New Geography

Author: Lisa Sturtevant

  • More Clouds Over Sky-High Metro Housing

    Can housing costs get so high that they repel new migrants, and stunt a metropolitan area’s economic growth?

    For potential migrants looking for a job, metropolitan areas—and in particular large metropolitan areas—are the places to be. People move in because that’s where the jobs are. More than 93% of non-farm jobs in the U.S. are in the 100 largest metropolitan areas. The biggest 15 metro areas account for 34% of the nation’s jobs. Big places also have the benefits of cultural amenities, educational institutions, and impressive retail and restaurant environments.

    The combination of population growth, a constrained supply of land, and local land use policies, however, work together to put upward pressure on housing costs. Wages in big cities are typically higher than in other places. In theory, this compensates for higher housing costs. But is this notion a thing of the past?

    One measure of housing affordability in metropolitan areas is the Housing Opportunity Index (HOI), produced quarterly by the National Association of Home Builders, and defined as the share of homes sold in a metropolitan area affordable to a household earning the local area median income, using standard mortgage underwriting criteria. The nation’s largest metropolitan areas—the so-called Mega metropolitan areas—are the least affordable places to live as measured by the HOI. They also experienced the biggest drop in affordability over the 2000-2007 period: The average HOI for the Mega metropolitan areas dipped from 54.3 to 34.5 between 2000 and 2007 (Table 1). In other words, 54.3% of homes in the nation’s largest metropolitan areas were affordable to the median income household in 2000, but only 34.5% were in 2007.

    Metropolitan Area Hierarchy
    Population
    Mega metro greater than 2.5 million
    Major metro 1 to 2.5 million
    AAA metro 500,000 to 999,999
    AA metro 250,000 to 499,999
    A metro less than 250,000
    Source: D. A. Plane, C. J. Henrie, and M. J. Perry.  2005. Migration up and down the urban hierarchy and across the life course. PNAS 102(43).

    Given the run-up in home prices, household migration to the Mega metropolitan areas should have slowed between 2000 and 2007, while migration to smaller, more affordable places should have been relatively higher.

    Table 1. Housing Opportunity Index by  Metropolitan Area Type: 2000 and 2007
    Percent of Homes Affordable to Median Income Household
    Metropolitan Area Type 2000 2007 Pct. Change
    Mega metropolitan 54.3 34.5 -19.8
    Major metropolitan 68.3 57.5 -10.9
    AAA metropolitan 70.7 53.2 -17.5
    AA metropolitan 66.4 56.2 -10.3
    A metropolitan 70.3 62.5 -7.8
    All metropolitan areas 68.7 58.3 -10.4
    Source: National Association of Home Builders and author’s calculations

    In-migration rates (Table 2) fell between 2000 and 2007 for all sizes of metropolitan areas, with somewhat larger drops in the Mega metros.

    Table 2. Household In-Migration Rates (per 10,000 housing units) by Metropolitan Area Type: 2000 and 2007
    Metropolitan Area 2000 2007 Pct. Change
    Mega metropolitan 401 367 -8.8
    Major metropolitan 449 423 -5.8
    AAA metropolitan 467 440 -5.8
    AA metropolitan 544 501 -7.9
    A metropolitan 560 526 -6.1
    All metropolitan areas 527 492 -6.6
    Source: IRS County-to-county migration files and author’s calculations

    Is there a statistical relationship between a metropolitan area’s housing affordability and household in-migration rates in a given year? The short answer is no. A simple correlation between metropolitan area HOI and in-migration rate for both 2000 and 2007 show a weak negative relationship between housing affordability and household in-migration rates.

    What does seem to matter is the change in housing affordability over time. There is a positive and large correlation between the change in HOI between 2000 and 2007 and in-migration rates in 2007. That is, places where affordability dropped between 2000 and 2007 generally had lower rates of household in-migration in 2007. Places that became more affordable between 2000 and 2007 had higher in-migration rates in 2007.

    Why would the change in housing affordability be related to in-migration when a metro area’s current affordability is not? It may be that changes in housing affordability are actually signals for a multitude of economic changes. Generally, we think of home prices as being influenced by changes in the local economy. When a region loses jobs, as some metro areas in the mid-West have, home prices fall and the region becomes more affordable.

