Non-Farm Payroll Stunner, In More Ways Then One...

After earlier data, including Wall Street layoffs , Chicago PMI, ADP Private Payrolls (which is usually off but to the upside, this one came in shockingly low) and the ISM Manufacturing Index, all showing lower employment in the sub-indicies, the expectation was for a pretty bad NFP number today, in fact consensus was at 135k vs a prior print of 200k, but at 8:30 today, NFP printed at an incredible 243,000. That's a stunner! Here's another...

As I half kiddingly joked yesterday about the "Bureau of BS, or rather BLS", it seems Goldman warned this morning that there would be significant changes to the way employment was calculated as the BLS revised it's model this month. From Goldman:


 'Non-farm payroll growth should slow on seasonal and weather-related factors'
 We forecast that payroll employment growth slowed to +125k in January from +200k in December. The deceleration largely reflects less of a weather-related boost (weather remains milder than usual, but not much more so than in December) and a reversal of positive seasonal adjustment distortions in the December survey (last month the category of “couriers and messengers” was boosted by 42k on temporary hiring for package delivery persons during the holidays; this increase is likely to be reversed in the January report). Even with slightly slower payroll growth, we still think the unemployment rate should fall to 8.4% from 8.5%, although this is a close call.


Today’s report will also include annual benchmark revisions to both the payroll and household surveys. In the payroll survey, the BLS we realign the level of employment with unemployment insurance records. In September 2011, the BLS estimated that this would add 192k to the level of employment. Today’s final announcement should be close to that figure. Historical data will also be revised for new seasonal adjustment factors and updates to the “birth/death” model. Separately, the household survey will be adjusted for new “population controls” incorporating data from the 2010 Census (essentially these are the scaling factors used to translate the monthly survey into a national estimate). These revisions mean that the raw change in household employment will have virtually no meaning. The BLS will report adjusted household employment estimates which remove the effect of revised population controls, and investors should focus on these figures.
Payrolls: GS: +125k; Consensus: +140k; Last +200k. 
Unemployment: GS 8.4%; Consensus: 8.5%; Last 8.5%. 
Earnings: GS +0.2%; Consensus: +0.2%; Last: +0.2%


From Bloomberg:


With the release of January 2012 data on February 3, 2012, the establishment survey will introduce revisions to nonfarm payroll employment, hours, and earnings data to reflect the annual benchmark adjustment for March 2011 and updated seasonal adjustment factors. Not seasonally adjusted data beginning with April 2010 and seasonally adjusted data beginning with January 2007 are subject to revision.

And the net effect of revisions?

1.2 million people dropped out of the labor force in 1 month! This sent the civilian labor force to a 30-year low of 63.7!  An election year stunt?

Look at the charts...

Persons not included in the labor report calculations...

That's a sharp 1 month jump.

Labor Force Participation Rate...

This chart goes back to January 1984, a 30 year low.


From Bloomberg:

About 113k of NFP gain from “low wage jobs,” David Ader, strategist at CRT Capital Group, writes in note. Additionally, “we didn’t see the drop in courier and messengers as expected - but suspect we will." Moreover, ‘‘long-term stress remains at the U6 measure at 15.1% is still high, but likely falling due to people leaving labor  force, and duration on unemployment remains over 40 weeks."