Also relevant is that 12 of these 15 metro areas touch either an ocean or the Gulf of Mexico (and the three that don’t — Riverside-San Bernardino, Napa and Vallejo-Fairfield — aren’t far from the Pacific). Coasts both attract people who can live anywhere and constrain an area’s ability to expand, neither of which is good for housing affordability.
Constraints on expansion of the housing supply are what has driven the real estate crisis in coastal cities. Some are physical: Most of the metropolitan areas in the above charts come in near the top of Massachusetts Institute of Technology economist Albert Saiz’s ranking of metros with the smallest amount of developable land (land that is pretty flat and not underwater) within 50 miles of the core city. California’s Oxnard-Thousand Oaks-Ventura, squeezed between the Pacific Ocean and the Santa Susana and Topatopa Mountains, is both the least developable and least affordable metro area in the country.
Since the 1960s, the East and West Coasts (California in particular) have also experienced what economists Chang-Tai Hsieh of the University of Chicago and Enrico Moretti of the University California at Berkeley term a “property rights revolution” of anti-development activism and regulation that’s made it even harder to build enough new housing to satisfy demand. This revolution coincided, and to some extent collided, with a big increase in agglomeration. That's the economic value created by gathering lots of skilled knowledge workers in the same place — a phenomenon described by Moretti in his influential 2012 book “The New Geography of Jobs.”
In a paper that was published in the American Economic Review in 2019 but circulated widely for several years before then, Hsieh and Moretti estimated that if land-use restrictions in just the New York, San Jose and San Francisco areas had been the same as that of the median U.S. city from 1964 to 2009, thus enabling them to house more people than they actually did, U.S. gross domestic product would be 3.7% higher. A few weeks ago, George Mason University economist Bryan Caplan discovered that Hsieh and Moretti made a simple mathematical error and that, according to their model, GDP would actually be 14% higher. That’s a lot of lost output!
The sudden move to remote work occasioned by the pandemic seemed to offer a workaround. If knowledge workers could be as productive gathered virtually as crammed into New York high rises or Silicon Valley office complexes, then maybe those development constraints in high-priced cities wouldn’t matter so much. Moretti, in an interview last month with Vox, said he doubted “the economic geography of the U.S. will be profoundly different in the long run” because remote working probably won’t permit companies to access “those particular advantages that come from agglomeration.” But he also acknowledged that the share of remote work will likely remain higher than it was before Covid-19.
I’m not going to try to predict here how many of us will be working from home 10 years from now, but the shift over the past year has already been enough to send rents and home prices spiraling downward in some in-demand cities. If lots of new housing is built in these places, the price pressure could ease. Boise had an earlier stint as a fairly expensive place during the 1990s tech boom (chipmaker Micron Technology is based there), and price increases subsequently slowed to less than the national pace as lots more housing was built.
Still, the characteristics that attract remote workers — most of the metropolitan and micropolitan areas with the highest percentages of remote workers before the pandemic are in the mountains or on the beach — also serve to constrain new development. The remote workers themselves may add to those constraints by bringing new attitudes to town. Boise now has its share of Nimbys, too. In a few smaller, even more scenic metro and micro areas such as Boulder, Colorado; Bozeman, Montana; and Bend, Oregon, the combination of mountains, rising opposition to development and high demand have driven home-purchase prices up to or past the levels of metro Boston and New York, although they’ve still got a ways to go to catch San Jose and San Francisco.
The larger non-Californian-or-Northeastern metro areas on the least-affordable list are generally there for different reasons. In Las Vegas and Phoenix, overbuilding during the housing bubble led to a wipeout from which local home builders are still struggling to recover, with employment in construction still down 40% from its 2006 peak in Las Vegas and 25% in Phoenix compared with less than 2% nationally. It’s down 18% in Florida, which also suffers from chronic housing-affordability issues because wages are so low.
Wages are higher in the big metro areas of Texas, whose wide-open spaces and loose regulatory policies have been the great counterexamples to the constrained cities of the East and West coasts. But even these may finally be reaching the point where their affluence and sprawling size become limitations. This is most obvious in Austin, but neither San Antonio nor Dallas-Fort Worth score particularly well on affordability either, with each requiring almost 62 hours of median-wage work to pay the average monthly rent on a two-bedroom apartment. Again, all these places still look cheap to a coastal Californian. But without coastal-California wages (and in particular San Jose-San Francisco wages), they’re really not so affordable anymore.
Where is housing still affordable even for the locals? I’ll start this time with the most affordable of all metro areas, a list full of places that it’s hard to envision remote workers flocking to anytime soon.