Thursday, October 8, 2015

Housing, A Series: Part 67 - Better Numbers for City Rent Inflation

I have used numbers from Zillow that estimate absolute price/rent ratios for each metro area, which gives me a better estimate of home prices and expected rent inflation in each city.  I have updated the numbers from yesterday's post here.  I have used Zillow's home value data for each metro area to establish relative Price/Rent ratios in 2014 for each city.  Then I have used my BEA, Financial Accounts of the USA, and Case-Shiller data to complete the time series.

As I suspected, pegging the Price/Rent ratios to equal values in 1995 did cause home prices to be overstated in the low cost cities and understated in the high cost cities.  Even with that data, the four largest cities (identified as light blue dots in the scatterplots) lined up pretty well on the 45 degree line between actual rent inflation and expected rent inflation needed to justify market home prices.  This adjustment pulls the other cities more in line with a 1:1 relationship.

To review the last post, the 125% increase in home prices from 1995 to 2005 can roughly be divided into three causes. (1) higher intrinsic values of homes because lower long term real interest rates increase the present value of future rent payments and (2) higher prices simply due to the excess rent inflation over that 10 year period, which was significantly higher than non-shelter inflation.  This leaves the source of only about 30% of home price gains unidentified.  I have argued that this can be attributed mostly or completely to the persistence of rent inflation, which creates a growth  factor in the intrinsic value of home ownership.  In aggregate national data, this appears to bear out.  The rent inflation needed to justify home prices appears to be close to the amount of rent inflation that persists.

To confirm this idea, I have compared the actual rent inflation since 1995 with the rent inflation required to justify home prices in each individual city.  There is a lot of noise here, because home prices will be based on long term expectations, and short term local rent inflation can fluctuate significantly.  But, generally the cities with higher rent inflation have seen a similar rise in rent expectations, because our supply problems are regulatory in nature and persistent.  So, rational expectations appear to be the cause for most of that last 30% of home values at the height of the boom, leaving little or none of the boom remaining to be explained by credit expansion, monetary accommodation, or speculative overreach.

The return of high rent inflation while home prices remain well below the levels that demand this level of rent inflation means that there is an unprecedented amount of policy space available for real interest rates to rise in an expanding economy, while building and home prices both rise substantially.  There is ample room, and great need, for all of these things to happen.  Keep in mind that nationally, rent inflation is again running at more than 1% above core CPI inflation, so as we move past the crisis period without a homebuilding response, the dots in the second scatterplot will drift to the right.  Rent inflation expectations now should be higher than recent actual rent inflation.

Wouldn't it be a wonderful problem to have if a decade from now, we have disruptive levels of rent deflation in Silicon Valley, which really do lead to massive capital losses for real estate owners?  I daydream of a country that could make that happen, and that would allow 4% inflation for a few years while that did happen, in order to help prevent defaults and failures among those real estate owners without calling stability a bailout.  Millions of additional young idealists could move there and join in the technological renaissance that produces so many good things for all of us.  Let's teach those builders and speculators a lesson in boom and bust cycles.  Let's let them build!  Let's see what happens when there is a market supply response!

The city charts updated with the new required rent inflation data are below the fold.  As before, expectations were not out of line in any of the high cost cities.  And, with this new data, the anomaly that made low cost cities appear to be the only places with unrealistic rent inflation expectations, has disappeared.  With Price/Rents adjusted down, housing markets in Detroit and Cleveland now appear to be reasonably pessimistic.


  1. Nice blogging.

    Unfortunately, both left-wingers and right-wingers seem to accept that a city has a right to zone property. And forget about push-cart vending.

    So the Fed will continue to suffocate the economy.

    Maybe it will become hip to move to Buffalo or Milwaukee, and this problem will solve itself...

    1. Maybe we can get Stanford to set up a satellite campus in Houston. :-)

  2. TravisV here.

    In case you missed it, you might be interested in this Alex Tabarrok post:

    1. Yes. I'm a broken record on this. There is tremendous networking value in the cities that house cutting edge industries, and instead of being turned into consumer surplus, it is being captured by the firms, workers, and ultimately real estate owners, in those cities. I've become convinced that this issue explains all the anomalous economic trends of our time - stagnant wages, income variance, high capital income, low interest rates. Not only are costs higher, but there are millions of productive workers locked out of these engines of opportunity. So, we have both the problem of eliminating real production and the problem that what production we have managed to create is being claimed by real estate owners instead of consumers.

  3. Yes, and those who work in these areas but are forced to deal with long commutes to do so adds to these deadweight losses.