Neighborhood Data Show That U.S. Suburbanization Continues (Wonkish)

In yesterday’s post, I used newly released Census population estimates for 2015 to show that suburban counties are growing faster than urban counties, and by a widening margin. In that post I noted that “… many counties, especially large counties, include both urban and suburban neighborhoods. …  Other data sources with greater geographic detail confirm that population growth is generally faster in suburban than urban neighborhoods, though not uniformly.” This follow-up post provides a closer look at trends in where Americans live, using much more detailed neighborhood data.

To track the trend in urban versus suburban (and rural) living, I looked at the share of U.S. households that live in Census tracts with different densities. For 1990, 2000, and 2010, I calculated the density, in households per square mile, in each Census tract, from the full-count decennial Census. (Neighborhood density is the best predictor of whether someone describes where they live as urban, suburban, or rural. An alternative method is to look at distance from the city center, but I like that measure less: for instance, three miles in one direction from downtown might be a lot more urban than three miles in the other direction; also, some metros are much more urban than others – which means, for instance, three miles from midtown Manhattan is much more urban than two miles from downtown Charlotte.)

In previous work, I showed that people in neighborhoods with at least 2,213 households per square mile typically consider their neighborhood to be urban; neighborhood density of 102-2,213 is perceived to be suburban, and less than 102 is rural.  Accordingly, I divided Census tracts into categories at those density breakpoints, and then added a few arbitrary breakpoints for additional granularity, for a total of eight categories.

Then, for each Census year, I calculated the percentage of households living in each density category for the U.S. overall. In addition to 1990, 2000, and 2010, I used tract-level household estimates from the 2014 five-year American Community Survey as a more recent datapoint, though the ACS counts are estimates and reflect an average of the five years from 2010 to 2014, whereas the decennial Census data are full household counts at a single point in time. (For the 2014 ACS estimates, I used tract densities calculated from the 2010 decennial Census.)

Here’s the punchline: Americans are getting more suburban, less urban, and less rural. The share of households in urban neighborhoods declined in the 1990s, 2000s, and 2010s to date. This is true whether we group all urban neighborhoods together (that is, all categories with density of 2,213 or more), or look just at the highest-density urban neighborhoods: the share of households in Census tracts with at least 10,000 households per square mile has declined, slightly but consistently, from 4.0% in 1990 to 3.9% in 2000, 3.8% in 2010, and 3.7% in 2014. Rural neighborhoods have also had a declining share of population, too. But suburban neighborhoods have all gained population share over the entire time period, with the exception of the higher-density suburban neighborhoods (1,500-2,212 households per square mile) in the most recent years (2010-2014). (Note that the % of households in each category reflects the breakpoints, some of which are arbitrary — what matters is the TREND in the % of households in each category.)

neighborhood density

Time for all the usual caveats. First: this analysis shows the broad national trend, but there are always some places – specific neighborhoods or even metros overall – that buck the national trend. Second: the composition of neighborhoods matters, not just the total number of people who live there. Some demographic groups – such as young adults with college degrees – have indeed become more urban. That group has been important in shaping a narrative about urbanization, but they are the exception, not the norm. Third: where people live isn’t necessarily what they “want.” The cost and availability of housing matter, too, which in turn is shaped by land-use regulations, housing subsidies, and other public policies. Suburban population growth, by itself, doesn’t prove that demand for suburban living is increasing — just as rising urban prices, by themselves, don’t prove that demand for urban living is increasing. And, fourth: from a policy perspective, it doesn’t matter so much whether cities are winning. Public policies that affect location decisions (like the mortgage interest deduction, or parking policies) should be based on externalities that arise from where people live, not on trends in where people are moving.

With those caveats, the main point is that America is becoming more suburban. Both up-to-date annual county population estimates (yesterday’s post) and neighborhood data that’s less current and less frequent (this post) show that same trend. While it is important to remember that, in theory, county-level analysis could give a misleading picture of overall urban and suburban population trends, it is also important to know that, in fact, it does not.

2015 Population Winners: The Suburbs and the Sunbelt

Local population growth trends are reverting to pre-2000 patterns as the housing bubble and its aftermath recede.

Today the Census Bureau released its 2015 population estimates for counties and metropolitan areas. After volatile swings in growth patterns during last decade’s housing bubble and bust, long-term trends are reasserting themselves. Population is growing faster in the South and West than in the Northeast and Midwest, and faster in suburban areas than in urban counties; both of these trends accelerated in 2015. Before getting deeper into these broad patterns, though, let’s start with the metros that saw the fastest growth and steepest declines in the past year.

Top Metro Winners and Losers

The Villages – a small metropolitan area near Orlando with lots of retirees – led all metros in population growth. Six of the ten fastest growing metros in 2015 were in Florida and Texas, while none were in the Midwest or Northeast.

Metros with the Fastest Population Growth
# Metro YoY population change, 2014-2015
1 The Villages, FL 4.3%
2 Myrtle Beach-Conway-North Myrtle Beach, SC-NC 3.5%
3 Cape Coral-Fort Myers, FL 3.3%
4 Midland, TX 3.3%
5 Odessa, TX 3.3%
6 Greeley, CO 3.2%
7 Austin-Round Rock, TX 3.0%
8 Bend-Redmond, OR 2.9%
9 Punta Gorda, FL 2.8%
10 Fort Collins, CO 2.7%
Among all metropolitan areas

Among large metros, Cape Coral – Fort Myers and Austin had the fastest growth. All of the fastest-growing large metros were in the South and West.

Large Metros with the Fastest Population Growth
# Metro YoY population change, 2014-2015
1 Cape Coral-Fort Myers, FL 3.3%
2 Austin-Round Rock, TX 3.0%
3 North Port-Sarasota-Bradenton, FL 2.7%
4 Orlando-Kissimmee-Sanford, FL 2.6%
5 Raleigh, NC 2.5%
6 Houston-The Woodlands-Sugar Land, TX 2.4%
7 Charleston-North Charleston, SC 2.4%
8 Provo-Orem, UT 2.4%
9 Lakeland-Winter Haven, FL 2.3%
10 Las Vegas-Henderson-Paradise, NV 2.2%
Among metropolitan areas with at least 500,000 people

Population growth in these metros was driven more by domestic migration than by international migration or “natural increase” (that is, births and deaths). All ten of the fastest-growing large metros had more in-migrants from the rest of the country than out-migrants. However, natural increase boosted population growth in metros with lots of young adults of child-bearing age (e.g. Provo-Orem UT) but was a drag on growth in metros with older populations (such as Cape Coral – Fort Myers and North Port – Sarasota – Bradenton). And among the large metros with the highest growth rate from international migration (Miami, San Jose, New York, Honolulu, Orlando, and Washington DC), only Orlando also had positive net domestic migration and therefore strong overall population growth. (Among all metros, the correlation between overall population growth and net domestic migration was 0.86; the correlation between overall population growth and natural increase was 0.40, and the correlation between overall population growth and international migration was just 0.19.)

Note the presence of oil towns on these lists. Houston was among the ten fastest growing large metros, and Midland and Odessa, TX, were among the fastest growing of all metros. Two other places with an energy boom – Williston and Dickinson, ND, were the fastest-growing “micropolitan” areas in the country. These Census population estimates are as of July 1 of each year, so the 2015 data are actually a snapshot from almost nine months ago. It’s highly likely that population growth has slowed in energy-producing areas since July 2015: more recent jobs data have shown that falling oil prices hurt energy-sector employment.

