After the peak of the housing bubble in 2006, U.S. house prices fell for six years, until 2012. Are these memories getting a little hazy?

The Federal Reserve, through forcing years of negative real short-term interest rates, suppressing long-term rates, and financing Fannie Mae and Freddie Mac to the tune of $1.8 trillion on its own vastly expanded balance sheet, set out to make house prices go back up.  It succeeded.  Indeed it has overachieved.  Average house prices are now significantly higher than they were at the top of the bubble.  This is shown in the following 20-year history of the familiar S&P Case-Shiller national house price index.

Picture1 pollock

If you already own a house, the price boom makes you feel richer.  But if you are trying to buy, it makes houses less and less affordable.  House prices rise not only faster than inflation, but faster than your income.

So far, we have considered nominal house prices.  But since the old 2006 peak, we have had more than a decade of general inflation, as the Fed strives for perpetual depreciation of the dollar’s purchasing power at a rate of 2% a year.  The aggregate increase of the Consumer Price Index from 2006 to 2017 was about 24%.  We need to consider house prices on a real (inflation-adjusted) basis.  In real terms, the same period of history is shown in Graph 2.

Picture2 pollock

From 1987 to 2000, the average real (inflation-adjusted) annual increase in U.S. house prices was 0.3%. This 0.3% annual rate is the same as the very long term trend increase in U.S. real house prices, as calculated over the 117 years from 1900 to 2017, by the Credit Suisse Global Investment Returns Yearbook for 2018.

In other words, in addition to giving you a nice place to live, it appears that over time on average, houses provide a good inflation hedge, plus a little, but not plus very much.

As Graph 2 shows, after 2000, in real terms the housing bubble expanded and contracted quite symmetrically, bottoming out in 2012 just about on its trend line.  But it did not resume its trend behavior.  The Fed was on the case, and up real house prices went rapidly again, rising over 5% a year on average from 2012 to 2017.  Their current real level is equal to that of mid-2004, when the bubble was already well inflated, and far over—28% over—their trend line as extended from 2000.

Graph 3 shows Nobel Prize-winner Robert Shiller’s estimate of U.S. house price increases since 1953, compared to the increase in the Consumer Price Index.  The two track very closely for decades, swiftly part company as the housing bubble inflates, get pretty close by 2012 as house prices correct, then once again dramatically diverge, bringing us to now.

Picture3 pollock

Looking at the pattern of real house prices, I conclude that because of the Fed, house prices are too high, and real house prices will fall.  Of course, this does not mean that nominal prices must fall—they could go basically sideways for some time, while inflation continues year after year, as the Fed plans, and real prices would be falling.  The supply of new house construction has been much more constrained than in the last cycle, which is consistent with this scenario.

That would be a “soft landing,” as is always wished for, and might turn out to be consistent with the Fed’s gradualism in undoing its house price inflation program.  The Fed is letting its mortgage-backed securities portfolio run off through maturities and prepayments, not selling any of the $1.8 trillion it still owns, and the same for its $2.4 trillion of long-term Treasury debt.  It is raising step-by-step its target short-term interest rates, but they are still very low and negative in real terms.  Nominal short rates of 1.5%, compared to inflation of 2%, obviously gives you a negative ½% real yield.  In a “normalized” financial world, the real yield will be positive, not negative.

Falling real house prices might also be composed of part inflation and part declining nominal house prices.  That would be consistent with rising nominal and real long-term interest rates.  Over five decades, since 1971, 10-year Treasury note yield has averaged 2.5% more than the CPI inflation rate.  Interest rates on 30-year mortgages have averaged over these years 1.7% more than the 10-year Treasury.  If inflation runs at 2%, that gives us 2% + 2.5% + 1.7% or in total 6.2% as the normalized mortgage interest rate.  That is a lot higher than the current 4.5% and would certainly restrain or dampen house prices.  The timing, and whether the landing will be soft or not, is just what nobody knows or can know.

What can the Fed do about this?  Nine years after the 2007-09 crisis, it needs to withdraw its radical interest rate and investment interventions.  I am certain that the Fed does not want to find out what would happen in the market if it actually put its MBS and long-term Treasuries out for bids from Wall Street.  So about all it can do is continue with the gradualist program, keep up its rhetoric about how very gradual everything is, and hope house prices have a soft landing.  It is said that “hope is not a strategy.”  But that’s the best the Fed has at this point.  The rest of us should constrain leverage in the housing sector as the fall in real house prices unfolds.

Image credit: SukanPhoto


Featured Publications