    But it is also possible that the relationship can work differently. Can home prices influence job growth, particularly when prices rise quickly and wages cannot keep up? Perhaps the phenomenon is a result of employers not being able to expand their businesses and add jobs because they could not find workers that could afford to re-locate. Perhaps new firms did not locate in places with rapidly escalating housing costs because the wage premium got too high. Perhaps the disadvantages of large metropolitan areas started to outweigh the advantages, at least on the margins.

    Job data from the Bureau of Labor Statistics provides some preliminary indications on the economic effects of housing unaffordability in large metro areas. In metropolitan Miami, for example, the HOI dropped 49 points, from 58.8 in 2000 to 10.0 in 2007. The job growth rate in Miami slowed considerably in 2007. The number of jobs in the Miami metro area grew by about 3% annually in 2004, 2005 and 2006; however, in 2007, jobs grew by only 0.6%. In the Washington DC metro area, where the HOI dropped 39 points between 2000 and 2007, job growth proceeded at a stable 2.0 to 2.5% annual growth rate in 2004, 2005 and 2006. In 2007, however, job growth was at 0.8%.

    In many places where housing affordability improved—or at least did not decline too much—job growth held steady or increased in 2007. Many of these were Major metro areas (with populations between one and 2.5 million). For example, the HOI for Denver increased six points, from 58.5 in 2000 to 64.5 in 2007. The job growth rate in Denver in 2007 was 2.2%, up from 2.1% in 2006, 2.0% in 2005 and 0.8% in 2004. In Pittsburgh, the HOI was up 8.6 points and annual job growth rates were steady between 2004 and 2007. In Rochester, NY, the HOI was up five points and job growth in 2007 was stronger than in 2006.

    So—big drops in housing affordability may be a problem for a metropolitan area’s economic health. High housing costs make it more difficult to attract labor. The wage premium needed to offset the high housing costs becomes untenable at some point. As a result, new firms stop locating in high cost areas and existing firms are unable to add jobs.

    Eventually, of course, job losses would put downward pressure on housing costs and the metropolitan area’s economy would stabilize. Large metropolitan areas will end up with a smaller share of the nation’s jobs after this stabilization process. If large places are good for economic activity because of the agglomeration benefits they provide, the dispersion of economic activity may not be a good thing for them.

    How can the biggest cities best prosper? Large, high cost metropolitan areas can stem economic slowdowns through policy changes that increase the supply of housing, thereby reducing housing costs. These policy initiatives would include changes to local land use and building regulations to encourage the construction of more housing. A regional approach to increasing the housing supply would focus on promoting housing near transit and employment centers.

    In the last two years, home prices have declined across many high cost metropolitan areas; however, job growth has already slowed disproportionately in many of these regions. This trend will likely continue in the near future. If the nation’s largest places are to be the dominate location for job growth in the future, regional and local policymakers need to develop comprehensive housing strategies that treat housing as an integral part of a regional economic development policy.

    Photo by limonada (Emilie Eagan), Late Afternoon and Almost Stormy

    Lisa Sturtevant is an Assistant Research Professor at George Mason University School of Public Policy, Center for Regional Analysis.

  • A Return to the City or a New Divide in the Nation’s Capital Region?

    Census data continue to suggest that fringe areas still grow faster than cities, but some have continued to argue that the flight to the suburbs has ended, or at least slowed, and that we are experiencing a resurgence of urban living. In a 2005 article for the Journal of the American Planning Association, Robert Fishman predicts a new pattern of migration – a so-called Fifth Migration – that will revitalize inner core neighborhoods that were depopulated through decades of suburbanization. In a 2004 study of the New York region, James W. Hughes and Joseph J. Seneca contemplate the beginning of a “third transformation,” or a post-suburban regional geography, characterized by the end of population dispersion and the beginning of recentralization.

    Anecdotes, rather than hard data, have tended to drive the back to the city case. Data brought to bear on the issue usually shows the suburbs are still going strong. Yet it also appears that all suburbs are not created equal and population data may be missing subtle population shifts within a metropolitan area. The flow of households within a metropolitan area can show early signs of a change in a region.

    This analysis considers the extent to which there is the beginning of a back to the city movement in the Washington DC metropolitan area using county-to-county migration data from the Internal Revenue Service. We must start with the assumption that the Washington area is unique among American metropolitan areas. The presence of the national government largely shapes the structure and geography of the regional economy. A large share of the region’s jobs is concentrated in the core, due to the role of the Federal government in the region. However, in addition to being the seat of the Federal government, the Washington DC metropolitan area also serves a varied set of private sector employers, and is home to a diverse population with growing suburban and city neighborhoods.