One last point about the fastest-growing metros: 2015 is the first year since 2010 when Austin was not in the top spot among large metros. Cape Coral – Fort Myers, which bumped Austin out in 2015, was also the leader in the bubble years of 2004-2006.

Metro with the Fastest Population Growth, by Year
Year Metro
2001 Las Vegas-Henderson-Paradise, NV
2002 Las Vegas-Henderson-Paradise, NV
2003 Las Vegas-Henderson-Paradise, NV
2004 Cape Coral-Fort Myers, FL
2005 Cape Coral-Fort Myers, FL
2006 Cape Coral-Fort Myers, FL
2007 New Orleans-Metairie, LA
2008 Raleigh, NC
2009 Provo-Orem, UT
2010 Colorado Springs, CO
2011 Austin-Round Rock, TX
2012 Austin-Round Rock, TX
2013 Austin-Round Rock, TX
2014 Austin-Round Rock, TX
2015 Cape Coral-Fort Myers, FL
Among metropolitan areas with at least 500,000 people

Now for the other extreme. Of the 381 metropolitan areas, 96 lost population in 2015 while 285 gained. Those with the steepest declines were all smaller metros, including several with military bases, spanning most regions of the country except the far West.

Metros with the Steepest Population Decline
# Metro YoY population change, 2014-2015
1 Farmington, NM -4.2%
2 Hinesville, GA -2.6%
3 Elizabethtown-Fort Knox, KY -2.2%
4 Pine Bluff, AR -1.1%
5 Watertown-Fort Drum, NY -1.1%
6 Decatur, IL -1.0%
7 Charleston, WV -1.0%
8 Albany, GA -1.0%
9 Saginaw, MI -0.9%
10 Wichita Falls, TX -0.9%
Among all metropolitan areas

Among larger metros, 13 lost population in 2015 while 91 gained. The ten with the steepest declines were all in Ohio, Pennsylvania, upstate New York, and Connecticut.

Large Metros with the Steepest Population Declines
# Metro YoY population change, 2014-2015
1 Youngstown-Warren-Boardman, OH-PA -0.7%
2 Scranton–Wilkes-Barre–Hazleton, PA -0.3%
3 Pittsburgh, PA -0.2%
4 New Haven-Milford, CT -0.2%
5 Syracuse, NY -0.2%
6 Rochester, NY -0.2%
7 Cleveland-Elyria, OH -0.2%
8 Hartford-West Hartford-East Hartford, CT -0.2%
9 Toledo, OH -0.1%
10 Buffalo-Cheektowaga-Niagara Falls, NY -0.1%
Among metropolitan areas with at least 500,000 people

Youngstown had the sharpest losses among large metros for the past six years and for 11 of the past 15 years. Metro Detroit had the biggest losses in 2008 and 2009; in 2015 Detroit was 14th from the bottom among large metros, with a growth rate of just 0.01%.  New Orleans had the steepest population decline of all large metros in 2006 thanks to Hurricane Katrina (the 2006 data reflects the year leading up to July 1, 2006, which includes when Katrina hit in August 2005). The early 2000s tech bust put San Jose at the bottom in 2002.

Metro with the Steepest Population Decline, by Year
Year Metro
2001 Youngstown-Warren-Boardman, OH-PA
2002 San Jose-Sunnyvale-Santa Clara, CA
2003 Youngstown-Warren-Boardman, OH-PA
2004 Youngstown-Warren-Boardman, OH-PA
2005 Youngstown-Warren-Boardman, OH-PA
2006 New Orleans-Metairie, LA
2007 Youngstown-Warren-Boardman, OH-PA
2008 Detroit-Warren-Dearborn, MI
2009 Detroit-Warren-Dearborn, MI
2010 Youngstown-Warren-Boardman, OH-PA
2011 Youngstown-Warren-Boardman, OH-PA
2012 Youngstown-Warren-Boardman, OH-PA
2013 Youngstown-Warren-Boardman, OH-PA
2014 Youngstown-Warren-Boardman, OH-PA
2015 Youngstown-Warren-Boardman, OH-PA
Among metropolitan areas with at least 500,000 people

 

Population Trends Favor the Sunbelt and the Suburbs

The lists of the fastest and slowing growing metros hint at general patterns in recent population growth. Looking at all counties in the U.S. – not just those in metropolitan areas – reveals three trends.

The first is the accelerating shift of population toward the Sunbelt. Among the four Census regions, the South and West both had population growth of 1.2% in 2015, far ahead of the 0.2% growth in both the Northeast and Midwest. Population growth in the South and West has outpaced that in the Northeast and the Midwest for decades, as well as in each year since 2000 throughout the housing bubble, bust, and recovery. The gap narrowed somewhat after the bubble burst, as population growth quickened in the Northeast between 2008 and 2012. Since 2013, however, population growth in the South and West has accelerated, while growth in the Northeast and Midwest has slowed – thus widening the gap between those two Sunbelt regions and the rest of the country.

four regions

The second trend is the recent slowdown in population growth in urban counties (defined as those with tract-weighted density of at least 2000 households per square mile). Both higher-density suburban counties and lower-density suburban counties had faster population growth than urban counties in 2015, and the gap between suburban and urban county growth was larger in 2015 than in 2014. In short, suburbanization accelerated in 2015.

While population growth in urban counties has clearly recovered from the housing bubble, during which urban counties lagged for many years and even lost population in 2006, the rebound in urban population growth was brief. Urban counties outpaced all other areas only in 2011, and urban growth in 2015 slowed to its lowest level since 2007.

Growth in small towns & rural areas – the lowest-density counties – remained behind that of urban, higher-density suburban, and lower-density suburban counties in 2015, even though small towns & rural areas grew in 2015 at the fastest pace since 2010.

Keep in mind that many counties, especially large counties, include both urban and suburban neighborhoods. The Census will release sub-county population estimates for 2015 later this year. Other data sources with greater geographic detail confirm that population growth is generally faster in suburban than urban neighborhoods, though not uniformly: for instance, high-density downtowns have grown faster than moderately dense suburban neighborhoods though slower than the lowest-density suburbs.

county density quartiles

The third trend is that metropolitan areas with at least one million people grew faster in 2015 than midsize and smaller metros did, just as in every year since 2008. While this is a reversal of the bubble years in the early 2000s, when midsize metros grew faster, it is a return to the pre-bubble pattern: in the 1980s and 1990s, as in the post-2008 period, population growth was faster in million-plus metros than in midsize metros, smaller metros, and non-metropolitan areas. (Micropolitan areas counted as metros in this analysis.)

metro population

It might seem surprising that urban counties are growing more slowly than suburban counties even though larger metros are outpacing smaller metros: after all, the most urban counties in America are in the large metropolitan areas of New York, Boston, Washington DC, and San Francisco. In fact, though, there’s no contradiction. Even the largest metropolitan areas typically include both urban and suburban counties. Most of the fastest-growing counties in America are suburbs in large Sunbelt metropolitan areas. Of the six counties (among those with at least 50,000 people) where population grew at least 4% in 2015, five are suburbs of large Sunbelt metros: Hays county (in Austin metro), Broomfield (Denver), Comal (San Antonio), Fort Bend (Houston), and Forsyth (Atlanta). (The sixth is Sumter FL, the county that constitutes The Villages metropolitan area.)

 

The Longer View: Population Trends Are Getting Back to Old Patterns

Since 2000 population trends have reflected the housing boom, bust, and recovery. The boom, lasting until 2006, favored the suburbs, where most new single-family homes were built (or overbuilt). Then, in the housing bust, patterns reversed, with urban counties and large metros rebounding while suburban and rural growth slowed. Now, as the recovery continues, old patterns – from before the 2000s – are returning.

