Thursday, April 12th, 2012

US Metro Population Growth Slows

Last week the Census Bureau released 2011 county and metro area population estimates that showed overall slowing population growth and particularly showing slow to halting growth in exurban counties. I’ll come back to the exurbs in a minute, but first a look at a map of metro area growth last year:

Metro area percent change in population, July 1, 2010 to July 1, 2011. Source: Census Estimates via Telestrian

Here’s the county map:

County percent change in population, July 1, 2010 to July 1, 2011. Source: Census Estimates via Telestrian

Someone once said to me about Chicago’s Mayor Daley that if he did something you liked, he was a visionary genius leader, but if he did something you hated, he was a corrupt machine dictator.

That seems to be how too many urbanists view the Census Bureau.

Back in the 90s when the Census estimates showed cities growing more slowly than boosters believed, they pressured the Census Bureau into adjusting the estimates to provide higher values. As it turned out, in most cases even the original estimates for cities proved inflated. In fact, the 90s were actually better for a lot of major cities than the 2000s were (e.g, New York, Los Angeles, and Chicago). This led to a new narrative that the Census had undercounted cities somehow.

Now this new data shows slowing exurban growth. All of a sudden, the Census Bureau has become once more a source of Gospel Truth, and I’ve seen many articles suggesting that the exurbs are dead, killed by rising gas prices and new Millennial preferences.

Let’s not get ahead of ourselves here.

Yes, exurban growth slowed recently. While cities on the whole fared more poorly than expected in the last census, we did see strong growth in downtowns and adjacent areas. I myself wrote about improving migration trends for core cities. That’s good news worth celebrating for cities. But don’t overstate the case.

I have a different though admittedly speculative take on the exurbs. I think a chunk of the fringe migration was from very low end home builders skipping out beyond established jurisdictions into unincorporated territory with few buildings restrictions. They threw up dirt cheap homes there and often sold them to people with marginal credit and income who had no business buying homes, using a variety of gimmicks to do so. (I know someone who sold homes for one of these builders, so I heard about some of these). Loose credit policies and government guarantees fueled this. The housing crash killed this market. Now that subsidies for this type of growth aren’t available, so that market is probably never coming back.

But when the economy improves and the market normalizes, I’d expect some level of suburbanization to resume. Part of the logic is simple math. If you add up the population of the municipalities of New York City, Los Angeles, Chicago, Philadelphia, San Francisco, Boston, Seattle, Washington, Portland, and Miami you only get 20 million people. That’s only about 20% of what the Census Bureau is projecting for just population growth by 2050. With the difficulties of building in urban areas, there’s no way to accommodate just the new growth even if everybody wanted into the city. In other words, there’s just no way there is going to be some massive back to the city movement. I hate to break it to you, but that’s reality.

Here’s the full list of large metros, sorted by population growth percentage:

