In previous post on the efficient market hypothesis, I sympathized with Scott Sumner’s overly demanding definition of identifying a bubble, because his definition prevents people from taking credit for spotting a bubble an implausible number of years before it popped. The example I gave was Dean Baker, who I recalled as predicting the current housing bubble in 1948. Dean corrected me in the comments, reminding me that it was in fact 2002 when he “called” the bubble, and he links to the paper where made the call.

I strongly suggest that anyone who harbors a notion that Dean called the bubble read his paper, from August of 2002, and consider whether he was calling the housing bubble that popped in 2006, or whether he was calling a housing bubble that hadn’t begun in any appreciable way.

Why don’t I think Dean actually called the bubble? First, in his paper Dean specifically refers to the period of time before we can expect prices to fall as “months”. In his conclusion, he sums up his predictions like this: is likely that the HPI will follow the rent index in the months ahead, first showing considerably slower growth. In later months, it is likely that that the HPI will fall in real terms, and possibly in nominal terms, until it is back near its pre-bubble position relative to the rent index.

He is referring here to slower growth in the “months ahead” and a fall in the house price index in “later months”. In fact, house prices didn’t peak until the second quarter of 2006, four years or 48 months after Dean wrote this paper. From his choice of words here he makes it pretty clear that he did not have anything like that in mind.

Another problem is that national house prices are still above where they were when Dean spotted a “bubble”. According to the most recent Case-Shiller house price indices, the U.S. national house price average is now where it was in late 2003, a year after Dean claimed that house prices were going to fall 11% to 22%. Prices falling well below where they were when you cried bubble is a bare minimum requirement for calling a bubble. Dean has not met this very minimal requirement.

You might think prices are going to fall another 25% and prove Dean right, but prices would need to fall much more than that. If Dean was correct that there was an 11% to 22% bubble in 2002, then after the huge buildup in housing supply that occurred from 2002 to 2006, you’d expect the eventual fall in prices to bring us much lower than 11% to 22% below the 2002 price level. To put it simply, if I said “the supply of gold right now is too high to support the prices we are seeing, prices are going to fall 10% from here”, and then the supply of gold quadruples, prices need to fall more than 10% to prove me correct. This is even more true when you consider that the popping of a huge housing bubble is more destructive than the popping of a small housing bubble, so prices should be lower than Dean forecast due to a much larger destruction of wealth than he expected.

Dean’s a nice guy and all, but I think it really is inaccurate to characterize him as having called this bubble.