For starters, compare population growth in metros by the severity of their local housing bust. In the hardest-hit metros, where prices climbed during the bubble and then fell 30% or more, population growth slowed dramatically from 2006 to 2009. Note that population in these metros started to slow before the bubble reached its height in 2006, as rising prices hurt affordability, and continued when the bubble burst as people lost their homes and local job markets suffered. In contrast, in metros with a relatively mild housing bust (price declines of 15% or less), population growth accelerated in 2007-2009: their economies held up better in the recession than the hardest-hit metros did. But since 2011, the metros with the severest housing bust have once again had the fastest population growth, and their lead over metros that had a moderate or mild bust has grown. Lower housing prices and stabilized local economies have attracted people back to metros that suffered the worst. In five of the ten large metros with the fastest population growth in 2015 (the four Florida metros plus Las Vegas, as shown above) home prices fell more than 45% in the housing bust.

housing bust severity

But it’s not just that population growth patterns today more like they did during early years of the bubble. Rather, local population growth trends increasingly look like they did before the bubble, in the 1980s and 1990s.

To see this, compare the list of the ten fastest-growing large metros in 2015 (shown at the top of this post) with the ten fastest-growing metros over the twenty-year period from 1980 to 2000. Five of the top ten in 2015 were also in the top ten in 1980-2000, including Cape Coral – Fort Myers, Austin, Orlando, Raleigh, and Las Vegas. Another four of the top ten in 2015 were in the top third in 1980-2000: North Port – Sarasota – Bradenton, Houston, Provo – Orem, and Lakeland – Winter Haven. Only Charleston SC was in the top ten in 2015 but had just middling growth from 1980 to 2000. On the flip side, seven of the ten steepest declining metros in 2015 were also among the bottom ten in 1980-2000; the other three in the bottom ten in 2015 were in the bottom twenty in 1980-2000.

Put another way: the correlation between population growth in 2015 and in the 1980-2000 period is quite high. Using all counties, rather than just large metros, the correlation between 2015 growth and 1980-2000 growth was 0.68. What’s surprising is not just that the correlation is high, but that it has increased in recent years: the correlation between 2009 growth and 1980-2000 growth was just 0.47, for example. In general, we’d expect these correlations with the 1980-2000 period to decline with each passing year – in most ways, things today are more like things were one year ago than five, ten, twenty, or fifty years ago. And, until 2009, the correlation between current-year growth and 1980-2000 fell as expected (except in 2006). But since 2009, the correlation has gone up, which is to say that local population growth patterns now look increasingly like the pre-bubble period of 1980-2000.

annual vs endcentury correlations

As local population patterns look more like the pre-bubble period, with accelerating growth in the suburbs and the Sunbelt, it becomes clearer that some of the population shifts during the housing bubble and bust were temporary and reflected the extreme housing cycle. In particular, the acceleration of population growth in the Northeast in 2009-2011 and moment when urban growth surpassed suburban growth in 2011 look like reactions to a housing bubble that brought unsustainable growth to the suburbs and the Sunbelt. That’s not to say that nothing has changed: there have been dramatic shifts since the pre-bubble years in the composition of local populations. College-educated young adults are much more likely to live in high-density urban neighborhoods than they used to, while seniors are increasingly likely to remain in suburban single-family homes. But, in aggregate, local population growth in 2015 looks ever more like it used to before the housing bubble, with the Sunbelt and the suburbs widening their leads.

Notes: all population data are from the Census Bureau. Metropolitan areas consist of one or more counties; the latest Census population estimates and all historical analyses in this post use consistent metropolitan area definitions from 2013, listed here. All population cutoffs are based on 2010 decennial population. There are 917 metropolitan and micropolitan areas altogether, of which 381 are metropolitan areas. There are 104 metropolitan areas that meet the 500,000 population threshold used throughout this post.

Where Real Home Prices Are At Record Highs

In one-third of large metros, nominal home prices hit a new record high in 2015. But, adjusted for inflation, real home prices are breaking records in just 7 of the 121 largest metros.

Home prices have been climbing nationally and in nearly every metro for several years. The rebound in prices after last decade’s housing bust and economic recession has been steep and widespread.

At first glance it looks like prices today in many metros are at all-time highs. In 43 of the 121 metropolitan areas (and metro divisions) with at least a half-million people, home prices hit a new local record high in 2015, according to the latest FHFA home price index.

But home prices, like almost all other prices, tend to rise over time thanks to general inflation. In fact, if real (that is, adjusted for inflation) home prices never changed, inflation alone would cause nominal (that is, unadjusted) home prices always to be at a new record high. Price indices like FHFA, Case-Shiller, CoreLogic, and others report nominal home prices.

Looking instead at real home prices, few metros are breaking records. (I adjusted FHFA’s nominal price index using CPI-U excluding shelter.) In only 7 of the 121 largest metros are real home prices now at record highs. These are:

  • Austin
  • Buffalo
  • Denver
  • Honolulu
  • Nashville
  • Pittsburgh
  • San Francisco

These record-setting metros all have booming housing markets today or had a very mild housing bust last decade, or — in the case of Austin — both. In a few other metros, real home prices are within 5% of their record highs but not quite there, including Durham-Chapel Hill, Houston, and San Jose.

In the vast majority of large metros — 97 out of 121 — the standing record for real home prices is held by last decade’s bubble. For nearly all of those 97 large metros, real home prices reached their peak in 2005, 2006, or 2007. Detroit and Indianapolis set their record a bit earlier, in 2003; Raleigh bloomed later, peaking in 2009.

Furthermore, in a handful of large metros, the record high came before last decade’s bubble. In 17 of the 121 largest metros, real home prices were at their highest in the late 1970s or 1980s (the FHFA index starts in the late 1970s for most large metros). These include much of Texas and Oklahoma, including Houston, Dallas, Fort Worth, San Antonio, and Oklahoma City, though notably not Austin, which is currently at a record high. Several mid-size metros in the Midwest and Northeast also peaked in the 1970s and 1980s, like Hartford, CT; Rochester, NY; Dayton, OH. No metro set a record high for real home prices between 1989 and 2002 that still stands today.

Many metros are still — despite the price rebound — far below their record high in real terms.
Real home prices are now 17% below peak in Atlanta, 29% in Chicago, 32% in Miami, and 44% in Las Vegas. There may be a lot about the housing market to worry about, but it’s hardly 2006 all over again.

(Download a spreadsheet with (1) the peak quarter for real home prices and (2) the difference between today and the peak, for all large metros, here.)

The rebound in home prices is therefore less dramatic than suggested by headlines about nominal home-price records. Certainly nominal home prices matter in some ways: neither mortgage payments nor capital gains are indexed to inflation, so nominal price changes can have real effects on homeowner behavior. But as an indicator of bubble risk, affordability, or other market dynamics, real home prices paint a more accurate — and less alarming — picture.

Should Your Tech Firm Have an Economist?

I get asked this often by companies thinking about hiring an economist – especially around this time of year, as firms are putting together annual hiring plans and as economists are doing their annual job-market dance. Tech companies that posted openings this year on the main job board for new Ph.D. economists include Airbnb, Amazon, AOL, eBay, Facebook, iCIMS, Pandora, Uber, and Wealthfront. Many other tech firms, most prominently Google and Microsoft, also have economists.

My answer comes in two parts. First: what do economists actually do in tech firms? And, second, do tech firms really need economists to do these things? My answers come from my own experience and that of my economist friends and peers. (Two good recent takes by others on why tech firms have economists are here and here.) Some of this applies to firms outside of the tech world, too, though that’s not my focus; and my answers surely don’t apply to all tech firms or to all types of economists.

Economists in tech firms tend to do four things:

  1. Advise on strategy. Economists often help firms plan and make investments by analyzing and forecasting economic trends, while thinking about how those trends will affect the business. They might help size a market; estimate prices for something the firm is selling; or estimate prices for something the firm is buying, like employees, office space, or even other firms. Economists might run the corporate analytics or business intelligence team and work closely with the CEO and the finance team.
  2. Build or improve the core product. Economists often work directly on the core product, especially in firms developing new advertising models or new transactions platforms. Economists are naturally drawn to the challenges that these firms wrestle with, and cutting-edge economics research can be valuable to these businesses. Which firms are these? Some evidence: at the just-completed annual economists’ conference, the companies most often named in academic paper titles were Uber and Amazon. Plus, economists from those firms as well as from eBay, Facebook, Google, Microsoft, and Pandora presented at the conference.
  3. Evaluate economic impact. Firms often want to understand or trumpet their economic impact. They might do this as part of an antitrust or competitiveness-practices case; to lobby for particular regulatory or legal actions; to build goodwill; or – believe or not – as part of a genuine effort to make a positive social contribution. Key internal partners for economists evaluating economic impact might be the legal, government relations, or PR teams.
  4. Build brand awareness, credibility, and thought leadership. Economists often do research, typically using a firm’s own data, to develop new insights about the sector or market that the firm operates in. Their output might be research reports, blog posts, tweets, or presentations. The audience for this research might include consumers, industry customers, investors, policymakers, or the media. The economist might be the company’s public-facing spokesperson on these topics, or work primarily behind the scenes. At sites that help consumers do important, complicated things — like finding a job (Glassdoor, Indeed), finding a home (Trulia, Zillow, Redfin), and finding a professional (Thumbtack) — chief economists create content that helps build the brand and boost site traffic.

It’s rare that one economist will do all of these things at a tech firm. A firm doesn’t necessarily need all of the things that economists could do, and even if it did, economists come in many flavors. Just as you wouldn’t want a dermatologist to deliver your baby, you probably wouldn’t put your in-house pricing-algorithm expert on CNBC to talk about market trends.

Most tech firms get by without economists. By the time a firm starts thinking about bringing on an economist, it might already have a team of data scientists and analysts with excellent quantitative intuition, strong coding skills, and model-building experience. The best data scientists are comfortable with a wider range of analytical and statistical methods, as well as general-purpose coding languages, than most economists are, though economists are increasingly learning these tools.

But economists come with a differentiated set of skills, training, and temperament that can be hugely valuable. (To be clear, I’m thinking about the flavor of economists I personally know best — applied microeconomists with Ph.D.’s from “saltwater” departments.) Economists’ comparative advantages include:

  1. Knowledge of economics frameworks. Many of the concepts that are second nature to economists are essential for making sense of markets, prices, and behaviors. Economists interpret almost everything that happens through the lenses of both supply and demand. We think carefully about how incentives affect behavior. We think not only about how actions affect prices, but also how prices, in turn, affect other actions. And we don’t believe there’s such thing as a free lunch (tech firms’ catered lunches notwithstanding). Economists consider these frameworks so essential that we’re easy to caricature: years ago, many of us joked that a particular famous economist with newborn twin daughters might name them Supply and Demand.
  2. Data detective and mash-up skills. Much of the data used at tech firms are those created or harvested internally, and often the challenge is how to manage overwhelming amounts of data. In contrast, economists often have to go out and find data for their research. Many economists develop strong detective skills for sniffing out new or overlooked datasets; know how to find and use Census and other established, benchmark data; and have a sense for how to combine multiple datasets in new and unexpected ways.
  3. Hypothesis-driven statistical modeling. Economists tend not to let data “speak for themselves” – they (and many statisticians and other quantitative social scientists) try to develop theoretical models or hypotheses before unleashing statistical tools, while making assumptions (sometimes reasonable, sometimes not) about the data. These methods are designed to infer cause-and-effect relationships between variables. In contrast, commonly used data science methods are more geared toward making predictions; these often involve black-box models that offer superior predictive accuracy but less transparency about the relationships between variables. This post nicely summarizes the differences between how economists and data scientists traditionally approach statistical modeling.
  4. Cleverness about experimentation. Economists sometimes conduct proper randomized, controlled experiments, but for many questions that’s not possible (like estimating the economic value of an additional year of education). Instead, economists come up with clever ways to construct “natural experiments” by seeking out random events, arbitrary rules, or beyond-human-control conditions that arguably resemble a randomized, controlled experiment. As a result, many economists are sophisticated about experimental design, keenly aware of statistical bias, and often able to squeeze lemonade insights out of lemon data.
  5. Culture of internalized data scrutiny. Economists typically make damn sure their numbers are right. Anyone who has written an economics dissertation has seen someone – perhaps themselves – get ripped to shreds at an academic seminar. Economists gossip about their peers’ data errors. Economics is a culture of anticipating objections; finding your own mistakes; and getting it right before you present your work. This methodical training can admittedly be a liability in a fast-paced start-up, but you want that kind of self-scrutiny and caution before your data are presented to the judge, the regulator, or the Wall Street Journal reporter.

Economists do not have a monopoly on these advantages. Many other quantitative social scientists, statisticians, and researchers share these qualities, as do many data scientists and analysts, some of whom have solid economics training themselves.

So should your tech firm have an economist? Forgive the classic two-handed economist response, but: it depends. On one hand, economists don’t come cheap; you might not urgently need what economists can do; and you might already have others on staff who are close enough substitutes for economists. On the other hand, economists can have a profound impact on strategy, product, impact-evaluation, and brand-building, using powerful tools and a unique approach. And if you’re still not sure whether the benefits outweigh the cost, that’s another thing that economists can probably help figure out.

Thanks to Selvin Akkus, Roy Bahat, Nikesh Parekh, and Eric Rice for thoughtful feedback on previous versions, and to several other investors, executives, and economists with whom I’ve had this conversation over the years. This post is based on my presentation at an April 2015 meeting of the San Francisco chapter of the National Association of Business Economics. Apologies to any companies I missed in my lists of those posting jobs and presenting at the annual meetings – let me know and I’ll make updates.

New Census Report Lowballs Household Formation

Today the Census’s Homeownership and Vacancy Survey (HVS) reported that household formation was just 462 thousand in 2015 Q4 compared with a year earlier. Household formation is a key indicator of housing demand, and that 2015 Q4 estimate rivals those at the worst of the recession.

But there’s no need to panic. I don’t see how that household formation number can be right. The problem might be that the HVS updated the housing-unit controls used in this report. The result is an implausible time series in the number of households, which jumps suddenly in 2014 Q4:

total households

Household formation equals the year-over-year change in total households. Therefore, household formation was reported to have jumped up for four quarters (2014 Q4 – 2015 Q3) and then to have plummeted in 2015 Q4:

household formation

Fortunately, there’s an alternative method for estimating household formation that sidesteps the HVS housing-unit controls. (See this previous post on household formation.) The number of households equals the number of adults times the headship rate, which is defined as the number of households per adult. Among adults 16+, the headship rate is roughly 50% — in other words, there are about two adults age 16+ in the average household. We can combine several facts from other Census sources:

  1. The U.S. civilian non-institutionalized population age 16+ was a hair under 250 million in December 2014.
  2. The U.S. civilian non-institutionalized population (all ages) was estimated to have grown 0.8% in 2015. Separately, the total resident population age 16+ had been projected to grow 1.0% in 2015. There’s no good estimate of the growth of the civilian non-institutional population age 16+ specifically, but somewhere in the 0.8%-1.0% range is a safe estimate.
  3. The headship rate for civilian non-institutionalized adults age 16+ fell ever so slightly from 50.06% in 2014 Q4 to 50.02% in 2015 Q4, according to my calculations from Current Population Survey basic monthly microdata files.

