Thursday, June 7, 2007

It seems that anyone with an opinion can be an expert

(Warning, this post is all text and no pictures).

I guess I'll voice my opinion, and see if anyone ever calls me an expert.

Here's a Seattle Times article, just one of a steady stream on the Seattle housing market, and the national market at large.

The bullet points are:

  • Number of homes sold went down
  • Number of homes on the market went up
  • Median price went up
They cite an analyst, Glenn Crellin, who claims that the shift toward a buyer's market caused people to buy more house, hence the increase in home price.

To me, this explanation is a load of hooey.
  1. It carries the testable prediction that the average size, or quality of homes went up. They don't report if that's true. It seems like a key piece of evidence that would be screamed from the rooftops if it were true.
  2. The claim just doesn't make sense. It might make sense if home size/quality went up at a flat median price. Instead, they're claiming that many home buyer's have the train of thought: "Wow, homes are cheaper than I anticipated! They're so cheap I think I'll blow my budget on one." ...

    huh?
I'll propose an alternate theory. I have no idea if it's true but it makes a bit of sense. In the interest of full disclosure I'll freely admit that this theory fits with what I want to believe.

The housing market is not entirely composed of rational and informed individuals. I don't know the percentages, but a significant number of players in the housing market are poorly informed, misinformed, or just plain irrational. If you want examples, look at all the human-interest stories about sub-prime loans. They invariably feature idiots who make $35k a year with 4 kids, "buying" a $300k house.

The three bullet points above can be explained by the actions of two separate groups of rational and irrational players in the market.
  • Rational home owners see that the market is going to go negative sometime soon. Those with the right circumstances are trying to sell their homes now. In a market, you haven't actually made money until you cash-out. They're trying to cash out, hence the glut of homes on the market.
  • Rational home buyers aren't buying right now.
  • Uninformed or irrational buyers are still buying. This group intersects with people who either buy too much house and/or pay too much for what they buy.
In short, the average sale price went up not because prices went up, but because bargain hunters aren't buying.

Buried inside this argument is an important statistical truth that is often lost on people. Comparing means requires comparing similar groups. If groups aren't matched, a difference in means may reflect a difference in sampling bias. If you eliminate bargain hunters from the market, the average sale price will "go up," but only because you've changed your sample group, not because values actually went up.

Keeping things scientific, here are a couple of testable predictions (that I won't bother to test):
  • Home purchases by successful real-estate speculators have gone down.
  • Demographics of home buyers have shifted (which really is a broader version of the first prediction).

No comments: