Why Real Estate Prices Won't Bottom for Another 5 Years

I am very underinvested in real estate, and have been since we sold our apartment in NY in 2005.  At that time, I thought the market had peaked.  I was wrong, but not too far off.  We rented for 2 more years, then moved to NC where we bought a nice house. However, I have bought no land, nothing commercial, nothing in the equity markets and hence have stayed quite underinvested in real estate for years now.  Today, the Case-Schiller numbers hit a low not seen since April 2009.  Is it time to buy yet?

First of all, take a look at a long term chart of real estate prices, courtesy of Barron's this weekend.

The price scale on the right is the Case-Schiller index value, and today was reported to be 139.  That was down 3.3% from a year ago.   Home prices are still falling.  The peak index value was over 200 during our (now obvious) bubble, and before the bubble, prices hovered in the 110-120 range.  So perhaps we aren't far away from the bottom.  However, housing fell way below its long term trend during the 1920s through the 1940s.  That is entirely possible here.  This chart scares me to be honest.  What will stop the falling knife?

As far as the bull case for housing right now, usually it goes something like the following: "Housing starts (ie new builds) are today only 500,000 homes per year.  And with 1.5mm of new home demand per year on average, then we are depleting inventory rapidly.  There are 3.6mm homes for sale today, so we will have housing shortages very soon."  Hmmmm.

DEMAND FOR HOUSING
To gauge demand for housing, you have to exclude retrades and resales, and look at what really drives long term demand for housing.  While average population age, income and birth rates all matter, essentially net population growth in the US is what drives NEW demand for housing.  Furthermore, if you eyeball the housing demand chart over a very period of time it has equated to around demand growth of 1.4-1.5mm homes per year.  The NAHB also forecasts that long term housing demand for single-family units will grow at 1.5mm homes.  (Personally I think it will be lower, as immigration rates have slowed in the US, and aging populations tend to downgrade from houses to condos or smaller homes, lessening demand).

Nevertheless, using 1.5mm of annual housing demand less 0.5mm of new builds equates to net new demand of around 1mm houses per year in the US. 

Next, I'll take a look at our housing supply.  Because at the end of the day, until supply gets absorbed, we are very unlikely to see any meaningful housing appreciation.

SUPPLY OF HOUSING
First of all, the bull argument for housing entirely ignores the fact that the inventory of homes for sale is greatly understated.  Yes there are 3.6mm homes on the market right now.  But there are also piles of homes owned by banks, and also homes in the foreclosure process that will eventually hit the market.

The breakdown of our country's housing stock more accurately looks like:

1.  Houses for Sale Today:  3.6mm.
2.  Houses Foreclosured On (that is owned by Banks):  2mm
3.  Shadow Inventory:  2 - 3mm homes.  Shadow inventory refers to housing that is "seriously delinquent."  When a homeowner is 90 days or more behind on his mortgage, or is in the process of foreclosure, then that home is considered shadow inventory.

Total Housing Inventory:  around 8mm homes.  Alan Abelson was right on this weekend in Barron's.

So, comparing 8mm of home supply by the net new demand for housing of 1mm homes, means that we have 8 YEARS of supply in the market. That is a far cry from 6-12 months of supply, the level that is commonly cited as bullish for prices.  Or to use the SAAR number of 4mm (just the seasonally adjusted number of homes sold per year), when we have true inventory below 6 months, ie 2mm homes of inventory, then prices tend to rise.  8mm homes is years away from there.  At least 4-5 years. 

(By the way, much of this data came from the Mortgage Bankers Association.)

One of the issues backing up the pipeline of homes for sale has been the bruhaha over foreclosure proceedings.  Because so many mortgages were made, then sold, then re-sold, the paperwork became a disaster.  The well publicized robo-signers, who admittedly had no idea what mortgages that they were transferring, have enabled many receiving foreclosure notices to stay in their homes much longer.

In fact, last year there were 2.9mm homes that received foreclosure notices.  However, only 1mm of those were actually foreclosed on.  The process has gotten extremely backed up not only due to sheer volume, but also because banks are being much more careful now in the foreclosure process.  But that means we have significant shadow inventory that will eventually hit the market, and continuing to add to the supply glut over the next 2-3 years.