Row Metro Area 2010 2011 Total Change Pct Change
1 Austin-Round Rock-San Marcos, TX 1,728,247 1,783,519 55,272 3.20%
2 Raleigh-Cary, NC 1,137,297 1,163,515 26,218 2.31%
3 Dallas-Fort Worth-Arlington, TX 6,400,511 6,526,548 126,037 1.97%
4 San Antonio-New Braunfels, TX 2,153,891 2,194,927 41,036 1.91%
5 Houston-Sugar Land-Baytown, TX 5,976,470 6,086,538 110,068 1.84%
6 Charlotte-Gastonia-Rock Hill, NC-SC 1,763,969 1,795,472 31,503 1.79%
7 Denver-Aurora-Broomfield, CO 2,554,569 2,599,504 44,935 1.76%
8 Washington-Arlington-Alexandria, DC-VA-MD-WV 5,609,150 5,703,948 94,798 1.69%
9 Miami-Fort Lauderdale-Pompano Beach, FL 5,578,080 5,670,125 92,045 1.65%
10 Oklahoma City, OK 1,258,111 1,278,053 19,942 1.59%
11 Salt Lake City, UT 1,128,269 1,145,905 17,636 1.56%
12 Seattle-Tacoma-Bellevue, WA 3,447,886 3,500,026 52,140 1.51%
13 New Orleans-Metairie-Kenner, LA 1,173,572 1,191,089 17,517 1.49%
14 Orlando-Kissimmee-Sanford, FL 2,139,615 2,171,360 31,745 1.48%
15 Riverside-San Bernardino-Ontario, CA 4,245,005 4,304,997 59,992 1.41%
16 Nashville-Davidson–Murfreesboro–Franklin, TN 1,594,885 1,617,142 22,257 1.40%
17 Atlanta-Sandy Springs-Marietta, GA 5,286,296 5,359,205 72,909 1.38%
18 Portland-Vancouver-Hillsboro, OR-WA 2,232,896 2,262,605 29,709 1.33%
19 Tampa-St. Petersburg-Clearwater, FL 2,788,151 2,824,724 36,573 1.31%
20 Phoenix-Mesa-Glendale, AZ 4,209,070 4,263,236 54,166 1.29%
21 San Jose-Sunnyvale-Santa Clara, CA 1,841,787 1,865,450 23,663 1.28%
22 San Diego-Carlsbad-San Marcos, CA 3,105,115 3,140,069 34,954 1.13%
23 San Francisco-Oakland-Fremont, CA 4,343,381 4,391,037 47,656 1.10%
24 Indianapolis-Carmel, IN 1,760,826 1,778,568 17,742 1.01%
25 Sacramento–Arden-Arcade–Roseville, CA 2,154,583 2,176,235 21,652 1.00%
26 Minneapolis-St. Paul-Bloomington, MN-WI 3,285,913 3,318,486 32,573 0.99%
27 Columbus, OH 1,840,584 1,858,464 17,880 0.97%
28 Jacksonville, FL 1,348,702 1,360,251 11,549 0.86%
29 Las Vegas-Paradise, NV 1,953,927 1,969,975 16,048 0.82%
30 Los Angeles-Long Beach-Santa Ana, CA 12,844,371 12,944,801 100,430 0.78%
31 Richmond, VA 1,260,396 1,269,380 8,984 0.71%
32 Louisville/Jefferson County, KY-IN 1,285,891 1,294,849 8,958 0.70%
33 Boston-Cambridge-Quincy, MA-NH 4,559,372 4,591,112 31,740 0.70%
34 Kansas City, MO-KS 2,039,766 2,052,676 12,910 0.63%
35 Memphis, TN-MS-AR 1,318,089 1,325,605 7,516 0.57%
36 Baltimore-Towson, MD 2,714,546 2,729,110 14,564 0.54%
37 New York-Northern New Jersey-Long Island, NY-NJ-PA 18,919,649 19,015,900 96,251 0.51%
38 Philadelphia-Camden-Wilmington, PA-NJ-DE-MD 5,971,589 5,992,414 20,825 0.35%
39 Chicago-Joliet-Naperville, IL-IN-WI 9,472,584 9,504,753 32,169 0.34%
40 Milwaukee-Waukesha-West Allis, WI 1,556,953 1,562,216 5,263 0.34%
41 Virginia Beach-Norfolk-Newport News, VA-NC 1,674,502 1,679,894 5,392 0.32%
42 Birmingham-Hoover, AL 1,129,068 1,132,264 3,196 0.28%
43 Cincinnati-Middletown, OH-KY-IN 2,132,415 2,138,038 5,623 0.26%
44 St. Louis, MO-IL 2,814,722 2,817,355 2,633 0.09%
45 Pittsburgh, PA 2,357,951 2,359,746 1,795 0.08%
46 Hartford-West Hartford-East Hartford, CT 1,212,491 1,213,255 764 0.06%
47 Rochester, NY 1,054,723 1,055,278 555 0.05%
48 Providence-New Bedford-Fall River, RI-MA 1,601,065 1,600,224 -841 -0.05%
49 Buffalo-Niagara Falls, NY 1,135,293 1,134,039 -1,254 -0.11%
50 Detroit-Warren-Livonia, MI 4,290,722 4,285,832 -4,890 -0.11%
51 Cleveland-Elyria-Mentor, OH 2,075,540 2,068,283 -7,257 -0.35%

Topics: Demographic Analysis

25 Responses to “US Metro Population Growth Slows”

  1. Will T says:

    Fascinating. When you publish maps by county like these, is there any chance you could also publish them as cartograms showing population? I love the visuals, but am always left feeling like I have no idea what’s happening in Manhattan, and I often forget exactly which county cities like Portland, Phoenix, and the Research Triangle cities are in.

