However, I am making an important concession to the baying hounds on comment sections of other posts I’ve written about affordability, who rightly say that there is no evidence that simply increasing the number of housing units means that prices will drop. In fact, there are weird examples of places like Phoenix and Las Vegas where prices went up even as supply increased.
When critics of my supply-side rants have called on me to cite the evidence of this, I decided I'd look for that evidence. I didn’t find anything that “proved” that simply building more units always lowers price. What I did find is a great deal of head-scratching in the literature about housing price and what accounts for it.[pullquote]My wager is based on the idea that even though people have other options here—like moving to Maple Valley—if we create enough housing here of all kinds, there are enough factors to make moving to Seattle better than moving to outlying areas.[/pullquote]
The two best things I found on this were a complex study on interest rates and housing price (Housing Price Booms and the Current Account) and an article about that study by Edward Glaeser. The study and Glaeser's review of it rehearse most of the discussions about housing prices. It’s all very deep, complex, and fun stuff, including Baye's Rule and discussions about how “belief dynamics can temporarily delink house price from fundamentals, so that low interest rates can fuel a house price boom” (from the study).
Glaeser sums up the problem nicely in his review:
While supply elasticity appears to impact how bubbles play themselves out, there is plenty of variation that is not explained by supply elasticity. Both Phoenix and Dallas appear to have fairly elastic supplies of housing. Dallas experienced little price appreciation during the boom—it built a vast number of new homes.
Glaeser points out next that, “Phoenix managed to see both huge increases in construction and huge increases in prices” (emphasis mine). What’s true about both Phoenix is that they are pretty libertarian places where getting a gun, a bottle of booze, and a building permit are pretty easy.
What is a supply hawk like me to do with such inconclusive evidence? There are lots of wonky silver linings.
First of all, there is a difference between demand for housing and demand for credit for housing. That is, what people demand is housing, but in order to get the housing they have to get credit for an enormous loan first. If I am going on the housing market, I am really on the market for a loan for housing, not the actual house itself. Once I get the credit at a rate I can afford, then I go get my house. The Adam, Kuang, and Marcet framework shifts the discussion of the housing boom away from the idiosyncrasies of the housing market—location, number of bathrooms, color of the garage doors—to the market for the money to buy housing.
This is a pretty satisfying shift, but it still doesn’t fully grapple with the way people look for housing, and how the market responds to that demand. Location is critical, for example. If I had free land on the waterfront in Nice and built really crappy, tiny little apartments facing the Mediterranean, I could probably charge just about whatever I wanted for them, and get my price. On the other hand, a three-bedroom house in Omaha hasn’t appreciated much over the last 20 years, notwithstanding the recent housing bubble.
All of this is very interesting, but some important facts here in Seattle haven’t changed even without drop-dead evidence that increasing supply will lower price. Affordability is still a relationship to price and is therefore a pretty subjective and bad measurement for understanding the housing market. A $1,000,000 condo might be affordable to you but not to me, but the price is the same for both of us.
Also, the Seattle housing market is constrained by some pretty unmovable geography that means we simply can’t sprawl like Dallas or Phoenix. Additionally, existing single-family neighborhoods can’t add anymore single-family homes. When it comes to the traditional single family home, we’re all full; we simply can’t meet the demand by building more single-family Craftsman-like homes.
Can I promise that by increasing the number of condominiums and apartments in Seattle that the overall price of all housing will start to fall? No, I can’t promise that. But I’d bet on it. My wager is based on the idea that even though people have other options here—like moving to Maple Valley—if we create enough housing here of all kinds, there are enough factors to make moving to Seattle better than moving to outlying areas.
And that’s where the Seattle City Council has dropped the ball. Those other factors are things like proximity to transit, living in vibrant neighborhoods, and having access to good schools and affordable day care. When the council dithers on promoting more new development, they just add more costs to the production of new housing, and when they fail to aggressively court new growth and development while other cities do, they eliminate the barriers for people in the housing market to look in Burien, Kirkland, or Renton.
We should be competing for growth with those cities the way the United States competed with the Soviet Union in the space race, or the way western towns in the 19th century competed for railroad stations. That means letting private developers, both for-profit and non-profit, build more housing here of all types.
We aren’t Phoenix, where supply—even in an environment with fewer regulatory hurdles—can’t keep up with demand. In Seattle building permits aren’t handed out like candy, land isn’t cheap, and credit isn’t loose anymore. If we all set our biases aside for a second, I think this makes sense: we are one of the best places to live on the planet, and doing things to limit supply can’t possibly help lower housing price, even if nobody can prove that increasing the supply will lower it.