The Cleveland Fed issued a report that looked at the historical relationship between lagged GDP growth and the steepness of the yield curve. They compared the ten-year with the three-month bill yield. They used this information to make a probability estimate for a double did recession (click on chart to enlarge). Not surprisingly, they concluded that the probability is only 10%. I would take that bet. If anyone in Cleveland is interested, let me know.
I don’t question the Cleveland Fed’s (“CF”) analysis. If you rely on history to predict the future then the probability of a slowdown is low. This argument has been made by a number of pundits of late. While historically correct, I think this is dangerous logic.
It’s not clear to me that the CF believes in the predictive capability of steepness of the yield curve. Their caveat:
Other researchers have postulated that the underlying determinants of the yield spread today are materially different from the determinants that generated yield spreads during prior decades.
When they say this they really mean, “We don’t trust these results. There is too much “noise” in the current data. We are publishing this to try to blunt the ton of negative media about another recession. We want consumer sentiment to be as high as possible!”
Some reasons why the predictive powers of the yield curve may not hold up this time around.
- Sure the yield curve is steep. That is because we have ZIRP. With 3-month bills at an 1/8th everything looks steep. The bill yields since 1960 ()
We have never had a period like this. We are living in a dirty float. These are not true market rates. Nor are they sustainable without substantial consequence. We are on steroids. The results we achieve while on these powerful drugs should be discounted.
- The NY Fed’s 1.75T QE operations are having a continued impact. The result is extraordinary excess liquidity. Don’t trust current bill yields. This will change at some point.
- The yield curve that the CF relies on has collapsed of late.
- The global macro story has benefited short-term liquidity in the US for the past six months. This too is temporary. It may be unwinding as I write.
- Almost all other economic indicators are telling us a different story.
- Unless Paul Krugman gets his way and we get another $500b of useless spending, the fiscal side of the equation is heading into a wall. This is the most compelling side of the recession story.
A true double dip recession is a very rare occurrence. Once in the last 60 years. I am not forecasting one because of that. I think it is much more likely that we are entering a period of prolonged sub par growth. I see us bumping back and forth around zero for several more years. The NBER will not call that a recession. It sure will feel like one.
The analysis by the CF should be tempered given the “materially different” conditions that exist today. The probability of recession is significantly higher than they have suggested. Similarly, any forecast of future GDP growth should be discounted. With that in mind consider this graph () from the same report:
Projecting forward using past values of the spread and GDP growth suggests that real GDP will grow at about a 1.00 percent rate over the next year.
How much should we discount the CF forecast of 1%? I would suggest about 1%. We will have very little net growth over the next eight quarters.
The CBO and the SS Trust Fund look at the longer term using growth rates of 3%. We are going to be lucky to get a third of that. Without growth at or near 3% our deficits will explode. Our social obligations will overwhelm us.
A second recession will kill us. A few years of slow growth will scare us to death.