ECRI: PREMATURE TO CALL FOR DOUBLE DIP
24 September 2010 by Cullen Roche
6 Comments
Lakshman Achuthan, managing director of the ECRI says it’s premature to be calling for a double dip recession. The ECRI’s leading indicator fell to 122.2 from 122.6 this week while the index’s annualized growth rate rose to -8.7% from -9.3%. The much reported growth rate has improved in recent weeks. All in all Achuthan believes it’s too early to be too bearish:
“After a brief plunge in the late spring, the WLI has been fairly stable throughout the summer and into September, suggesting that it is still premature to predict a new recession.”
http://www.reuters.com/article/idUSTRE68N2UY20100924
“After a brief plunge in the late spring, the WLI has been fairly stable throughout the summer and into September, suggesting that it is still premature to predict a new recession,” said Lakshman Achuthan, managing director of ECRI.
ECRI will never call a recession unless we are clearly in one no matter how dire the numbers become.
… and the bears have no shares
Equities globally still to move from overshoot on downside fuelled by ‘pullback’ in ECRI (and similar) to overshoot on upside as panic buying sets in
ECRI makes their WLI data available at their resources page. My counts differed slightly from Rom Badilla’s counts that you posted in June, but were very similar. When counting negative growth rates, I didn’t require them to be consecutive. I allowed positive readings that were minor and sporadic “noise.” These “noise” readings were not included in my count of weeks with negative readings.
I found seven “slowdown” periods whose counts of negative readings ranged from 12 to 34. The worst growth rate in each period ranged from -1.0 to -6.8. Contrast this with the seven recessions (as determined by NBER): The worst growth rate in the double dip recession of July 1981 to Nov 1982 was -5.2. The worst growth rate in the other 6 recessions ranged from -8.2 to -29.7. Note that extended negative growth rates resulted in seven slowdowns and seven recessions, which is consistent with ECRI’s claim that there is a 50-50 chance of a recession.
In the current period we have 16 weeks with negative growth rates, and 14 of them are worse than -6.8, which was the lowest reading in the seven slowdown periods. The current period is certainly more pronounced than prior slowdowns, but it is not yet persistent enough for ECRI to call a recession. I disagree with Rom Padilla’s use of 19 weeks as a guide for what is persistent enough to call a recession. It is based on his calculation: “The ECRI Growth Rate Index stayed in negative territory for about 19 weeks on average before the onset of the seven recessions (16 weeks if we count only consecutive negative readings).” The average isn’t appropriate here. Padilla found that four of the slowdowns had negative periods of more than 19 weeks. That’s more than a 50% error rate in calling a recession based on persistence. I’m only aware of two calls that ECRI made. In March of 1991 they called the recession after 34 weeks of negative readings. In March 2008, they called a recession after 30 weeks of negative readings. I don’t know what their criteria for “persistence” is, but it must be more than 19 weeks.
Sorry for twice misspelling Rom Badilla’s name.
Annualized growth rose from minus 9.3 to minus 8.7% ? That’s still over 8% in the red ! Did he get too much heat from the government wonks in D.C. ?