JOHN WILLIAMS’ SHADOW GOVERNMENT STATISTICS
 
FLASH UPDATE
 
April 4, 2008
__________

 

Heavily Gimmicked Payroll Contraction Was Much Worse in Reality

Other Key Data Also Confirm Sharply Deteriorating Inflationary Recession

__________

 Please Note: The March SGS Financial-Weighted Dollar and the March SGS-Alternate Unemployment Rate have been updated, and the preliminary SGS-Ongoing M3 estimate will be updated this weekend on the Alternate Data tab at www.shadowstats.com.The Hyperinflation Special Issue should be posted by Monday, April 7th.
–Best wishes to all, John Williams

 

March Payroll Decline Easily Topped 120,000

When a Fed Chairman begins talking recession, a recession is in place. Chairman Bernanke’s comment on Wednesday that the U.S. economy "even could contract slightly" in the first half of 2008 was more reporting than a prognostication. He certainly had an advance idea of the March employment data that now show a decline in average first-quarter 2008 payrolls versus fourth-quarter 2007, where seasonally-adjusted March 2008 payrolls are down at an annualized 0.7% rate from December 2008.

Despite the bad news in the monthly jobs data, the reported numbers still were overly Pollyannaish, thanks to extreme gimmicking. As anticipated, the industrial production benchmark revisions showed considerably weaker economic activity than previously reported, while the purchasing managers survey again showed a deepening economic contraction and surging inflation.

Also on the inflation front, money supply M2 continued to surge in the latest weekly reporting (up at a seasonally-adjusted, annualized 24.3% in the week ended March 24th), with annual growth in March M3 now a fair bet to top 17.2%, up from the 16.9% historic high set in February. The money supply numbers will be updated over the coming weekend on the Alternate Data tab at www.shadowstats.com, after tonight’s data releases.

Jobs Data Should Continue Fueling Recession Forecasts. The reported third consecutive decline in monthly payrolls, as of March, will do much to reinforce recession outlooks, but the data remain severely gimmicked, understating the monthly declines in payroll employment, thanks to the usual statistical shenanigans at the Bureau of Labor Statistics (BLS). Net of gimmicks, the decline in payrolls and the rise in the unemployment rate were statistically significant.

Payroll Survey. The BLS reported a seasonally-adjusted jobs loss of 80,000 (a loss of 147,000 net of revisions) +/- 129,000 for March 2008, following a revised 76,000 (previously 63,000) jobs loss in February. The prior period revisions, however, were stretched further back in time, likely aimed at muting the published revision to February changes. The revised February decline otherwise would have been 130,000. Although officially the March decline was statistically indistinguishable from a gain, the apparent decline net of gimmicks and revisions would be a statistically-significant contraction. Annual growth in total nonfarm payrolls slowed further to a recessionary 0.35% in March, from 0.59% in February. Annual payroll growth is likely to turn negative in the next month or two.

Bias Adjustment. One element boosting the numbers was the monthly bias factor (birth-death model), which was a net addition of 142,000 jobs in March, following a net addition of 135,000 jobs in February. The March add-factor mindlessly spiked construction jobs by 28,000 and financial activities jobs by 6,000, irrespective of anecdotal evidence of serious trouble in those areas  

Seasonal-Factor Gimmicks. Year-to-year growth should be virtually identical in both the seasonally-adjusted and unadjusted series, and applying the unadjusted annual change to the seasonally-adjusted year-ago numbers for February and March suggests that the seasonally-adjusted month-to-month change should have been a contraction of 124,000. This reporting gimmick is made possible by the "recalculation" each month of the monthly seasonal factors. If the process were honest, the suggested differences would go in both directions. Instead, the differences almost always suggest that the seasonal factors are being used to overstate the current month’s relative payroll level, as seen last month and the month before.

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 those of multiple job holders), showed household employment dropped by 24,000 in March against a 255,000 decline in February.

The March 2008 seasonally-adjusted U.3 unemployment rate showed a statistically-significant increase to 5.08% +/- 0.23% from 4.81% in February. Unadjusted, U.3 held at 5.2% in March. The broader U.6 unemployment rate rose to an adjusted 9.1% (9.3% unadjusted) in March, versus 8.9% (9.5% unadjusted) in February. Adjusted for the "discouraged workers" defined away during the Clinton Administration, actual unemployment, as estimated by the SGS-Alternate Unemployment measure, rose to 13.0% in March, up from 12.8% in February.

Employment Environment. The continued and deepening employment decline reported in March, though still far short of reality, at least continued moving in the same direction suggested by some of the better-quality employment-environment indicators, with collapsing February help-wanted advertising, surging new claims for unemployment insurance, and recession-level employment readings, again, for both the March purchasing managers surveys.

Industrial Production Takes a Hit in Revision. With the index of industrial production based on average 2002 production equal to 100.0, February 2008 had been reported at 113.7. As of the benchmark revision published on March 28th, February 2008 now stands at 111.9, some 1.6% lower. The bulk of the downside revision was in 2006, the latter part of which is when the current recession began. Further detail of the revisions and implications for the annual benchmark revisions of GDP, due later this year, will be discussed in the next newsletter.

Purchasing Managers Surveys Show Inflation and Recession. The stock market truly is irrational if it rallies sharply on a minor upswing in a still-negative purchasing managers survey (manufacturing in March was 48.6 versus 48.3 in February). The alternatives are that either silly hype can rally these extremely vulnerable markets, or that some analysts have a compulsion (or real need) to explain all market movements in terms of any published news, regardless of actions or market manipulations by major players and/or government/Fed. Both factors likely are at play.

Meaningful numbers out of the purchasing managers manufacturing survey, in terms of the component diffusion indices (a reading below 50 means contraction, 50 and above expansion): new orders fell to 46.5 in March from 49.1 in February, employment rose to 49.2 from 46.0, while prices paid exploded to 83.5 from 75.5. On the services side, employment held at 46.9, while prices paid rose to 70.8 from 67.9. The markets still largely are ignoring evidence of a rapid pick-up in inflation.

Week Ahead. The February trade deficit report on Thursday (April 10th) should show sharp deterioration, based on underlying fundamentals. With the concept of recession gaining some official credibility, perhaps such reporting will be allowed. Given U.S. dollar vulnerability, however, odds still favor continued understatement of the deficit.

The various economic data will be reported on more fully in the upcoming mid-April newsletter.

__________

Continuing market turmoil, central-bank/government intervention and systemic shocks remain within the general outlook, which is unchanged.

Publication of the Hyperinflation Special Issue should follow by Monday, April 7th.

Flash Updates and Alerts will be posted as needed. The next regular newsletter is targeted for mid-April.