Multiplying the adult population in each time period by the headship rates yields an estimate for the number of households. In 2014 Q4, that equaled 124.9 million. In 2015 Q4, that equaled something in the range of 125.8 – 126.1 million, depending on whether you use the 0.8% or 1.0% estimate of adult civilian non-institutionalized population growth. The change over time is household formation, which this method suggests is in the 900-1150 thousand range. Even the low point of that range is nearly twice the rate of household formation of 462 thousand reported by the Census.

The stagnant headship rate is a disappointment and a surprise. The aging of the population alone should raise the headship rate since older adults live in smaller households than younger adults do. Furthermore, the economic recovery should encourage more young people to form households, further lifting the headship rate. So the slight decline in headship, and resulting household formation of 900-1150 thousand, is nothing to celebrate. But it’s far from the disaster that the HVS estimate of 462 thousand suggests.

The Urban Jobs Comeback, Continued: Follow the Money

Yesterday’s post showed that suburban counties of large metro areas have had the fastest job growth, both over this economic cycle as a whole (2000-2015) and even, by a slim margin, in the most recent year (2014-2015). Job growth is a natural measure of overall economic activity. But not all jobs are the same: some jobs pay better – and in that sense represent more economic activity – than others.

This post repeats some of yesterday’s analysis, but for wage growth rather than job growth. (See notes from yesterday’s post for details on data and methodology, including about geography. See additional notes at the end of this post.) It looks at growth in both (1) average wages per job and (2) total wages. Total wages equal average wages per job times the number of jobs.

Average wages per job are highest in urban counties of large metros, by a wide margin. Within large metros, urban-county wages-per-job are 20% higher than in higher-density suburbs and 37% higher than in lower-density suburbs. After all, many higher-paying industries, like finance, are clustered in big-city downtowns, and many companies put headquarters offices in big urban centers and back-office functions in lower-cost suburbs or smaller metros.wages per job 2015

But the key question is growth in average wages per job – and in total wages – across different places. Just as total wages equal wages per job times the number of jobs, the growth in total wages is a combination of wages-per-job growth and job growth.

Over the period 2000-2015, wages per job grew most in urban counties of large metros, small metros, and non-metropolitan areas. Non-metro and small-metro wages per job were helped by energy jobs, many of which are both high-paying and far from big cities:

wages per job 2000 2015

However, the differences across place types in wages-per-job growth are small. As a result, total wage growth over the same period was still fastest in lower-density suburbs of large metros, just as job growth was. Wages-per-job growth was faster in urban counties than in lower-density suburbs, but not by enough to make up for the faster job growth in lower-density suburbs. Whether we look at jobs or total wages, growth in the urban counties of large metros was slower than in higher-density suburbs, lower-density suburbs, mid-size metros, and small metros, over the period 2000-2015:

total wage 2000 2015

The story is different in the most recent year. Yesterday’s post showed that in 2014-2015 job growth in urban counties of large metros was still behind the suburbs, but only slightly. However, total wage growth between 2014 and 2015 was fastest in the urban counties of large metros – well ahead of all other types of places. (Note the scale, too: total wage growth for the U.S. as a whole last year was far ahead of its annualized average over the 2000-2015 period.)

total wage 2014 2015

To sum up: measuring economic activity by total wages rather than jobs, growth was still fastest in lower-density suburbs over the period 2000-2015. But in the most recent year, total wage growth has been fastest in urban counties. The possibly rosier future for urban growth that yesterday’s post discussed looks even better for total wages than it does for jobs.

 

Notes:

Wages typically include bonuses, options, and other forms of compensation.

All wage data were converted to constant 2015 dollars using the CPI-U for May of each year.

See notes to this post for all other data and methodological details.

City Limits: How Real is the Urban Jobs Comeback?

Last week, General Electric announced it will move its headquarters from Fairfield County CT to downtown Boston. GE is not alone: it joins other companies that have added jobs in big-city locations. Are these examples part of a larger trend of the urbanization of economic activity?

Key evidence in favor is that since 2007 job growth in large metropolitan areas has strongly outpaced job growth in the rest of the U.S. However, a more nuanced story emerges with a closer look at the data. In this post I look at urban versus suburban counties within large metropolitan areas, and at multiple time periods. Overall, while non-metropolitan areas have suffered, and the urban core of metropolitan areas have rebounded in recent years, the long-term trend remains that suburbs of large metros have the fastest job growth.

Suburbs are Where the Job Growth Is

Large metropolitan areas – those with at least a million people — are home to a majority (58%) of all U.S. jobs and are hardly uniform. Some metropolitan areas are far more urban than others: the density of the typical neighborhood in metro Chicago is 6 times that in metro Charlotte. Therefore, job growth in Chicago would be better evidence of an urban comeback than job growth in Charlotte. Also, large metropolitan areas include both dense urban neighborhoods as well as sprawling suburbs: the New York metropolitan area, for instance, includes Pike County PA and Putnam County NY, where the typical resident lives at a density of more than 2 acres per household. So, job growth in Manhattan would be better evidence of an urban comeback than in Pike County.

It turns out that job growth in the suburban portions of large metropolitan areas has been faster than in their urban portions. I divided all counties in large metropolitan areas into three categories based on their tract-weighted household density: urban counties, higher-density suburbs, and lower-density suburbs (see notes at the end of this post for details about data and methodology). Of the 58% of U.S. jobs in large metropolitan areas in 2015 Q2, approximately 46% are in urban counties; 38% in higher-density suburban counties; and 16% in lower-density suburban counties. Here’s the comparison of job growth from 2007 Q2 to 2015 Q2 in each type of large-metropolitan counties, alongside mid-size and small metropolitan areas and the non-metropolitan remainder of the country:

jobs 2007 2015

Job growth in all parts of large metros significantly outpaced job growth in smaller and non-metropolitan areas, as noted at the start of this blogpost. But within large metropolitan areas, job growth was fastest in lower-density suburban counties, and slightly faster in the higher-density suburban counties than in urban counties. Therefore, suburbs of large metros have led job growth in the economic recovery.

Take the Long View

The 21st century has begun with an extreme economic cycle. The housing bubble formed in the early years and peaked in 2006, after which home prices plummeted, employment declined, and the economy fell into the Great Recession, followed by the recent years of recovery. The housing bust, recession, and recovery were a reaction and correction to the bubble.

Using 2006 or 2007 as a starting point to compare urban and suburban growth shows us only what happened in the recession and recovery. It ignores the first half of the cycle, when the bubble fueled rapid homebuilding, population growth, and job growth in the suburbs, especially outlying, lower-density suburbs.