ANOTHER METHOD
Another way to look at the housing supply is by examing census data on vacancy rates.  Over time, vacancy rates of homes in the US have been remarkably stable, averaging around 14mm homes.  This is true since 1990, and also true going back to the early 1970s.  Many of these are summer homes, and many are bank owned, foreclosed homes.  I assume for simplicity, that all of the 14mm average number, are summer homes.  In effect, this somewhat adjusts automatically for rental houses, ceterus paribas, that is, assuming generally that the rental market is somewhat stable as well over time.

So today, we have 18.7mm vacant houses in the US.  Comparing that to our LT average of 14mm, I think its pretty clear that we have around 4.7mm excess homes in this country.  Combining this with shadow inventory (which likely aren't vacant, yet), and you also get around 7.2mm of housing inventory.

Very close to 8mm of housing inventory.  Way too much supply.

THE RENT OR BUY ANALYSIS
To me, our first housing purchase was simple.  It cost one $2500-3000 to rent a one bedroom apartment in NY, or with a mortgage, you could buy a place and pay $2000 a month for a similar one bedroom.  This was 1999, and I didn't quite understand the disparity.  I mean, with a mortgage, I am essentially locking in my rent for 30 years (with a fixed rate mortgage)!  With a rental unit, who knows what my landlord will charge me next year, or in 5 or 30 years.  But rent was definitely going higher.  Clearly, real estate prices were poised to go up, at least until buy prices caught up with rent prices.

I point this out because today I see houses in our neighborhood that rent for $2700 a month, but would cost you well over $3,000 a month with taxes and a mortgage.  The "buy" monthly payment is still higher than the "rent" monthly payment.  Depending on where you live, to me this is critical as a first step in examining where the real estate market is heading.  I suspect that the "buy" price will be lower (perhaps much lower) than the "rent" price when we reach the bottom.  However, looking at the chart below tells me that much of the excess premium to buy a home is clearly gone.  Still though, it costs more to buy, and bubbles are almost always characterized by bottoms well below trend. 

INTEREST RATES
I also hear that it's the best time to buy now because interest rates will never be lower.  That is true, and to me locking in a 15 year or 30 year fixed mortgage is a great idea.  If you have an adjustable mortage, even if it's 5 years out, I would refi into something fixed.  We are going to inflate our way out of massive fiscal deficits, and that means rates eventually are going higher.  Maybe much higher.  Don't be lazy and get blasted in 4 years when your ARM resets.

However, I wouldn't bet that low rates alone are enough to entice buyers and end the real estate bear market.  Even in 2009, with $8,000 tax credits on new homes, and artificially low rates, real estate prices barely stayed flat. 

What I think people miss about interest rates is, the entire mortgage market in our country has been nationalized.  Fannie Mae and Freddie Mac today purchase 70% of all mortgages made in the US.  Who is going to lend you money on a mortgage when they get phased out?  (that is Tim Geitner and the Obama administration's plan).  While this will likely take years, it means that mortgage rates are artficially low today, subsidized by taxpayers as lenders, compounded by the mortgage interest deduction (which may or may not stay in place).  At the end of the day, the private market cannot absorb the size of our mortgage market at current rates, rates will have to go up. 

And, as rates go up, prices generally fall.  During the 1970s, housing dragged as rates skyrocketed.  See below, prices were actually down for much of the decade, adjusted for inflation. 

CONCLUSION
The rent or buy analysis seems to indicate that housing hasn't fully normalized.  The supply of housing also still seems very high.  And prices compared to long term trendlines have still not quite gotten to normalized levels.  All of this tells me that housing prices will continue to suffer. 

While prices have fallen from 200 to 139 per the housing index, a 30% fall, it seems likely that prices will fall another 10-20%.  I think its most likely we'll see 120 again on the housing index.  Most housing bear markets last for many many years.  New York prices fell from 1987 until 1995.  Texas housing fell from 1981 after the oil bust, and didn't recover until 1999. We had a tremendous national housing bubble, and working off the excess is not a five year endeavor.  More likely, its a ten year readjustment period, perhaps longer.  Meantime, my recommendation is to keep real estate exposure as low as possible.

If anyone has any short ideas in the equity markets (apart from St Joe JOE, or XHB, neither of which I can get a borrow on), please forward my way.  Thanks!