  2. Unfortunately I don’t have cartogram capability, but I’d love to add it at some point. My software will let you generate a map that’s as large as you want though (or generate state maps)

  3. Chris Barnett says:

    Subtitle: Sprawling Sunbelt Metros Continue Fast Growth; Midwest State Capitals Continue to Lead Region.

  4. jbcmh81 says:

    I disagree with the idea that there can’t be a “massive” move back to the city, because it’s all relative. If Cleveland added 50,000 people to its urban core over the next 10-20 years, that would be a pretty massive positive move to the city. I agree that the suburbs will always exist and they will not completely die, but at the same time, if the subsidization of suburban life no longer exists, at the very least, suburbs will have to be built with urban qualities in mind. Meaning that they need to have amenities, grocery, walkability and transportation, etc that you would typically only find in an urban environment. So while the suburbs are not dead, unplanned sprawl likely is for the most part. And this, ultimately, can only be bad news for the boom times of the Sun Belt, which relied almost entirely on suburbanization and sprawl to gain growth.

  5. Matthew Hall says:

    These are all rough estimates of metros in the midst of the ‘great reset’. With much more change to come to the economic and political dynamics of the U.S. these shouldn’t be taken evidence of long term trends.

  6. @jbcmh81, I think clearly you are going to see a trend towards more dense and quality suburbs.

  7. kantor says:

    I agree that the projected population growth (and I stress the word “projected”) cannot be accomodated all inside the major cities.

    However the USA do not consist only of big cities and suburbs (in any of the various incarnations); what if this population growth shifts towards the many historical small towns that have been deactivated in the past 20/30 years?
    After all not everyboy wants to live in a dense urban core, but the alternative might not be a cheap stucco palace in the high desert….

  8. Chris Barnett says:

    Kantor, the answer to your question is likely found in the last 100 years of US Census data.

    Some huge percentage of the US population is “urbanized”, and as one sees in Aaron’s sidebar promo graphic for Telestrian, the “small town” and rural US counties continue to empty out. The exceptions are typically those places with US military bases and state universities (large government spending).

    Small towns aren’t particularly sustainable (in the economic, not ecological sense) until a European or Asian carmaker drops in a factory, or if there is an accident of history where a major US company grew up in one (thinking of my favorite, Columbus, IN).

  9. DBR96A says:

    Somebody on a different message board broke down the numbers for the Pittsburgh MSA between the official 2010 Census results and the 2011 estimates, and noted that a net positive migration of +7,513 — +5,708 domestic, +1,805 international — was basically cut in half by a natural population loss of -3,864, for a total population increase of +3,461.

    It’s sad to see every MSA surrounding Pittsburgh still declining in population except for Morgantown, WV. And the MSAs around Lake Erie look like a rash on that first map. Better days have to be ahead.

  10. CityBeautiful21 says:

    @WIllT – the Research Triangle counties are Wake (Raleigh), Durham (Durham), and Orange (Chapel Hill). The high-growth collar counties are Johnston, Chatham, and to a lesser degree, Alamance.

  11. EJ says:

    Aaron: “@jbcmh81, I think clearly you are going to see a trend towards more dense and quality suburbs.”

    The Columbus, Ohio suburb of Dublin is definitely moving in this direction.

    Granted, their financial resources run quite a bit deeper than those of most other burbs, so a complete retrofit is very much possible for them. It may not be the case for many other suburbs and exurbs however, particularly those that were hammered by the foreclosure crisis and commercial vacancies and don’t have a large and diverse enough tax base to offset these losses and fuel a second act. Probably every metro has a couple of Dublins, but also many more suburbs and exurbs that are going to have a much tougher road ahead of them.

