On Recession Odds: Have They Declined?

I wrote last week about the predictive power of the treasury yield curve, and how a curve inversion which started in August 2006 forecast a decent chance of a recession between April 2007 and March 2008.

The WSJ's Greg Ip points us to another forecasting model for recessions which combines credit spreads with Treasury spreads. This model is very similar to one developed by Fed Chairman Ben Bernanke's staff.

Ip asked Brian Sack, an economist at Macroeconomic Advisers, to run the model using up-to-date data and reports that "'the odds of a recession [in the next 12 months] are less than 10% (see chart below), down from over 40% in early 2007,' Mr. Sack says."

While that's very similar to the New York Federal Reserve's prediction, it's also slightly misleading.

The issue here is that most recession prediction models forecast probabilities about one year ahead. So, while the 10 percent chance for a recession applies to today through roughly early 2008, the 40 percent figure Sack cites above in all likelihood covers the current time period.

So, if you think we're in a recession now, the Bernanke model wouldn't disagree with you.