JOHN WILLIAMS’ SHADOW GOVERNMENT STATISTICS

FLASH UPDATE

February 6, 2009

__________

January Jobs Loss Was 716,000 Net of
Concurrent Seasonal Factor Bias (Instead of Official 598,000 Loss)

Downside Benchmark Growth-Pattern Revisions
Were Mostly Pre-Election

Alternate-Unemployment Rate Hits 18%

__________

PLEASE NOTE: A full newsletter is targeted for Monday, February 9th. The initial estimate of the SGS-Ongoing M3 for January will be posted over the weekend on the Alternate Data tab, with detail following in the newsletter.

– Best wishes to all, John Williams

Stimulus of $900 Billion Would Be $69,200 Per Officially Unemployed Individual. Beyond the impact of the annual benchmark revisions, payroll employment continued to sink faster than initially reported by the Bureau of Labor Statistics (BLS). Monthly revisions continued to sharp downside revisions to prior months, and the Concurrent Seasonal Factor Bias (CSFB) continued to show major ongoing distortions to the headline reporting of payrolls, as detailed in the accompanying graph and as discussed below. Nonetheless, the year-to- year percentage decline in January 2009 payrolls tumbled further, now rivaling the employment-contraction trough seen in the double-dip recession of the early 1980s.

As to the annual benchmark revision, it was worse than had been advertised, with unadjusted payrolls for the benchmark month of March 2008 being revised lower by 89,000, instead of the initial estimate of a 21,000 downward adjustment to payrolls. In terms of seasonally-adjusted monthly payroll levels, that translated into a downward adjustment in payroll levels in excess of 200,000 for each month from May 2008 on. The effect was to suggest weaker growth patterns in place before the 2008 election. The annualized quarter-to-quarter contraction in seasonally-adjusted payrolls deepened to 1.3% in the second quarter of 2008, versus the earlier estimate of 0.6%; the annualized quarter-to-quarter contraction deepened to 1.5% in the third quarter, versus the earlier estimate of 1.1%; but the annualized quarter-to-quarter contraction narrowed to 3.6% in the fourth quarter, versus the earlier estimate of 3.7%. A more complete analysis of the benchmark revision will follow in the newsletter.

Payroll Survey. The BLS reported a statistically-significant, seasonally-adjusted jobs loss of 598,000 (down 909,000 net of revisions) +/- 129,000 for January 2009, following a revised 577,000 (previously 524,000) jobs loss in December. Annual contraction (unadjusted) in total nonfarm payrolls continued to deepen, down 2.58% in January — the weakest since a 2.67% decline in October 1982) — versus a revised 2.11% (was 2.03%) annual contraction in December. The seasonally-adjusted series also contracted year-to-year, down by 2.53% in January, versus a revised 2.15% (was 1.88%) contraction in December.

Concurrent Seasonal Factor Bias. The pattern of impossible biases (see the Reporting/Market Focus in SGS Newsletter No. 43 of June 10, 2008) being built into the headline monthly payroll employment continued with January 2009 reporting. Instead of the headline jobs loss of 598,000, consistent application of seasonal-adjustment factors — net of what I call the concurrent seasonal adjustment bias — would have shown a more-severe monthly jobs loss of about 716,000. This upside reporting bias has been seen in 11 of the last 12 months, with a rolling 12-month total upside headline-number bias of 1,185,000. A worksheet on this is available upon request.

Birth-Death/Bias Factor Adjustment. An element that may have intensified the reported January jobs loss was the monthly bias factor (birth-death model). Never designed to handle the downside pressures from a recession, the model adds a fairly consistent upside bias to the payroll levels each year, but with a seasonal distribution that has one month of negative bias: January. The downside adjustment to unadjusted January 2009 payrolls was 356,000, versus a subtraction from employment of 378,000 in January 2008. The impact on the adjusted monthly level supposedly is neutered by seasonal adjustments.

Household Survey.  The usually statistically-sounder household survey, which counts the number of people with jobs, as opposed to the payroll survey that counts the number of jobs (including multiple job holders), went through further regular revisions in the latest reporting, in order to adjust for new population estimates. 

Accordingly, with reported December 2008 and January 2009 employment levels not comparable, the BLS did not report the usual month-to-month changes in key data. The BLS, however, estimated that net of the population changes, household employment would have dropped by 832,000 in January, versus a decline of 806,000 in December.

On the new population basis, the official unemployment count was 11,616,000 (it would have been 42,000 worse without the population adjustment). As noted in the opening heading, the proposed $900 billion stimulus package works out to $69,200 per formally unemployed individual, but that is based on the not-seasonally-adjusted unemployment count of 13,009,000. There are 1.4 million unemployed people who are "not unemployed" on a seasonally-adjusted basis, due solely to the illusions of the government’s statistical adjustments. The above unemployed count does not [corrected] include discouraged workers or those considered to be only marginally attached to the workforce.

The January 2009 seasonally-adjusted U.3 unemployment rate showed another statistically-significant increase, to 7.56% +/- 0.23% from 7.19% in December.  Unadjusted U.3 rose to 8.5% in January from 7.1% in December.  The broader January U.6 unemployment rate jumped to an adjusted 13.9% (15.4% unadjusted) from a revised 13.5% (13.5% unadjusted) in December.  

During the Clinton Administration, "discouraged workers" — those had given up looking for a job because there were no jobs to be had — were redefined so as to be counted only if they had been "discouraged" for less than a year. This time qualification defined away the bulk of the discouraged workers. Adding them back into the total unemployed, actual unemployment, as estimated by the SGS-Alternate Unemployment Measure, rose to 18.0% in January, from 17.5% in December.

Employment Environment. The continued broad deterioration in January’s employment environment broadly was in line with deterioration in the better-quality employment-environment indicators: December newspaper help-wanted advertising was at an historic low, with a deepening annual fall-off in January online help-wanted advertising (Conference Board); new claims for unemployment insurance have continued to surge sharply, setting a 28-year high in annual growth; and employment readings continued in the deepest recession territories for both the January manufacturing and nonmanufacturing purchasing managers survey. Since the employment and unemployment indicators tend to be coincident markers of broad economic activity, weaknesses in these series are signaling an ongoing and deepening recession in place.

__________