Well, Wall Street’s eggheads had it half right. We did wind up with a true-up, but not where they expected. For starters, the Street’s misconception is easy to understand. There is a very strong correlation between the monthly ADP numbers and the private payroll numbers reported by the government. Since 2001 the coefficient of determination (R2) between the ADP report and the BLS is 0.925 and since the start the Great Recession the R2 is 0.97.
Intuitively, that means that over the last ten years anywhere in between 93% and 97% of the movement in the BLS number could be explained by the ADP report. Furthermore, over the last 117 months the ADP has been presenting its report is has overstated the BLS number 60 times and understated it 57 times.
Of those 60 overstatements the average was 49K with a 99% confidence interval the overhang would fall in between 37K and 61K; hence the Street’s confidence heading into Friday’s job’s report. In this vein, the BLS number for private payrolls wound up falling 3.9 standard deviations above the historical mean!
Where did the Street go wrong? It went wrong in assuming the ADP number was correct. Consider this, for the 13 months in between November 2009 and November 2010 the ADP reported a net gain in private payrolls of 112K which worked out to an average gain of 9K per month. Over the same time step the BLS reported a net gain of 1,225K or 94K per month. So either the BLS was overstating private payrolls by a factor of 11-to-1 or it was understating them by a factor of 0.09-to-1 relative to ADP.
As analyzed in today’s issue of The Schork Report, last Wednesday’s reported gain by the ADP was 2.4 standard deviations above its mean since the start of the decade. Not only was it the largest reported gain on record, it was also 3.2 times greater than the largest month-on-month gain since the official end to the recession. Therefore, last week’s ADP number was highly suspect.
Why then did Wall Street place so much emphasis on it?
Stephen Schork is the Editor of The Schork Reportand has more than 17 years experience in physical commodity and derivatives trading, risk systems modeling and structured commodity finance.