Economist Writing Every Day

In a post from July 2021, I discussed housing affordability and “zoning taxes” — in other words, how land use restrictions such as zoning were driving up the cost of housing in some US cities. San Francisco, Los Angeles, Seattle, and New York stood out as the clear outliers, with “zoning taxes” adding several multiples of median household income to housing costs.

The paper I was summarizing used data from 2013-2018, and it’s a very well done paper. But so much has changed in the US housing market since that time. In my post, I pointed to a map from 2017 showing that a large swatch of the interior country still had affordable housing — loosely defined as median home prices being no more than 3 times median income.

To see how much has changed so quickly, consider these two maps for 2017 and 2022 generated from this interactive tool from the Joint Center for Housing Studies.

Notice that in 2017, much of the interior of the country from the Great Plains down to Texas, and over to the East Coast (except for MSAs right on the coast) was generally what we would consider “affordable”: dark blue in color (housing prices under 3 times income levels) or at least in light blue (housing prices 3-4 times income levels).

But just 5 years later in 2022, the dark blue MSAs are almost completely gone, and the light blue ones are vanishing too.

In my July 2021 post I concluded by saying that “As demand increases, and supply-side restrictions remain in place, we would predict the same thing will happen to Columbus as happened to San Francisco. But probably not for decades.”

I was wrong. It didn’t take decades for Columbus and similar cities start to see zoning regulations really start creating a binding constraint on supply. It only took a few years (and by the time I was writing in 2021, it was basically already true, we just didn’t have the data yet). The following chart shows annual data for this same home value-income ratio for Columbus and three other cities that were relatively affordable from 1999-2017 but are now creeping up into unaffordable territory.

The old “zone of affordability” that I talked about a few years ago is largely gone.

But to make matters worse, let’s consider four other cities that are generally considered the epicenter of the early 2000s “housing bubble” in the US. Important note: the y-axis is very different from the prior chart!

The pattern for Las Vegas et al. looks a bit different. These MSAs went from affordable in 1999, to unaffordable in 2003-2007, then back to affordability for a few years. Now they are clearly back to unaffordable territory, with home prices at 6 or even 7 times incomes. For each of these four cities, that’s even higher than the peak of “housing bubble.”

I put “housing bubble” in quotation marks several times above for a specific reason. You could look at both of those charts and say “wow, it looks like we are in another housing bubble, and now it’s spreading across most of the country, not just a few cities.” In fact you can find people (including Robert Shiller) saying as far back as 2013 (more than 10 years ago!) that Las Vegas was once again in a housing bubble.

But this reaction is contingent on granting that the increase after 2003 in Las Vegas et al. was in fact a bubble, which most of us take to mean an asset bubble that was inflated by speculation which was, in hindsight, unwarranted. It looks that way in the chart! And indeed, most observers at the time and since have called it a “bubble.” But what if it wasn’t a bubble? What else could it be?

The alternative explanation of the rise in housing prices in the early 2000s is clearly stated by Kevin Erdmann and others as a situation where demand in particular cities was growing much faster than supply was able to expand. The increase in housing prices was bad from a social perspective, but it made perfect sense given the constraints (zoning, etc.). In this telling of the history, it was a housing boom, not a bubble, but a boom that ran up against inflexible supply constraints.

Why then did housing prices crash? That seems to indicate “bubble” is a better explanation than “boom.” The alternative explanation for the crash (stated most clearly in Erdmann’s book, but this summary by Tim Lee is also very good) is that through a variety of federal policies (including Fed policy) we killed demand. The bubble didn’t pop because of its irrational nature. It “popped” because policy popped it. The resulting recession in 2007-2009 was thus possibly avoidable, in this telling of the history.

I hope I’ve explained the Erdmann hypothesis correctly, as I’ve just come to fully digest it myself. I will say that I’m convinced it’s probably correct. Maybe you aren’t convinced yet (read his book), but grant for a moment that he is correct about the history. What does it imply for the housing market today?

One implication is clear: don’t “pop the bubble”! This won’t solve the problem of unaffordable housing, and it will probably lead to a recession. You can only solve the unaffordability problem by increasing the supply of housing. Subsidizing demand won’t work — it will only increase prices further. Suppressing demand won’t work — it will leave people in other unaffordable situations (much of the increase in Las Vegas et al. was due to migration from cities with even worse housing situations), and may cause a recession.