Comparing 2007-2015 with 2000-2007, urban counties of large metros win the most-improved award. They alone had faster job growth in 2007-2015 than in 2000-2007, when their job growth was flat. In that sense, the recession and recovery have brought a rebound in urban job growth.

jobs 2000 2007 2015

For evidence of a longer-term, structural – not cyclical — shift in the location of jobs, we need to look at the entire cycle to date, 2000-2015. It’s clear that job growth in this economic cycle as a whole has not been driven by the urban counties of large metros. The fastest job growth by far been in lower-density suburban counties of large metros. Growth in urban counties of large metros has lagged not only other parts of large metros but also mid-size and small metros.

jobs 2000 2015

Another way to look for long-term, structural shifts is to compare 2000-2015 with an earlier period. Comparable data are available starting in 1981. Using the same geographic definitions, this chart compares job growth in the 21st century so far with the last two decades of the 20th:

jobs 1981 2000 2015

Though job growth since 2000 has slowed in all types of places compared with previous decades, the patterns across place types are similar. Urban counties of large metros grew more slowly than suburbs of large metros, mid-size metros, and smaller metros, in both 1981-2000 and 2000-2015. This table shows the same data, along with the difference in growth between the periods:

Job Growth, Difference Between 2000-2015 and 1981-2000

Job Growth, 2000-2015

Job Growth, 1981-2000

Difference, 2000-2015 vs. 1981-2000

Large metros: urban counties

0.2%

1.2%

-1.0%

Large metros: higher-density suburbs

0.7%

2.6%

-2.0%

Large metros: lower-density suburbs

1.3%

3.2%

-1.9%

Mid-size metros

0.4%

2.0%

-1.6%

Small metros

0.5%

2.0%

-1.6%

Non-metropolitan areas

0.0%

1.6%

-1.6%

Note: numbers were rounded after differencing.

The slowdown in job growth between 1981-2000 and 2000-2015 was mildest in urban counties of large metros, and steepest in both higher- and lower-density suburbs of large-metros. In that sense, the suburbanization of jobs within large metro has slowed between 1981-2000 and 2000-2015. But that’s a narrowing, not a reversal, of the longer-term trend of suburbanization. Job growth in urban counties of large metros continued to lag all other place types except non-metropolitan areas since 2000.

Looking Forward

Two pieces of evidence point to a possibly rosier future for urban job growth. The first is that the most recent year of data available – 2014 Q2 to 2015 Q2 – shows that urban counties in large metros were only slightly behind suburban counties in large metros: the gap was narrower than any other year in decades.

jobs 2014 2015

Second, changes in the structure of the American economy favor big cities. The occupations that the Bureau of Labor Statistics expects to grow most over the next decade are disproportionately concentrated in big, dense cities. That’s no guarantee of faster growth, of course, especially if commercial and residential construction fails to keep up with growing demand from businesses that want to be in cities and their employees who want to live there. Still, these occupational changes reflect deeper shifts in economic activity that should favor big cities.

Aside from these nascent indicators, though, it’s hard to make the case that economic activity has fundamentally become more urban. Although urban counties of large metros have bounced back in the recovery, any trends that look at only part of the cycle since 2000 should be presumed cyclical until proven structural, rather than the other way around. Looking over the full cycle in order to see the longer-term trends, the most striking change is the slowdown in job growth in non-metropolitan areas, where growth was slower than in all other places. The fastest job growth in 2000-2015 has been in the lower-density suburbs of large metros – as it was in the two previous decades and, surprisingly, even in the post-2007 correction to the suburban boom of the bubble years. While large metropolitan areas have led the recent economic recovery, it’s a step too far to conclude that America’s economy is in the midst of a broad, fundamental shift away from suburbs toward urban areas.

 

Notes:

All data in this post are from the Bureau of Labor Statistics’ (BLS) Quarterly Census of Employment and Wages (QCEW). Data for 2000-2015 use the NAICS-classification totals for all covered employment, and data for 1981-2000 use the SIC-classification totals. 1981 is the first year for which comprehensive data are available for U.S. counties: the series begins in 1975, but data for some large counties are missing in the earliest years.

The classification into large, mid-size, small, and non-metropolitan areas follows the approach of Josh Lehner of the Oregon Office of Economic Analysis, published in City Observatory. Josh generously shared his data and methodology with me, which I was able to replicate exactly with the original QCEW data downloaded from the BLS site.

Metropolitan areas are defined as of 2013 and classified by their 2010 Census population into large (1,000,000+), mid-size (250,000-999,999), small (<250,000), and non-metropolitan. Metropolitan areas are defined by the Office of Management and Budget and consist of one or more counties.

I classified counties of large metropolitan areas based on tract-weighted density of households per square mile, as of the 2010 Census, into urban (2000+); higher-density suburban (1000-2000); and lower-density suburban (<1000). Tract-weighted density equals the average density of the Census tracts in a county, weighted by tract households; compared with a standard density measure, it is less skewed by large areas of uninhabited land. I used the same county-density classification in this post about population growth. Tract-weighted household density of 2000 is close to the level that survey respondents consider urban, as explained in this post.

As mentioned, the metropolitan and county-density classifications are based on 2010 population, 2010 household density, and 2013 metropolitan area definitions, for all time periods of employment and wage data shown. However, county density and metropolitan population are affected by earlier changes in employment, potentially introducing bias; as a check, I repeated the analysis for the full 2000-2015 period with an alternative county/metro classification based on year-2000 population, household density, and metropolitan definitions. Results were essentially unchanged, though 2000-2015 employment growth in year-2000-defined non-metropolitan areas was flat instead of slightly negative as shown in the post.

An alternative approach used in other studies (here and here) compares job growth downtown (within three miles of the central business district) with the remainder of the metropolitan area. This approach does not take inter-metropolitan variation in density into account. Despite these methodological differences from the approach in this blogpost, the results are similar. Those studies show that downtowns grew more slowly than the metropolitan remainder when both the bubble years and recession & recovery years are included; downtown growth outpaced metropolitan-remainder growth in the recession & recovery after lagging by a wider margin during the bubble.

Where the Fast-Growing Jobs Are

Urban occupations will grow faster over the next ten years than suburban and rural occupations. The metros with most favorable job mix for growth are Durham-Chapel Hill, New Haven, and San Jose.

The ups and downs of local job and housing markets are often tied to the fates of their key industries. Whether it’s tech in the Bay Area, oil in Houston, or manufacturing across much of the Midwest, the local job mix matters.

Yesterday the Bureau of Labor Statistics released its ten-year projections for occupational growth. This post shows which occupations are projected to grow or decline most; it also demonstrates the profound effect aging will have on labor demand and labor supply, and how the projections are good news for women and the well-educated. Healthcare-related occupations, along with personal care and computer & mathematical occupations, will grow fastest, while production (i.e. manufacturing) and farming/forestry/fishing occupations are projected to decline.

Which local markets are favored with fast-growing occupations or saddled with declining ones? I combined BLS national occupational projections for 2014-2024 with Census data on where occupations are located today, based on place of work (see note).

Urban Jobs, Rural Jobs

What do artists and economists have in common? Not much — aside from being the two most urban occupations. More than 60% of both actors and economists are in big, dense cities. But the most urban occupations also include in-person services where local demand is strongest in big cities, like taxi drivers, parking-lot attendants, and elevator installers and repairers.

most urban occupations

Occupations concentrated in urban areas are projected to grow faster than occupations in other areas. Based on where occupations are located today, urban jobs are projected to grow 6.8% over the next ten years, modestly ahead of suburbs, smaller cities, and rural areas. Rural jobs are disproportionately in manufacturing and natural-resource occupations, which are projected to decline. (This chart shows the weighted average of projected occupational growth in 2014-2014, weighted by today’s occupational mix in each type of area.)

densegroup projections

The differences in overall projected occupational growth across areas aren’t huge, and that’s because only some occupations tend to be clustered in certain areas. Many large occupations, like secretaries, cashiers, elementary school teachers, and retail salespeople, are almost everywhere, so the occupational mix varies less across different locations that you might think.

