schools or make our central cities more family friendly. Of course we should. But it strikes me that places like NYC are never going to be the destination of choice for families with school age children. Even if it captured more market share, it seems that it will always be exporting families.

Again, is it reasonable to expect that every place will be equally as attractive to people at all stages of their life? I don’t think so. Some places are better for young singles, some for families, others for retirees. So anyplace that is attractive to young singles is likely to be a structural source of out migration.

If you think again about New York, it takes in immigrants – raw recruits if you will – and spits out Americans. It takes in young singles – more raw recruits – and spits out up skilled people with families. There is huge value added in this. In a sense, New York City is a gigantic refinery for human capital. It’s a smelter for people. Perhaps we shouldn’t be any more sad about New York exporting people than we are about it exporting financial services. Taking in people, adding value, then exporting them is one of New York’s core competencies. Maybe we should be thanking it for providing this valuable service.

Whatever the case, any place that is a magnet for immigrants and young people, as most tier one cities are, are likely to be structural sources of out migration no matter what they do.

Jim Russell on Migration

It’s no secret I’m a fan of Jim’s thinking on talent and migration matters. His blog Burgh Diaspora is required reading, IMO. As I hoped he might, he chimed in with a comment here, which I’m including below so everyone will see it:

Interesting to see all the aspects of migration packed into one post. Still, some additional data categories should be considered. It’s not so much that migration matters, but demography matters. We tend to play fast and loose with net migration and population numbers. Declining population has become synonymous with net out-migration. That’s a gross distortion.

Each shrinking city has a unique demographic profile. Break it all down by age cohorts and the analysis gets very complicated. In my opinion, it also gets more useful. What’s the urban dependency ratio? A huge issue in cities with substantial legacy costs.

Concerning natural replacement rates, there is a striking variance among Rust Belt cities. Lower rates make the net out-migration problem more acute.

Overall, I think the net migration numbers tend to limit our thinking about how to make better cities. The population shift from Rust Belt to Sun Belt is dramatic, but the dominant narrative that folks such as Cox peddle won’t inform better policy. This is especially true if we insist on using state level data.

A good example is Texas, a big “vote with your feet” winner. A lot of good that’s done El Paso and other lesser-tier cities in the state. Houston, Dallas and Austin will Hoover up graduates from those places as well as the Rust Belt. They also suck up talent from the entire Sun Belt.

I think state policy is a red herring.

Migration, Properly Considered

In short, migration does matter. Any city that thinks it can be blasé about this is fooling themselves. On the other hand, surface numbers only tell us so much. We need to understand the dynamics going on underneath the hood.

This post originally appeared on January 7, 2010.

Topics: Demographic Analysis
Cities: Indianapolis, New York, Pittsburgh

11 Responses to “Replay: Migration Matters”

  1. Wad says:

    One other contribution to add to migration patterns is Bill Bishop’s theory of “The Big Sort.”

    People are moving to areas to be among like minds and like demographics. Bishop thinks it bodes ill for America.

  2. philip morris says:

    As always, interestng. But wrong about Birmingham. The city is solvent, though Jefferson County, which signed on to a bad sewer bond deal with JPMogan is threatened with bankruptcy (probably soon to be avoided). Very diversified economy, and the metro area modestly gaining population. Good school systems a mile from the urban core. Does not fit the model you project at all.

  3. MetroCard says:

    I agree that there’s more to the story than just economics, as push factors do change over time. Large cities are increasingly becoming lifestyle choices these days. Are jobs still a primary factor? Of course. But cities are evolving into what I’d almost call “niche” areas for different demographic groups. The media plays a big role in this, too–how many times have we seen those “Top Ten Cities for __________” lists floating around the Internet?

    As you touched on, the churn rate is very high in New York. Plenty of young families hightail it out of the boroughs for an anonymous subdivision in a fast growing state after the kids reach a certain age. Does this mean that New York is inherently bad for couples with children? No, perhaps it’s just that this particular demographic group might have reached a point where they feel they need a new “brand” or “niche” that works better for that lifestyle.

  4. Nice piece, but the retirement statement is way off base. There are far more nuances now in where retirees are moving (now the baby boomers). I would point you to the work of Dr. Peter Nelson at Middlebury College which demonstrate migration to many non-traditional areas (many non-metropolitan) such as the Upper Peninsula of Michigan, northern New England, the inter-mountain West, and northern Minnesota, of all places.

    There’s a nice discussion going on in the ULI group about this issue.

    Otherwise, you are correct that migration (esp. domestic migration) is a very important component that has both predictability (people change their location needs as they go through phases of a lifecycle) and much that has changed, especially with the recession/depression/downturn, etc.

    David Lindahl

  5. Greg says:

    Excellent take on NYC as a refinery of human capital.

    It’s amusing to think that some people might look at domestic migration data and think that New York is in decline or emptying out. The city’s economy is healthy and dynamic. The prosperity is spreading from Manhattan into new “hot” neighborhoods in Brooklyn and even Queens.

  6. George Mattei says:

    I think the legacy cost issue is a primary one for this topic. It affects everything. I think it would be fascinating to compare the growth of professional and technology jobs in Cleveland to that of Columbus or Indianapolis. I wonder if the growth of these sectors in Cleveland has been equal to or even superior to those cities. My guess is that if the industrial revolution had passed Cleveland by, we would be talking about the city in the same way we talk about Columbus or Indianapolis- a young rising star.

