Did the Recession Convince Americans to Move to More Productive Cities?
Last week, I wrote about the big disconnect between population and productivity growth in U.S. metro areas over the last decade. I found the two do not match up at all. I cautioned that the booming Sunbelt metros, many of which have seen rapid population growth with little improvement in their underlying productivity, were on a path to unsustainable or even fictitious “growth without growth.”
Over at Slate, Matt Yglesias picked up on this tension from the opposite direction – people aren’t moving to the places that are generating jobs. Indeed, the mobility of Americans, which hit new lows during the great crash, has yet to rebound substantially.
But what does the pattern of population and productivity growth look like since the Great Recession? In the wake of the crash, have population trends begun to line up more with underlying changes in the economy?
With the help of my Martin Prosperity Institute team, I looked at changes population and productivity growth since the crash period, from 2009 through 2012. The data for productivity – measured as the change in annualized rates of real economic output, or GDP per capita – come from the Bureau of Economic Analysis and cover the post-crisis years, 2009-2012. We compared these data to metro population growth over the same period, using data from the Census Bureau.* The data cover 381 metro areas nationwide.
The map above shows the average annual population growth by metro area for 2009 to 2012.
The pattern is a familiar one. The darkest blue areas span the Sunbelt and the Rocky Mountain West, while the yellow and green areas are in the Northeast, Midwest and Rustbelt, as well as metros hard hit by the housing crash, like Phoenix.
The table below shows the ten large metros (those with more than one million people) with the fastest annualized population growth rates between 2009 and 2012. Sunbelt metros dominate the list. The pattern here is mixed. Four of the top ten metros on the list also number among the top ten large metros in productivity growth. None of the top ten large metros make the top ten overall, for population growth across the country’s 381 metro areas. Austin, the top-ranked large metro in terms of population growth, ranks 2nd among large metros and 12th overall for productivity growth. The three others on both lists are also in Texas – San Antonio, Houston, and Dallas. This suggests that people may be starting to migrate to at least some of America's centers of productivity growth. But productivity growth remains more sluggish in at least half of the metros on this list. Orlando – a metro known to be hard hit by the housing crisis – ranks 8th in population growth and 35th in productivity growth among large metros (overall it comes in 183rd in productivity growth among all metros). Four of the top ten large metros in terms of population growth fail to make the top hundred metros overall in terms of productivity growth.
|Top Ten Large Metros with the Highest Annualized Population Growth (2009-2012)|
|Rank||Metro||Population Growth||Productivity Growth (Rank among large metros and overall)|
|1||Austin-Round Rock, TX||2.92%||5.85% (2, 12)|
|2||Raleigh, NC||2.30%||2.99% (13, 71)|
|3||San Antonio-New Braunfels, TX||1.99%||4.34% (6, 28)|
|4||Houston-The Woodlands-Sugar Land, TX||1.97%||4.07% (7, 33)|
|5||Dallas-Fort Worth-Arlington, TX||1.85%||3.97% (8, 35)|
|6||Washington-Arlington-Alexandria, DC-VA-MD-WV||1.84%||2.09% (30, 152)|
|7||Denver-Aurora-Lakewood, CO||1.77%||2.27% (26, 131)|
|8||Orlando-Kissimmee-Sanford, FL||1.73%||1.67% (35, 183)|
|9||New Orleans-Metairie, LA||1.66%||2.67% (20, 98)|
|10||Oklahoma City, OK||1.56%||1.95% (32, 166)|
The pattern is more pronounced when you open things up to all metros, regardless of size. One obvious outlier is the area with the fastest rate of population growth: The Villages, Florida – a micropolitan statistical area that bills itself as "Florida’s Friendliest Retirement Hometown" – which experienced a 3.88 percent rate of population growth between 2009-2012. It had a reasonable 4 percent rate of productivity growth, 34th in the nation. But second-place Kennewick-Richland, Washington, which saw its population grow 2.98 percent, ranked 263rd in productivity growth (0.75 percent). But as with large metros, a number of leading population growth metros have also seen significant productivity growth. Midland, Texas, which ranked 4th in population growth (2.47 percent), also ranked 8th in productivity growth (6.51 percent).
The second map (above) shows the growth rates of productivity – measured by GDP per capita – for America’s metro areas from 2009 to 2012. Nationwide, the metro average for annualized productivity growth was 1.74 percent over this period. Eighty-four percent of metros experienced positive productivity growth, while 47 percent experienced growth greater than the U.S. average. Just 40 percent of the metros that experienced productivity growth over this time period had population growth rates of over 1 percent.