Still, growth in occupations that don’t just serve the local population — like software developers or petroleum engineers — tends to boost growth in other occupations that do service the local population. For instance, if tech is booming and the oil industry is shrinking, then growth in other occupations catering to tech and oil workers — such as retail and personal services — could be faster in tech-heavy San Jose than in oil-dependent Houston. (Economists call this “local multipliers.”) Therefore, the projected differences across areas in occupational growth, shown above, probably understate the important of job mix to local growth.

The Local Markets with the Fastest-Growing Occupations

The same method shows which local markets have the most and least favorable job mixes. Among all metropolitan areas with at least half a million people, Durham-Chapel Hill, NC, has the most favorable job mix for future growth, followed by New Haven and San Jose. Several of these top ten have large local tech industries, including Durham-Chapel Hill, San Jose, Boston, and Washington, DC. Many are among America’s most expensive markets. What accounts for McAllen-Edinburg-Mission, a large, low-income Texas border metro? There, many people work in the growing occupations of home health and personal care, and relatively few in declining routine office occupations.

metros most favorable

The inland California metros of Bakersfield, Fresno, and Stockton have the least favorable mix. These metros have young populations and therefore relatively few people in healthcare jobs; furthermore, they have substantial agricultural industries, and Bakersfield has a significant oil and gas industry as well. Largely absent from this list are Midwestern markets, despite their historical reliance on manufacturing.

metros least favorable

The differences between metros are very small. Large metros aren’t as different in their occupational mix as you might think. Plus, remember that this approach might understate the impact of job mix on local growth because of the assumption that any given occupation will grow at the same rate in all places; that is, this approach ignores any multiplier effect.

But job mix isn’t necessarily local destiny. People move not only for jobs but also for personal reasons and for housing affordability. Places where geography or regulations (or both) limit new housing construction could have slower job growth than predicted by their job mix: failing to accommodate booming local industries with more housing could lead to higher housing prices rather than faster job growth. Other state and local policies, too, could also boost or constrain growth relative to what the job mix alone would predict. And places that aging baby boomers eventually move to could see larger-than-expected growth in healthcare jobs and other services that retirees consume.

Still, a local area’s job mix is a tailwind or headwind for future growth. Some fast-growing occupations are clustered in particular places, as are some fast-declining industries. Whether the occupations and industries that lead U.S. economic growth are as predicted or are full of surprises, the local job mix will be a blessing for some markets and a curse for others.

Note: Employment projections are from the Bureau of Labor Statistics (BLS), and the location of workers is from the American Community Survey (ACS). The 818 BLS occupation codes were combined as necessary to match the 473 ACS occupation codes.

Location of workers was based on Place of Work Public Use Microdata Areas (PUMAs), which are aggregations of residential PUMAs and are typically equivalent to counties in more urban areas or groups of counties in more rural areas. Place of Work PUMA’s were classified by tract-weighted household density into big, dense cities (weighted density of 2000+ households per square mile); big-city suburbs and lower-density cities (1000-2000); lower-density suburbs and smaller cities (500-1000); and small towns and rural areas (0-500). Approximately 30% of jobs are in big, dense cities; 30% in big-city suburbs and lower-density cities; 20% in lower-density suburbs and smaller cities; and 20% in small towns and rural areas. Data on the location of workers are from the ACS Public Use Microdata Sample (PUMS), 2012-2014.

ACS data were downloaded from IPUMS, which requests to be cited as: Steven Ruggles, Katie Genadek, Ronald Goeken, Josiah Grover, and Matthew Sobek. Integrated Public Use Microdata Series: Version 6.0 [Machine-readable database]. Minneapolis: University of Minnesota, 2015.

The Jobs and Workers of Tomorrow

The aging of the U.S. population will drive demand for health-care jobs, and nearly one-quarter of the American workforce in ten years will be 55 or older.

The latest employment projections from the Bureau of Labor Statistics reveal the effect of the aging population on both labor demand and labor supply. These projections, updated every two years, were released this morning and provide a comprehensive view of how the labor market might change over the next decade. I’ll be looking at these and related data as part of a new research project with Bloomberg Beta.

The fastest growing and declining occupations

Looking first at occupational groups, the fastest growing are projected to be healthcare support occupations (like nursing assistants and home health aides) and healthcare practitioners and technical occupations (like doctors and nurses). Many high-tech jobs are classified under computer and mathematical occupations, which are projected to grow quickly though not as quickly as the health-related occupations. Production occupations — two-thirds of which are in manufacturing firms — are expected to decline, as are farming, fishery, and forestry occupations.

major occupational group chart

At a more detailed level, the fastest-growing occupations are mostly health-related, led by large increases in physical therapy, occupational therapy, and nursing. Tailwinds are strong for wind turbine service technicians — whose ranks are expected to double — and the numbers are adding up for statisticians, too.

fastest growing occupations

The fastest-declining occupations include many that produce or transport physical goods, including locomotive firers. Plus, modern communication technologies continue to displace mail-processing, telephone, and switchboard operators.

fastest declining occupations

Faster-growing occupations require more education

Alongside its employment projections, the BLS reports the level of education needed to enter each occupation. I combined occupations requiring the same level of education and compared the total projected growth across the groups. The fastest growing are projected to be those needing a master’s, doctorate, or professional degree. Many of the fast-growing healthcare jobs require these advanced degrees. But the slowest-growing occupations aren’t those needing the least education. Rather, occupations requiring a high school degree are projected to grow more slowly than those requiring no formal credential. Occupations requiring a high-school degree accounted for 36% of jobs in 2014 and are important for middle-class opportunity. The slow growth of jobs requiring high school degrees will cause the labor market to continue to polarize into jobs requiring more education and those requiring less.

job growth by education

The fastest-growing occupations have something else in common. Six of the ten fastest-growing occupations — all of the therapy and nursing occupations — are dominated by women. Just three — wind turbine service technicians, commercial divers, and ambulance drivers — are mostly male. (Statisticians are roughly 50/50.) This pattern holds across the full set of jobs: female-majority occupations are projected to grow 7.7% over the next ten years, versus 5.1% for male-majority occupations. (Of course, the sex ratio of occupations could change in the future as well.)

Breaking it down further, by sex and educational requirements, the slowest-growing occupations are projected to be those that require a high school degree and are traditionally male — including many production (i.e. manufacturing) occupations. Employment in those occupations is projected to grow just 3.0% over ten years — less than one-fourth the growth rate of female-majority occupations requiring a master’s, doctorate, or professional degree.

job growth by education and sex

The aging workforce

Just as the aging population will change the types of jobs we’ll need, the workforce itself will age. Two trends will work together to make this happen. First, of course, is that baby boomers will be aging into their 60s and 70s. The number of adults 55+ is projected to increase 21% between 2014 and 2024. In contrast, the number of 25-54 year-olds — the “prime working-age population” — is projected to increase just 4% over the same period, despite the large number of young millennials (age 20-24) moving into prime working age.