  12. Alon Levy says:

    Aaron, I think you’re missing the point with the contrarian remark that,

    Now this new data shows slowing exurban growth. All of a sudden, the Census Bureau has become once more a source of Gospel Truth, and I’ve seen many articles suggesting that the exurbs are dead, killed by rising gas prices and new Millennial preferences.

    Since the 1990s, the position of urbanists and of the Census Bureau itself has consistently been that the decennial census is flawed and inaccurate. The Census Bureau wanted to adjust estimates using sampling; the House Republicans overruled it because it would add to the representation of Democratic areas, and so the US got stuck with inferior apportionment data. Over the 2000s the Census Bureau started using the ACS as a more accurate replacement for the decennial census.

    It’s the 2010-11 ACS trend that urbanists are trumpeting now and the 2000-9 ACS trend that we trumpeted two years ago. (The 2010 ACS is adjusted based on the decennial census.) The ACS is consistent with other data, for example housing growth and housing prices; the 2010 census is not. It’s not credible that New York and Atlanta grew so little over the 2000s.

  13. Alon, the IRS tax statistics (which are hard numbers) continue to show net out-migration for central cities (such as those with co-terminus city-county boundaries). This is the key source used by the Census in their estimates of domestic migration, which show core out-migration continues. It isn’t hard to imagine slow or even negative growth where you have net out-migration. NYC and a few other cases with high volumes of hard to count international immigration might be an exception, but I’m not sure. If you look at the maps of population growth in Chicago during the Census, you see very strong growth right where you’d expect to see it. The losses where concentrated in overlooked areas of the city that seldom feature in urbanist news stories.

    Given that the Census Bureau has historically been susceptible to political pressure to adjust its estimates, I am staunchly opposed to using any sort of sampling during the decennial census, even if it would improve accuracy if done right. There’s simply no reason to believe that the Census Bureau would be able to resist outside pressure to produce favored results given that they’ve already done it with their annual estimates in the past.

  14. NYC and Chicago’s 2010 figures I can understand. Its Atlanta that baffles me- a city that seemed to have had rampaging gentrification during the 2000s, and with one of the largest gaps between the 2009 Census estimate and the 2010 result (I think something like 20 percent of its population, wasn’t it)?

  15. Robert Hurst says:

    Why the assumption that the market will “normalize” from this point in the direction of continued expansion? On what time-scale can this expansion be considered normal?

  16. Robert Hurst says:

    The list includes city centers, suburbs, even some rather distant exurbs, all lumped together.

  17. Alon Levy says:

    Banning sampling is a political abuse of its own; this way it’s simple to undercount a difficult area just by underfunding the count to the point that it’s too expensive to count everyone there. It’s a neat, ostensibly neutral rule that ensures a highly biased result, much like how “two Senators per state” is ostensibly neutral.

    Mind you, back when it was debated around 1999 they didn’t talk in terms of cities vs. suburbs, but rather poor black and Hispanic neighborhoods vs. everything else. And it’s those poor neighborhoods where the decennial census missed people. I don’t think the census undercounted the Upper West Side.

    The New York City housing survey, also a hard number, shows growth in New York, at about the same rate predicted by the ACS. In contrast, the IRS misses a large number of people, and thinks average household size is much lower than it actually is.

  18. Chris Barnett says:

    Alon, “growth in housing supply” stats are meaningless without occupancy percentage. It used to be that the USPS stats were “best” for this purpose, but I suspect that’s not true any more. Receipt of US Mail is no longer a universal occupancy register.

    Builder/developers are more likely to use ACS to confirm market “build” signals because the granular decennial census income-by-tract data are so maddeningly behind and so dated when finally released. Lenders and investors prefer “new” data. (HUD still does not have a new list of “qualifying census tracts” for its low-mod income programs, based on a tract’s poverty rate, two full years after the enumeration.) So it is not surprising in the least that NYC’s count would track ACS projections if builders and developers use that data set.

