Comparing housing cycles since the 1980s

While reviewing the Gjerstad and Smith op-eds (1, 2, 3, 4), it occurred to me that it would be useful to compare a set of counterfactuals of historical real estate valuations. This will use Erdmann Housing Tracker data, so it will be for subscribers.

Basically, the op-eds I reviewed were an example of my “World’s Most Niche Meme”. When housing supply is inadequate, rents take an increasing portion of household income over time. But, since there are always cyclical fluctuations in housing and in the economy in general, the secular rise in costs is inevitably experienced as a series of increasingly excessive cycles. When cycles are rising, it feels like the worst cycle ever, and when cyclical pressure is negative, it wrongly feels like a release of pressure and a return to normalcy.

Figure 1

Of course Vernon Smith is an icon in economics and his work in experimental economics is pivotal. But, unfortunately, this setup is a trap practically optimized to capture him. Because, when the housing market inevitably feels like it is entering unprecedented levels of cyclical excess, Smith’s work is sitting right there to confirm it.

“Is this a bubble?”

“Of course it is! Smith proved that bubbles can even exist in lab experiments.”

It adds confidence to an intuition that, in this particular case, happens to be wrong but very tempting.

Behavioral economics and macroeconomics both suffer from the same problem. They use variables (herding, irrationality, etc. for behavioral and long & variable lags of Fed interest rate targets for macro) which are probably realistic reflections of some real effect. But, the variables give you practically infinite degrees of freedom. And, so they are theoretically correct and almost certainly practically wrong in many instances where they are used. I think, as a practical matter, it probably gives you a competitive advantage to ignore them until that problem can be fixed, even though they almost certainly contain some theoretical truth. That is what makes it so hard to ignore them.

Below the paywall, I will use the components of the Erdmann Housing Tracker to estimate the causes of variations in aggregate home values over time, to try to get to the bottom of what is cyclical and what is not.