Second is that labor force participation is projected to rise most for age groups 55+, especially those 65-74. The BLS projects that the share of 55-64 year-olds who are in the labor force (that is, people who are either employed or unemployed but actively looking for work) will rise from 64.1% in 2014 to 66.3% in 2024. For 65-74 year-olds, the labor force participation rate is projected to rise from 26.2% to 29.9%. In other words, people will be retiring later. In contrast, the BLS expects the labor force participation rate for younger age groups to stay flat or decline.

change in LFPR by age

The combination of faster population growth and rising labor force participation among older adults and means that nearly one-quarter (24.8%) of the workforce in 2024 will be 55+. That’s up from 21.7% in 2014 and just 11.9% twenty years earlier, in 1994. Seniors (65+) will be 8.2% of the labor force in ten years — nearly three times their share of 2.9% twenty years ago.

share of older adults in labor force

These projections reveal the significant impact the aging population will have on labor demand and labor supply. Health-care jobs will grow fastest, and nearly one-quarter of the labor force will be older than prime working-age (a term that might itself need to be retired). More generally, job growth will favor women and those with advanced degrees, while men with high school degrees may face the biggest labor-market challenges.

Note: all data, with the exception of the sex ratio of occupations, are from the Bureau of Labor Statistics. Sex ratio by occupation was calculated from the American Community Survey (ACS) Public Use Microdata Sample (PUMS), 2012-2014. The BLS provides more detailed occupation codes than the ACS, so in some cases the sex ratio for a broader occupation reported in the ACS was assigned to multiple, more detailed occupations reported in the BLS projections. ACS data were downloaded from IPUMS, which requests to be cited as: Steven Ruggles, Katie Genadek, Ronald Goeken, Josiah Grover, and Matthew Sobek. Integrated Public Use Microdata Series: Version 6.0 [Machine-readable database]. Minneapolis: University of Minnesota, 2015.

Why Millennials Still Live With Their Parents

The entire increase in young adults living with their parents over the past twenty years can be explained by demographic shifts. That means the high share of millennials living with parents today might be the new normal.

This morning the Census reported that more young adults are living with their parents in 2015 than during the recession. Despite widespread expectations (including my own) that young people would move out as the job market recovered, they are not. The share of 18-34 year-olds living with parents was 31.5% in 2015, up from 31.4% in 2014. (These Census data are from March of each year. See note at end of post on data and methods.) Using different Census data, Pew recently reported that in 2014 the share of young adults living with parents or relatives was at its highest level since 1940 for men and even earlier for women.

The chart below shows the trend. (The charts in this post differ slightly from the newly published Census tables. See note for why.) After dropping a bit from the late 1990s to the early 2000s, the share of 18-34 year-olds living in their parents’ home rose steadily from 2005 to 2012 and has remained near this post-recession high even as the economy has recovered and unemployment for young adults has dropped sharply.

in parents home 1834 112215

Because the share of young adults living with their parents rose suddenly after the housing bubble of the mid-2000s burst, it’s natural to explain this trend in terms of recent housing and labor market dynamics. After all, young adults with jobs are much less likely to live with their parents than young adults without jobs are. Plus, rising rents and student debt burdens might be holding young people back from moving out of their parents’ homes.

However, alongside recent swings in the housing and job markets, there have been profound long-term demographic shifts that are related to young adults’ living arrangements. For instance, an unusually high share of 18-34 year-olds are at the young end of that range, and younger young adults (18-24) are much more likely to live with parents than older young adults (25-34). An especially important trend is that people are waiting longer today than in the past to get married and have kids — so the share of 18-34 year-olds who are married with kids has plummeted from 49% in 1970 to 36% in 1980, 32% in 1990, 27% in 2000, 22% in 2010, and just 20% in 2015. Unsurprisingly, married young adults and those with children are far less likely to live with their parents than single or childless young adults.

married with kids 1834 112215

How much have these longer-term demographic shifts contributed to the increase in young adults living with their parents? Using regression analysis, I estimated how much these demographic shifts contributed to changes in young adults living with parents in order to extract the demographics-adjusted trend. I included a standard set of demographic variables: five-year age subgroup, marital status, presence of children, sex, race, ethnicity, nativity (i.e. native- or foreign-born), current school enrollment, and educational attainment.

Adjusted for demographic shifts, the share of young adults living in their parents’ home was actually lower in 2015 than in the pre-bubble years of the late 1990s. In other words, young people today are less likely to live with their parents than young people with the same demographics twenty years ago were. To be sure, even the demographics-adjusted share of young adults living with their parents has climbed back up since the housing bubble burst around 2006, but it remains below pre-bubble levels from the 1990s.

in parents home 1834 demographics adjusted 112215

In this kind of analysis it’s important not to explain too much away using demographics, especially if demographic trends might themselves be an effect of living with parents or of the recession. For instance, it’s possible that living with parents might delay marriage: sleeping in your childhood bedroom probably doesn’t help your social life. In fact, though, the decline in young adults being married with kids long pre-dates the recession and the rise in living with parents, and has been relatively steady for decades (see chart above). That means marrying later is not the effect of the post-2005 increase in living with parents.

The other demographic variables we might not want to adjust for are school enrollment (i.e. are you currently in college) or educational attainment (i.e. what’s the highest degree you’ve earned). School enrollment and educational attainment are affected by the economic cycle and affect whether you live with your parents. Plus, the Census counts full-time students in dorms as living with their parents in the data used in this analysis. As a check, I dropped the school enrollment and educational attainment variables and limited the sample to 25-34 year-olds only, few of whom are still in school. The story remains the same: while the demographics-adjusted share of young adults living with parents has increased since the mid-2000s, it remains slightly below the pre-bubble level of the 1990s.

in parents home 2534 demographics adjusted 112215

So that’s the punchline: the increase in young adults living with parents over the past twenty years can be explained entirely by demographic changes. The increase since 2005 is not an aberration; once demographics are taken into account, the aberration is the bubble years of the mid-2000s, when an unusually low share of young adults was living with parents.

Adjusting for demographics doesn’t make the recent increase in young adults living with parents — or the implications for today’s housing market — any less “real.” The increased share of young adults living with parents means that household formation is being driven not by millennials but by baby boomers, and helps explain the low share of first-time home-buyers.

But adjusting for demographics does change what we should expect from the future. Because the demographics-adjusted share of young adults living with parents today is similar to pre-bubble levels, long-term demographic shifts may simply have pushed up the share of young adults living with parents to a new normal. Unless demographic trends reverse, the share of young adults living with parents is unlikely to fall much. Today’s millennials will leave their parents’ homes as they age — they’re not going to live there forever. But it won’t be the sudden unleashing of pent-up demand we might have expected if the increase of living with parents were only about the housing bust and recession and not about longer-term demographic shifts.

Notes:

  • All original data and charts in this post are based on my analysis of the Current Population Survey’s (CPS) Annual Social and Economic Supplement (ASEC). For 2014, when the ASEC used a split sample, I combined the 5/8 and 3/8 samples and weighted them appropriately. The published Census tables on families and living arrangements, available here, are based on the same underlying data but with minor differences in how living at home or with parents is calculated.
  • The pre-bubble years (1994-1999) are the baseline, so the actual and demographics-adjusted shares of young adults living with parents, averaged over those years, are equal by construction. The analysis begins in 1994 because that’s the first year when the complete set of demographic variables is available.
  • I downloaded the CPS-ASEC data from IPUMS, which requests to be cited as: Sarah Flood, Miriam King, Steven Ruggles, and J. Robert Warren. Integrated Public Use Microdata Series, Current Population Survey: Version 4.0. [Machine-readable database]. Minneapolis: University of Minnesota, 2015. 
  • The Pew report cited in the blogpost used data from the American Community Survey and decennial Censuses, which yield different estimates than the CPS-ASEC but the same broad trends.