  19. Matthew Hall says:

    “The list includes city centers, suburbs, even some rather distant exurbs, all lumped together” Yes, that is what MSAs are and is the scale at which economies operate. Metros compete with each other.

  20. Robert Hurst says:

    Understood that metros compete with each other. But I thought this was about suburbs competing with cities within the same metro area.

  21. dpjvodf says:

    The predicted population growth won’t be in San Francisco, Seattle, NYC or any already packed city. It’s going to be in Utica, Harrisburg, Syracuse, Pittsburgh, Scranton, Toledo, Cleveland, Buffalo… places that are already built, with the capacity to support millions. I think you’re seeing the initial sparks of it now, it’s going to catch on over the next decade and a half.

  22. Alon Levy says:

    Chris, the housing survey suggests that occupancy rates have remained the same. (If I remember correctly, the decennial census disagrees and says vacancy has increased). This can also be confirmed by looking at rising rents.

    In the gentrified and gentrifying parts of New York, developers look at regulatory approval more than anything. There was a boom in construction permits in the early 2000s due to the Bloomberg administration, and presumably also because developers responded to the population increases of the 1990s.

  23. Steve says:

    As a previous post noted, Atlanta’s numbers were far lower in the 2010 census than the 2009 estimate. I think the census picked up on the large influx of newcomers, but it missed the substantial displacement of lower income African-American families. Like Chicago, Atlanta aggressively redeveloped its public housing projects, and the foreclosing crisis hit low income black areas of the city hard at the close of the decade. And remember that the households leaving the city are more likely to have children and have a larger houseold size on average than the households entering the city (singles, young couples, empty nesters, and so forth).

    Major population changes are indeed occurring in Atlanta’s city proper. Do not pay so much attention to the actual growth between 2000 and 2010 (only 1% from 416,000 to 420,000). Pay attention to the drop in the black majority from 61.4% to 54.0%. Pay attention to the drop in the poverty rate from 27.3% to 22.6%. Pay attention to the drop in enrollment in the city school system (I cannot get the number at the moment but I think it is in the 15% – 20% range). What happened in the past decade was a major inflow of relatively affluent, mostly (but by no means entirely) white newcomers offset by a major outflow of lower income African-American residents. For whatever reason, the census estimates seemed to have picked up on the former trend and totally missed the latter trend. The big transformation occurring in our center cities today is not so much that they are growing in absolute terms or as a percentage of the total metropolitan population, but that they are becoming home to a more affluent population as urban living has become more desirable. Whether that is good or not can be debated of course. And this new more affluent population will actually fight (total NIMBYism) to stop higher density developments in many cases.

  24. Bow says:

    Bit of topic here but have you seen some of the writes up about Indy’s “urban” super bowl”. I t has received rave reviews. It actually had people chatting about walkability and urban design.
    For those that doubt Indy’s strategy.Here are some result:.2 major conventions booked Indy after the Super Bowl. They both sited the urban design and facilities. Expect the Super Bowl back in Indy. The list of review goes on and on. Here is one that really caught my eye out of London’s Sky.,25212,13283_7467354,00.html

  25. david says:

    Ultimately, everyone is looking for confirmation of their choices. But beyond that, I care about this issue because most suburbs and exurbs are built to be entirely car dependent, and are designed in a way that this will always be the case. They will never be served by public transit even if the denizens want it because of their geography. This car dependence has downsides but no upsides. Most people use cars for various purposes such as vacation or visiting friends and family at some distance. There is no benefit to any human, however, by sentencing to automobile travel to buy the theoretical gallon of milk. There is simply no benefit to that mandate.

    Exurbs and Suburbs can be built in a way that local errands can be done without automobiles. They aren’t built that way now, but they could be. So drawing a distinction between city and exurb isn’t as meaningful as drawing a distinction between walkable environments and non-walkable ones. There are many places in cities where the environments are non-walkable and public transit is mediocre. These might as well be called exurbs.

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