In a recent debate (which I discuss here), Brad DeLong referred to an apparently decisive argument from Paul Krugman regarding the "hangover theory." Here's Krugman:

[T]he hangover theory, which I wrote about a decade ago, is still out there.

The basic idea is that a recession, even a depression, is somehow a necessary thing, part of the process of “adapting the structure of production.” We have to get those people who were pounding nails in Nevada into other places and occupation, which is why unemployment has to be high in the housing bubble states for a while.

The trouble with this theory, as I pointed out way back when, is twofold:

1. It doesn’t explain why there isn’t mass unemployment when bubbles are growing as well as shrinking — why didn’t we need high unemployment elsewhere to get those people into the nail-pounding-in-Nevada business?

2. It doesn’t explain why recessions reduce unemployment across the board, not just in industries that were bloated by a bubble.

One striking fact, which I’ve already written about, is that the current slump is affecting some non-housing-bubble states as or more severely as the epicenters of the bubble. Here’s a convenient table from the BLS, ranking states by the rise in unemployment over the past year. Unemployment is up everywhere. And while the centers of the bubble, Florida and California, are high in the rankings, so are Georgia, Alabama, and the Carolinas.

I'm going to deal with objection (1) in a forthcoming Mises Daily article. But for here, I want to discuss objection (2).

First, note that the BLS table looks at the yr/yr change in unemployment (by state) from Dec 07 to Dec 08. Now is that really a good measure of whether the bursting of the housing bubble has anything to do with the recession? After all, the bubble had well burst by Dec 07. So if the Austrians--or in fact, any economist who thinks the current recession "started" in housing--are right, you would expect the connection between unemployment jumps, and housing price collapses, to be weaker, the farther along you get from the bursting of the bubble.

Nudged on by an email from the mysterious von Pepe, I decided to check on the relation during a time frame that more tightly captures the bursting of the housing bubble. The OFHEO data is quarterly, and I picked the top of the bubble as 2q 2006. We can quibble with that, but that's what I picked.

Then I picked the other variable to be the change in unemployment (in terms of absolute point changes, not percentages of percentages) from Jun 2006 to Dec 2008.

Then I ranked the states according to these two criteria, and looked at the worst 10 in both rankings. The 7th through 10th slots don't match up, but check out the top/worst 6 slots in both lists:

Ranking of States By Point Increase in Unemployment Rate, Jun 06 - Dec 08
1......Rhode Island (+4.9)
2......Florida (+4.8)
2......Nevada (+4.8)
4......California (+4.4)
5......North Carolina (+3.9)
6......Michigan (+3.8)

Ranking of States By Percentage Drop in OFHEO Housing Price Index, 2q06 - 4q08
1......California, -27%
2......Nevada, -26%
3......Florida, -22%
4......Arizona, -16%
5......Rhode Island, -11%
6......Michigan, -11%

Note that North Carolina and Arizona are the only ones that don't match.

I confess I haven't yet run Monte Carlo simulations to see how likely this result is, out of 50 states, if there were no causal relation. But I'm feeling pretty good about my hangover theory.