It’s become something of a mantra among urban economists: Increasingly unaffordable housing prices in cities like New York, London, and San Francisco are very often the consequence of onerous and out-of-date land use regulations. Whether it’s restrictions on the height of buildings or the density of development, these regulations effectively constrain the supply of housing. This year’s Economic Report of the President flagged such land use regulations as a major factor in skyrocketing housing prices and growing urban inequality. Another study I wrote about last year estimates that restrictive urban land use policies cost the U.S. economy around $1.6 trillion a year.
But something much more enduring than zoning and land use is also contributing to the deepening housing affordability problems of leading superstar cities and knowledge hubs. According to a recent study by Issi Romem, chief economist at BuildZoom, part of the explanation lies in the geographic characteristics of cities and metros—mountains, lakes, coastlines, etc.—that make it all but impossible to expand and add more housing.
Expensive vs. expansive cities
The graph below tells the basic story. It arrays metros across two dimensions: the increase in housing prices over the past three decades (1980-2010) and the expansion of residential development. What it shows is an inverse relationship between outward expansion and housing price growth. The more a metro is able to expand outward and create new housing supply, the less its housing prices tend to rise.
Metros in the upper left hand quadrant like San Francisco, New York, L.A., Seattle, and Washington, D.C. have high housing prices and have seen relatively little residential expansion. Romem dubs these “expensive” metros. Meanwhile, metros in the lower left hand quadrant like Las Vegas and Atlanta—as well as the tech hubs of Austin and Raleigh in the North Carolina Research Triangle—have seen more modest housing price increases alongside much more residential development. Romem dubs these “expansive” metros.
Taken as a whole, U.S. cities and metros have generally continued to expand over the past half century or so. According to Romem’s analysis, U.S. metros have added about 10,000 square miles—about the same area as the state of Massachusetts—per decade from the 1950s to the 2000s. (Romem included only developed areas, with 200 or more existing housing units per square mile, in his analysis.)
But the geography of expansion has been astoundingly uneven, with some metros expanding earlier and others expanding later, as the chart below shows. Today’s “expensive” cities expanded more during the 1950s, ‘60s and ‘70s, while “expansive” cities grew more in the 1990s and 2000s.
Take two examples of expensive versus expansive: San Francisco versus Atlanta. As the chart below shows, the San Francisco metro area expanded far more than the Atlanta metro area during the 1950s. But by the 2000s, San Francisco’s expansion was all but nonexistent. In contrast, Atlanta’s expansion climbed steadily from the 1940s to the 2000s, even through the latest housing crisis.
Naturally, there are exceptions to the rule. For one, the economic plight of Rustbelt metros such as Detroit, St. Louis, Pittsburgh, or Cleveland has made their housing prices slow to rise, even though they have been unable to expand geographically. Romem dubs these “legacy” cities. Some expansive Sunbelt metros such as Denver or Salt Lake City have also seen substantial increases in their housing prices over the past few decades, though not to the extent of the most expensive cities. Interestingly enough, New York remains one of the most expensive cities, despite being the fifth leading “expander” of all 40 metros covered.
Geography is a major factor in expensive cities. Cities such as L.A., Seattle, and San Francisco face hard physical boundaries like coastlines and mountains, which in addition to policy decisions that may limit height or density, can also serve to limit capacity for expansion.
Geographically constrained cities also tend to opt for more strict land use regulations, as residents seek to preserve things like views or what they perceive as a natural limit on growth, according to an influential 2010 study by the MIT economist Albert Saiz. As Saiz puts it, “geography is a key factor in the contemporaneous urban development of the United States, and help[s] us understand why robust national demographic growth and increased urbanization has translated mostly into higher housing prices in San Diego, New York, Boston, and Los Angeles, but into rapidly growing populations in Atlanta, Phoenix, Houston, and Charlotte.” In this sense, geography is one of the biggest factors in skyrocketing housing prices in expensive cities.
Geography proves more and more important as cities and metros grow ever larger, Romem adds, because the best land is typically taken up first. In other words, the same dynamic that currently plagues expensive cities may someday catch up with their expansive counterparts.
In addition to physical barriers to development, there are also jurisdictional obstacles. Older cities like New York, Boston, and San Francisco are ringed by suburbs that make it impossible to expand outward, while younger Sunbelt cities have been able to annex and integrate land at their periphery.
How cities can adjust
There are two ways the American economy and its cities may ultimately adjust to this new geography of expensive vs. expansive cities, according to Romem. On the one hand, the high prices in expensive cities are likely to cause more people and industries to move to expansive cities and metros. Indeed, this is already happening in some places, as expansive suburbs in the Sunbelt grow faster than expensive urban centers like New York and San Francisco. But, as noted above, these expansive places may eventually come face-to-face with their own geographic limits to growth and restrictions on development as they experience more economic activity.
On the other hand, expensive cities could eventually start to liberalize their zoning and land use restrictions in ways that make them denser and more affordable. As my former CityLab colleague Emily Badger points out, “there is no such thing as a full place.” Even cities and metro areas that cannot expand any farther outward can still increase their density of development and build taller structures. Still, their housing prices are likely to remain relatively high. As long as expensive superstar cities and tech hubs continue to be desired by the most affluent and the most advanced industries, they will continue to face intense competition for space in and around their downtown cores, which puts unrelenting pressure on real estate prices.