    The metro area is defined by 22 counties and cities in the states of Maryland, Virginia and West Virginia and has at its center city the District of Columbia. For this analysis, the Washington DC metropolitan area is divided into five sub regions: Center City, Inner Core, Inner Suburbs, Outer Suburbs and Far Flung Suburbs (see Map 1). According to Census Bureau population estimates, between 1987 and 2007 the population of the Washington DC metropolitan area grew from 3.92 million to 5.31 million people, an increase of about 35.5 percent. The population growth rates over this period varied considerably within the metropolitan area. The Center City experienced a 7.6 percent population decline between 1987 and 2007 while all of the other subregions in the metropolitan area grew. The fastest growing subregion was the Outer Suburbs, where the population grew by 109.6 percent over the 20-year period, followed by the Far Flung Suburbs (80.0%), Inner Suburbs (27.4%) and Inner Core (23.7%).

    Figure 1 shows that the subregions furthest from the region’s core, the Outer Suburbs and Far Flung Suburbs, consistently have the highest rates of net migration, which indicate that they have been net gainers of households from other parts of the metropolitan area over the past 20 years. For the Inner Suburbs, net migration is positive (but small) until 1998 when it becomes negative. Both the Inner Core and Center City have negative net migration over the entire period, reflecting losses of households to the rest (i.e. the suburban portions) of the metropolitan area.

    Looking at the entire 20-year period suggests that the suburbs of the Washington DC metropolitan area have gained population at the expense of the closer-in jurisdictions. However, in the last few years, since 2005, the net migration rates for the Outer Suburbs and Far Flung Suburbs have declined while the rates for the Inner Core and Center City have become less negative. These three years of data suggest that the more distant suburbs have started gaining households more slowly while the closer-in jurisdictions have lost households more slowly with net migration rates moving towards zero. While three years do not necessarily constitute a trend, this analysis suggests the possible beginning of a modest back to the city movement in the Washington DC area.

    However, household gains (or a slowdown in losses) in the inner jurisdictions may be coming more at the expense of the Inner Suburbs as opposed to the more distant suburban jurisdictions. The Inner Suburbs subregion is continuing the downward trend in net migration rates that began in the late 1990s, losing households to both the outer suburban and core jurisdictions. If this trend continues, the Washington DC metropolitan area may experience a relative population decline of its Inner Suburbs, while the more far flung suburbs continue to grow (albeit more slowly) and the population of the inner jurisdictions stabilizes or even grows slowly.

    Despite the beginning of a small back to the core movement, the suburbs of the national capital region will continue to gather most future growth, and that suburban growth will be even further out. Over the last three decades, jobs have followed people into suburban communities; a place like Tyson’s Corner in Fairfax County now has almost as many jobs as Washington’s downtown business district. Workers can live in the Outer Suburbs and Far Flung Suburbs, benefiting from the relatively lower housing costs and commute with relative ease to jobs in these suburban employment centers. Some share of the population will choose to move back to the city, but there will always be demand for suburban life.

    The Inner Suburbs are caught in the middle of population moving in and moving out. The Inner Suburbs have become more urban and congested, as well as more racially and ethnically diverse. These changes may cause some households – including both native born persons and upwardly mobile immigrants — to look even further out for a traditional suburban lifestyle. A younger metropolitan area, the result of the large millennial or “echo boom” generation, may lead to more people moving out of the congested Inner Suburbs to a “real” urban neighborhood in the core, which is also crowded, but has public transit and walkable communities. This trend, however, may well be short-lived if, when this generation hits their mid-30s by 2015, it acts like previous generations, starting to raise families and move again to suburbs, most likely in the further periphery.

    All this suggests, for the short run at least, a possibility that this new pattern of household redistribution could create a new divide in the national capital region. Different from the well-documented east-west divide, the emerging divide will be between the “urbanites” and the “far flungs.” The divide will be demographic, economic and political and will characterize the future challenges to forming transportation, housing and other regional policies.

    Lisa Sturtevant is an Assistant Research Professor at George Mason University School of Public Policy, Center for Regional Analysis.