10 Comments

Isn't one explanation for the Schiller "gap" simply ZIRP? It shows up in the price of equities, as well. In that sense, housing is like any other risk asset--it inflates when money is cheap.

The US population of working age adults--homebuyers and growth-makers--has been flat-to-down since ~GFC. We've had unit-growth (supply) in excess of people-growth (demand) for ~15 years, and yet prices have gone up, coincidental with rates coming down. I understand the supply-side arguments--I really do--but the elephant in the room is on the demand side. There is at least some evidence that when people-driven demand growth came to an ebb, cheap money stepped in to fill void (and has been doing so ever since, until just recently).

More here: https://www.therandomwalk.co/i/120973131/consider-the-everything-bubble

Expand full comment

I know it seems like a correlation (40 years of declining rates and 40 years of rising home prices), but it is spurious.

Besides the fact that empirically the correlation fails (lackluster construction, low consumer inflation, etc.) ZIR is not a P. Persistently low rates are not the product of loose money. If anything, it is quite the opposite.

Expand full comment

I'm not sure about 40 years, but certainly since ~GFC, which is when the growth of working age people began to flatline. Goods inflation stayed low, but asset price inflation most definitely did not. I agree that low rates drive loose money . . . that's kind of my point. Fed says "we'll lend against houses for next to nothing" and, wouldn't you know it, lots of money starts flowing to houses. Money goes where the Fed wants it to go.

There are lots of reasons that we're different than Japan, but the same observation about the decline of people-growth leading monetary stimulus applies. When Japanese working age pop flatlined, BOJ started to intervene. If high rates "kill demand" then low rates "bring demand to life," which is necessitated by the lack of other sources of demand, i.e. new people.

Expand full comment

What should have happened after GFC, according to your model of thinking?

Expand full comment

I'm not really sure how to answer that. All I'm suggesting is that the relationship between rates and asset prices applies to houses too. You observe an increase in home prices above inflation and attribute all of the variance to supply-constraints. But there is another possibility, which is subsidized demand (particularly when unsubsidized demand, i.e. the number of people who want houses, stays flat over the same period.

It's true that goods inflation was moderate over that period (and from that we concluded there were no consequences of low rates) but what if we were wrong (at least in part)? What if goods inflation stayed low bc we offset so much cost via offshoring and technology?

Expand full comment

The reason I ended up with the supply thesis is because the demand theses failed every time I tested them. Really, my first dip of a toe into the housing topic 8 years ago was a realization that many of the demand-side explanations for the pre-2008 market were baseless on their own terms. After trying to figure out what happened for some time, I stumbled into the supply explanations, which really explain everything that is important. Since I started using the empirical model that I update here at the substack each month, I can test the effect of interest rates thoroughly. I would have loved to have found some basis for at least attributing some price appreciation to rate variations, but I cannot. I can't even plug it in as a coherent control variable because once I account for supply constraints, interest rates just don't correlate with prices in a systematic way.

Here is a post with links to some of my research. Obviously, you aren't obligated to read it, but it lays out a robust set of empirical investigations about how supply constraints basically explain everything. (A small credit boom before 2008 was followed by a much larger credit bust after 2008, that do account for some change in prices. Interest rates do not - at least not in any way I have been able to document.)

https://kevinerdmann.substack.com/p/research-roundup

Expand full comment

I will take a closer look, but I suspect I'm too dumb to really interrogate the model. It just seems intuitive to me that houses, like other risk assets, particularly financed ones, are sensitive to rates. Especially when the "real" demand driver, ie people, stayed flat.

Expand full comment