I was reading an article on Market Watch today which made and interesting argument for why housing prices will continue to fall at least 20%. Basically, he argues that prices will fall until median home prices are a smaller multiple of the median family income.

Today, median home prices are 3.5 times the size of median annual family incomes. This may be down from the recent peak of 4.2 times incomes reached last year, but it’s way above the 2.8 times that home prices averaged during 1984-2000, when lots of homes were bought, sold and built.

And if you think 2.8 is low, check out the early 1970s. That was when home prices were only 2.3 times median family incomes, and housing was selling like gangbusters.

He goes on to say that for prices to get back to 2.8 times the median income they would have to fall another 20%, and to reach the 2.3 multiple would require a price drop of 38%, which he doesn’t think is out of the question.

Definite food for thought, but I have a couple of problems with the conclusions overall.

1. First and foremost, I can’t get behind any arguments based on nationwide medians. The housing market is just too granular for nationwide statistics to be meaningful. A 40% drop in an expensive market like Florida will push prices down even if less inflated areas, like Texas, continue to appreciate. A nationwide statistic is useless to buyers and sellers in both markets.

2. For better or for worse, people want more in a home now than in the 1970’s, and they are willing to pay for that. According to the National Association of Homebuilders, the average home size in the US increased from 1,400 square feet in 1970 to 2,330 square feet in 2004. (Yes, I know what I just said about nationwide averages, but I would argue the trend towards more size in housing is basically universal, it’s the price you pay that varies.) So while the median house currently costs 3.5 times the median income, an increase of about 52% from the 2.3 multiple in 1970, the median home size has increased about 66% in that time!

3. Remember that in the 70’s and 80’s inflation was much higher than today, and interest rates were consequently higher as well. For example, the National Average Contract Mortgage Rate Index (an average of rates for purchasing a single family home) in October 2007 was 6.5%. In October 1981, that rate was 15.47% (source). On a 30 year $100,000 mortgage the payment at 6.5% is $632.07, and the payment at 15.4% is $1,296.49. If you could afford a $100,000 mortgage at 15.4%, you could afford a $205,000 mortgage at 6.5% (disregarding the likely higher taxes and insurance). Focusing on the housing cost/mortgage balance misses the point that the monthly payment is what really determines affordability.

4. In the same vein as #3, there are a lot more financing options available today than in the 70’s and 80’s. Because those high interest rates made payments unaffordable (even with a median price only 2.3 times the median income!), new products were developed and mortgage bankers began appearing to offer them at competitive rates. These new products and increased competition contributed to making a home purchase affordable for more people. This eventually was taken to extreme and led to the current sub-prime meltdown, but the overall increase in accessibility remains.

All this is not to say that I don’t think prices will continue to fall in overheated markets like California, Florida, and the East Coast. But saying that a 20%+ nationwide price correction makes sense based on historic figures just isn’t realistic.

What do you think? Is a nationwide crash still in the offing, does it depend on your neighborhood, or is the worst already over?

Posted by dave on Tuesday, December 18th, 2007 at 10:16 pm.