Are There Too Many Homes In America?

Includes: ITB, IYR, REZ, XHB
by: Modeled Behavior

By Karl Smith

The inventory of for sale home continues to collapse. My best guess is that the rate of collapse is slowing. Yet, we still have a ways to go before we can expect there to be more homes for sale each month than there were the same month last year.

We have no really good estimates on the number of vacant homes in America. The Tom Lawler census pegged the number at around 3 million, or 2.4%, in 2010. It has fallen substantially since then. Apartment vacancies continue to fall. Perhaps the increase in multifamily construction that began two years ago will begin to reverse that trend, but it will be close.

Reis estimates 100,000 units coming online in the latter half of the year, which is nothing to sneeze at. Still, the first quarter of 2013 absorbed 36,000 units even though the first quarter tends to be the slowest for landlords. And, there is still the issue of household formation. Our latest estimates of aggregate weekly payrolls (workers x hours x hourly pay) for production workers still shows a downshift in 2012, although that is likely to be revised up. But even before the likely revision, the series looks to be turning upward.

This is a good sign for the types of workers most likely to be looking to form new households. Provisional estimates from the CDC suggest that the recession-induced decline in birthrates has also leveled out. And like all such data, this series comes in with a lag, is revised over time, and tends to miss turning points. If births were turning upward, this is what the real time data would look like.

Larger payrolls and a stabilizing -- if not increasing -- number of births portend increases in household formation. What about supply? Residential construction is on the mend, but completions are still at record lows.

Not to mention the deficit yet to be made up. Assuming an average flow rate of about 1.5 million homes per year, you can eyeball the stock deficit by comparing the mass above and below the zero on the following graph.

Then there is the financing. Can American households carry the cost of home ownership? The National Association of Realtors' Affordability Index may seem comical -- it suggested that affordability did not begin to wane until 2005 and then skyrocketed during the Great Recession.

That's in part because the importance of interest rates and inflation dynamics is missing. High inflation and tight Fed policy made buying a home extremely difficult in the 1980s, though it made keeping a home easier as mortgage payments fell off rapidly as a percentage of income. In part, however, the oddity of the index is a function of how the NAR chooses to chart its data. If you invert the index, implicitly asking how typical mortgage rates compare to income instead of the other way around, you get this:

The early 1980s were brutal, but they quickly gave way to a blissfully normal 12 years from 1992 to 2004. In 2005 to 2006, monthly payments quickly leaped 20% above what an average family could afford, but are now nearly 40% below. That's close to Jed Kolko's estimate that buying is 44% cheaper than renting. And, of course, debt-service payments have collapsed to record lows as well.

What I would add, however, is that they are highly likely to fall further. Every time an underwater homeowner executes a short sale, this ratio falls further. That's because the short seller's payment will be less than the payment of the new buyer. That's what it means to short-sell a home. A high payment, but likely to default, borrower is replaced by a low payment, but less likely to default, borrower. As short sales continue to plow forward, this ratio will continue to feel downward pressure -- likely to ever new lows.