Domestic Migration

Matt slaps down Wendell Cox for saying that New York City has been losing people because it’s been experiencing net domestic outmigration. Immigrants and people born in cities don’t count, apparently (perhaps we can count then as three-fifths of a real American, or something?).

In fact, Cox’s assertions are lamer even than Matt makes them out to be. To begin with, he ignores the reason for the outmigration in places like New York City — housing is expensive. I have linked to this Glaeser paper more times than I can count; in it, he explains that the latest wave of migration to the Sunbelt is primarily about housing costs. Another way of saying this is that too many people want to live in New York — enough so that the city can’t build enough housing, which makes housing prices go up, which encourages outmigration. Even so, the city has added over a quarter of a million people since 2000.

Meanwhile, it looks as if a substantial share of the past decade’s migration to the sunbelt was illusory — a “migration bubble” in William Frey’s words. The latest Census data tells the tale. From 2008 to 2009, Nevada experienced a net domestic outmigration of about 4,000 people. Florida lost over 30,000 people on net — the second year in a row there was a net outflow. Inmigration to Arizona was about one tenth its 2006 level. Meanwhile, the District of Columbia had a net gain of domestic migrants of 4,500. The biggest improvements in net migration numbers came in California, New York, and New Jersey, respectively.

And central cities continue to see improvement in population trends. Census recently reported that Philadelphia, long population loser, was once more gaining people. Have a look at this Brookings chart on trends in domestic migration. There are more like it here.

The data are very clear. Since the middle of the last decade, domestic migration numbers have steadily improved for old population centers on the coast and for urban cores within those centers. If these trends continue, New York, Chicago, Los Angeles, Philadelphia, and Boston will join Washington and San Francisco in experiencing net inflows of domestic migrants by the middle of the coming decade.


  1. bottomofthe9th says:

    I think it’d be more accurate to say that migration to real-estate-bubbly parts of the Sunbelt was illusory. Atlanta, Dallas, and Houston (which are, after all, the three biggest cities in the Sunbelt) have continued to grow.

  2. ryan says:

    Texas is in a category by itself. Inmigration to other southern states, even the non-bubble ones like Georgia and the Carolinas, has slowed significantly.

  3. I think the reason why the sunbelt growth has slowed and the northeast/rust belt has started growing is that there’s house price parity, at least relative to job prospects. I grew up in Tampa and when I left for college in 1997, it was pretty cheap. I bought my first house in the Chicago suburbs in 2007 and houses in Tampa weren’t much cheaper (although the crash has been a lot harder there so the difference is bigger now). Given that I have access to much better paying jobs here, I feel like it would be a quality of life loss to move to Tampa at this point. I’d be much more likely to move to either a tech center for a specific job (I’m a programmer) or to a robust but inexpensive Midwest city like Indy or Columbus, OH.

  4. Say what you will, but New York is gearing up to lose yet another seat in Congress in the next reapportionment.

    We’ll see about migration when the economy improves.

  5. Peter, both Indianapolis and Columbus have both domestic and international net in migration. An Indianapolis Star analysis last year showed that since 2000 Indy metro had added international migrants at a rate even higher than Chicago, albeit from a much lower base.

  6. Alex B. says:

    New York STATE is gearing up to lose a seat. This discussion is focused on core city populations, and New York City’s population is growing. Meaning, if NYS is to lose a seat, it won’t be the fault of NYC.

    A quick look at the stats shows NYC with a 4.4% increase in population between the 2000 Census and the previous 2008 pop estimates. NY State, over that same period, grew by 2.7%.

    Anyway, Ryan is clearly talking about the impact this has on cities, not Congressional representation. And the trend for cities is overwhelmingly positive, despite the disingenuous spin of Mr. Cox.

  7. Aaron, where do you think I heard about how well Indy and Columbus are doing? Thanks, Urbanophile!

  8. Thanks for reading, Peter!

    Alex, Cox’s original piece was entirely about state migration patterns and had nothing to do with cities.

    Having said that, I myself will admit to having posted a comment on the original suggesting that cities like Indy, Columbus, and Kansas City are both domestic and international draws.

    As for New York City, everyone knows it’s New York City. It’s a unique case in America. Clearly, it is a powerful draw to people from all over the world. I don’t know that there are many conclusions we can draw about the rest of urban America based on NYC’s experience, however. Some cities are doing well, others are not.

  9. benamery21 says:

    Some observations on the Brookings report:
    Phoenix still added more domestic migrants than any other U.S. city in 2007-08, into the teeth of the housing bust and recession, albeit at a less torrid pace than earlier in the decade.

    Interstate domestic migration was down 38% from 2004-05 to 2007-08 or 2008-09.

    Total does not add to 100 due to rounding but the decline in migration appears to be: 37% housing, 20% marriage/household, 15% employment, 8% college, and 21% other. Obviously the decline in most of these things is driven by the economy.

    The most impacted types of moves by percent decline were marital status and household formation (-66%), followed closely by housing (-62%) and by college moves(-58%). Moves for other reasons were less impacted (-30%), and moves for employment declined relatively slightly (-17%).

    Contributing factors:

    34% of people moving to another state in 2004-05 did so for job related reasons. While this percentage has increased for 08-09, the total number of people moving interstate for work related reasons declined by 17% (accounting for 15% of the total decline in interstate migration).

    People changing states on entering or exiting college declined 58%, accounting for 8% of the total decline.

    People changing states due to a change in marital status or to establish their own household declined 66%, accounting for 20% of the total decline.

    People changing states primarily for housing related reasons declined 62%, accounting for 37% of the total decline.

    People changing states for all other reasons (2/3rds unspecified family reasons, like taking care of a parent or moving to be near relatives) accounted for 30.2% of 2008-09 interstate migration. The number of such moves declined by 30%, making up 21% of the total decline.