Reader Steve Kopits again asserts that VMT, gasoline consumption, GDP all imply a recession in the first half of 2022write his firm arguments “Obviously the public also doesn’t like macro indicators much given the recent polling going into November.” I agree, the public isn’t very happy, as evidenced by the University of Michigan sentiment indicator. But that’s not to say we’re in a recession.
figure 1: The Misery Index is the sum of inflation and unemployment in % (blue, left scale), and the University of Michigan Consumer Confidence Index (brown, right inverted scale). The NBER uses shades of grey to define the peak and trough dates of the recession. Source: BLS calculations by FRED, Federal Reserve Bank of Cleveland, University of Michigan, NBER, and authors.
Obviously, sentiment falls during recessions, but it also falls at other times, so the match isn’t complete. The same goes for the pain index (what Arthur Okun calls the “discomfort index”). Otherwise, especially with regard to the sentiment index, we will be talking about the 2011 recession. Note that the “Misery Index” has declined since June, while the Sentiment Index has been rising since June.
Two observations (reviewing what I wrote in this post postal):
- According to the regression coefficient and the standard deviation (or “beta”) coefficient, unemployment is more weighted in the University of Michigan Sentiment Index than y/y inflation.
- Both the misery index and the sentiment index are poor predictors of a recession over the next 12 months.
As an additional observation, both are in fact worse than the 12-month lag of the 10-year to 3-month horizon while predicting a recession! ! ! ! (using McFadden R2 as metric). Of course, this isn’t obvious to someone who can’t understand what a probabilistic regression result means.