What we're seeing here are the twin pillars of America’s economic recovery, around its leading knowledge and energy metros. But the map also indicates the recovery of some older manufacturing regions. There are blue patches in knowledge economy centers on both coasts: San Jose and the Bay Area more broadly, Portland and Seattle in the Pacific Northwest, and Boston on the East Coast. Blue is also prevalent across the Energy Belt, from southern Texas through to Louisiana and out in energy-related economies of the Plains and Rocky Mountain West. But other patches of dark blue appear in the center of the country, in manufacturing regions like Detroit that were hard hit by the crisis.
|Top Ten Large Metros with the Highest Annualized Productivity Growth (2009-2012)|
|Rank||Metro||Productivity Growth||Population Growth (Rank among large metros and overall)|
|1||San Jose-Sunnyvale-Santa Clara, CA||6.46%||1.35% (18, 72)|
|2||Austin-Round Rock, TX||5.85%||2.92% (1, 3)|
|3||Portland-Vancouver-Hillsboro, OR-WA||5.69%||1.24% (21, 85)|
|4||Detroit-Warren-Dearborn, MI||4.45%||-0.15% (50, 348)|
|5||Nashville-Davidson-Murfreesboro-Franklin, TN||4.40%||1.44% (15, 63)|
|6||San Antonio-New Braunfels, TX||4.34%||1.99% (3, 17)|
|7||Houston-The Woodlands-Sugar Land, TX||4.07%||1.97% (4, 18)|
|8||Dallas-Fort Worth-Arlington, TX||3.97%||1.85% (5, 23)|
|9||Louisville/Jefferson County, KY-IN||3.94%||0.63% (35, 192)|
|10||Seattle-Tacoma-Bellevue, WA||3.68%||1.32% (19, 74)|
The table (above) lists the large metros (again those with more than a million people) that had the biggest gains in productivity from 2009 to 2012. San Jose is first, followed by the knowledge hubs of Austin and Portland (belying its characterization as a slacker paradise). Interestingly, Detroit comes in fourth with a substantial 4.45 percent gain in productivity.
Again, the pattern is mixed. Four of the top ten metros in productivity – Austin, San Antonio, Houston, and Dallas – number among the top ten in population growth as well. Still only one of them, Austin, cracks the top ten in population growth overall. Detroit, not surprisingly, ranks near the bottom overall in population growth (348th in the nation), while Louisville ranks 192nd. Population growth has similarly lagged behind productivity growth in Portland, Seattle, and San Jose. Six of the top ten large metros in productivity growth fail to crack the top 50 in population growth overall. But still, these data suggest that over the course of the recovery there has been some rebalancing of population growth in light of underlying productivity growth.
Overall, the metros with the highest rates of productivity growth since the crisis were mainly smaller and mid-sized metros, including energy and rebounding manufacturing centers. The manufacturing metros of Kokomo and Elkhart-Goshen, Indiana, top the list, with productivity growth of 18.3 percent and 13.1 percent respectively. Both number among the slowest growing regions in population, ranking 308th (0.06 percent) and 251st (0.35 percent), respectively. San Jose is only the large metro than makes the overall top ten, ranking 9th in productivity growth across all U.S. metros. Four out of the ten metros with the highest productivity growth across the country experienced less than a 1 percent increase in population over the course of the recovery.
There is good news and bad news in the relationship between population and productivity growth since the Great Recession. The bad news is that overall – across all of America’s metropolitan areas – population and productivity growth remain largely disconnected. But this is less true than it was before. Over the recovery period, there was barely any correlation at all between the two (for the statistically minded, the correlation between the two was a paltry 0.16 and barely significant, though that is better than the complete lack of a relationship for the broader decade). The good news is that population is increasing in some of the metros which are seeing substantial productivity growth. As the housing market rebounds, this nascent movement could intensify as those stuck in underwater homes in lagging regions are finally able to sell and relocate to more productive parts of the country. One can only hope this trend can pick up steam as the recovery unfolds.
*CORRECTION: The original version of this post used unadjusted Census data from 2009. However, as our friend Aaron Renn over at Urbanophile reminded us, the definitions and borders of several metropolitan areas have changed since the 2010 Census. The population growth map and accompanying table and explanation have since been updated accordingly.
All maps by MPI’s Zara Matheson.