    I have compared Columbus to Cleveland in the past, and said maybe the biggest difference is that Cleveland was “unlucky” enough (if you call it that) to have come of age in the industrial era. Cleveland was a city of almost a million in the 1950’s. Columbus was a city of about 350,000 back then. Now Columbus is 780,000 and growing, while Cleveland is now just below 400,000.

    Perhaps more telling, the Cleveland metro area was one of only a few to lose population (albeit a very modest loss), while Columbus metro continues to add at a fairly healthy clip.

    When you look below the surface, you see some surprises-Cleveland has added many health-related and research jobs, centered around the Cleveland Clinic, the NASA Glenn research Center and Case-Western University. University Circle is a bona-fide medical-tech cluster that many cities would envy. They have the best record in the State of producing start-up tech companies. Other companies such as Progressive Insurance, Eaton Corp and regional bank HQ such as Key and Charter One provide well-paying professional jobs. Click here to see the list of their top 10-mostly tech and business professionals.

    Why with this profile is it losing jobs and people? Well, it grew of age in the industrial era, and there is a huge legacy cost, in terms of government, infrastructure and just overall economic activity, that’s hard to overcome.

    Give it another 25 or so years, and the big cities currently growing will slow, only to have a new round begin to spring up. Places like Atlanta and Las Vegas are already showing significant signs of stress.

  7. Perhaps we should be looking at net population change, rather than (or instead of) counting the moving vans? That would also account for international migration, births, deaths, and other things excluded from domestic migration statistics.

  8. Scotty, so-called “natural increase” (births minus deaths) is the other component of population change that needs to be examined. I think you want to disaggregate to see what’s going on across all variables.

  9. Alon Levy says:

    Sure, but don’t forget, natural increase is also influenced by where parents move before and after they have children. For example: Tokyo has the highest natural increase in Japan. Does it mean that people who live in Tokyo have more children, or that young couples move in, have a kid, and then move back out?

  10. Will says:

    “Given that different regions within the same state perform very differently, state level policies like tax rates and right to work laws can’t be the only answer.”

    I argued the same point – that there were outlying cities like Columbus, Indy, and the twin cities – to someone with libertarian leanings this weekend. He countered that it was because of government spending (universities and state capitals) that those cities were relatively more successful to the other ones…. :(

  11. Wad says:

    @Will, sounds like the person was saying “cheater, cheater, pumpkin eater” when it comes to capitals. IOW, if government has no legitimate right or need to exist, all economic activities by or from government by extension are illegitimate as well.

    Interestingly, the better off privately a city is, the less it is economically dependent on the prevalence of government. It’s the small-government areas that are often more income-dependent on that small government.

    You can take the 50 state capitals of the U.S. and group them into three categories: primate regions, dynamic city regions and passive regions.

    Primate regions (where the capital has the largest population and the largest economic activity and widely lead the second-largest region) and dynamic city regions make up about half of all capitals. These areas have enough nongovernmental activity around them to be economically resilient. The others couldn’t leave their cities without making them ghost towns.

    The primates: Honolulu, Denver, Indianapolis, Atlanta, Providence, Des Moines, Boise, Phoenix, Little Rock and Salt Lake City.

    The dynamic city regions: Sacramento, Austin, Columbus, Madison, St. Paul, Olympia, Raleigh, Richmond, Nashville, Oklahoma City, Columbia, Harrisburg, Albany and Hartford.

    The passives: Montgomery, Juneau, Dover, Tallahassee, Springfield, Topeka, Frankfort, Baton Rouge, Augusta, Annapolis, Lansing, Jackson, Jefferson City, Helena, Lincoln, Carson City, Concord, Trenton, Santa Fe, Bismarck, Salem, Pierre, Montpelier, Charleston and Cheyenne.

    There are some wobblers. Olympia would be passive were it not for its proximity to the Puget Sound MSA. Effectively, it’s part of metropolitan Seattle’s economy. Trenton, N.J., is notoriously in decline but it happens to be the worst house on the block of a great neighborhood, with New York City to the north and Philadelphia to the south. Both Lansing and Baton Rouge recorded population growth, but considering that their respective states’ largest cities recorded massive population declines, this is more of a sign of dependency than dynamism.

    Harrisburg and Albany have the appearances of dynamic cities, but there’s a case for putting them as passive. Harrisburg is actually growing, but it’s mainly a case of an emerging CMSA and possibly due to declines in Pittsburgh and other parts of the state. Albany has the same problem. The CMSA has more than a million people, but it encompasses a large area and must factor in the offset from population and economic losses upstate.

The Urban State of Mind: Meditations on the City is the first Urbanophile e-book, featuring provocative essays on the key issues facing our cities, including innovation, talent attraction and brain drain, global soft power, sustainability, economic development, and localism. Included are 28 carefully curated essays out of nearly 1,200 posts in the first seven years of the Urbanophile, plus 9 original pieces. It's great for anyone who cares about our cities.

About the Urbanophile


Aaron M. Renn is an opinion-leading urban analyst, consultant, speaker, and writer on a mission to help America’s cities thrive and find sustainable success in the 21st century.

Full Bio


Please email before connecting with me on LinkedIn if we don't already know each other.



Copyright © 2006-2014 Urbanophile, LLC, All Rights Reserved - Click here for copyright information and disclosures