The Urban Institute released a great study about how the District has changed over the last thirteen years. It has a lot of information in convenient graphs that summarize difficult-to-understand Census data, and its publication has spurred lots of commentary from the D.C. land-use commentariat.
Emily Badger focused on the quick growth of expensive apartments since 2005. Aaron Wiener noted that, contrary to the dominant narrative, inexpensive family-friendly rental units have not disappeared. Rather, cheap one bedroom and studio apartments have largely disappeared. Lark Turner explained that Millennials played a large role in the District’s demographic shift and subsequent housing changes.
All three authors are correct, but they’re mostly focusing on demand-side explanations for why DC is so much more expensive than it was 13 years ago.
It’s easy to see why some neighborhoods were much cheaper in 2002: a lot of today’s hip neighborhoods weren’t great places to live back then. Places like Columbia Heights, U Street, and NE were filled with disamenties that made rent cheap. For example, before Target and the other DCUSA tenants came to Columbia Heights, 14th and Irving looked like this:

Replacing vacant lots and debris with stores, restaurants, and other things people like should lead to higher rents.
With the elimination of disamenities and the addition of amenities comes increased demand in the form of new residents. More people bidding on the same number of units pushes prices higher. Then, so it seems, rich people move into poor people’s homes and we have the city we currently live in. Right? Not exactly.
These changes are easy to spot, and account for some of the price changes. But demand only accounts for half the story, and it doesn’t explain how our current situation could be different. Under a more relaxed regulatory regime we might not have experienced the same spike in rental prices, home values, and displacement the District has undergone during the last fifteen years.
As Matt Yglesias wrote in response to the study, the supply of rental units hasn’t expanded very much because it’s often illegal to increase density where it is demanded most. This factor–legal restraints on building more places for people to live–is primarily to blame for D.C’s sky-high rents and real estate prices.
Even in neighborhoods with skyrocketing demand and new projects, our local regulations often make it illegal to respond by maximizing the number of units that can fit on a given plot of land. As I wrote earlier this year,
Columbia Heights is a case study of what happens to a newly-popular neighborhood under a restrictive land-use regime that doesn’t fully allow the processes described above. In 2000, there were 27,129 people living in the neighborhood… After a decade of construction (mostly renovation), new residents, and change, the total population only increased by 1,087 people.
According to the Urban Institute study, we’ve added 33,918 expensive rental units since 2005 while the total number units of rental housing only increased by 12,500. During the same time period our population increased by more than 50,000 people.
When land becomes expensive, developers usually respond by building apartments instead of single family homes. This, of course, makes them more money, but it also makes housing more affordable than it otherwise would be. It also prevents your cheap basement unit from being converted into a wealthy person’s wine cellar.
Major cities in general have seen a surge in demand over the last few years as more people want to live in urban areas, but the price response in the District is uncommon. Not only have local housing prices spiked more than most other cities, we’ve even outpaced the NIMBY Mecca–San Francisco.
There’s not much for policymakers to do about a surge in demand aside from making D.C. a worse place to live, but there’s plenty of room to improve on the supply side. For starters, the District Council could limit NIMBYs’ ability to block projects, eliminate parking minimums, and loosen zoning restrictions that limit population density.
Otherwise, expect more of the same.