Today's employment report reinforces two major themes: the economy is contracting, but the contraction is not yet too deep.

The data are better than feared, as yesterday's jobless claims figure caused many to downgrade their assessment of the labor market. Adjusting for the amount of net business formation continues ot be a problem, but even with an adjustment for this glaring fact, the figures are not yet as weak as in past recessions. It's more a slow bleed.

The economy lost jobs for a seventh straight month, shedding 51,000 jobs for a second month and bringing the average size of job losses for the seven months to 66,000. These losses are consistent with economic recession but better than past recessions.

For example, in the first seven months of the 2001 recession, job losses averaged 145,000. In the first seven months of the 1990-1991 recession, job losses averaged 117,000 (keep in mind the fact that the workforce constantly increases, which should result in larger job losses as time goes on).

The increase in the jobless rate, to 5.7% from 5.6%, will not play well on Main Street and will impact consumer confidence as well as business confidence, as many will fixate on this most understandable economic statistic. The increase follows a large half-point leap in May, which was the largest monthly increase since February 1986.

Minor solace will be had by observations about the wage figure, which increased an as-expected 0.3%. The gain is not large enough to elicit thoughts that wage demands are increasing much as a result of the recentacceleration of inflation.

The government's assumption about the amount of net business formation remains curious. The data are represented by statistics computed using the so-called birth/death model.

In July,the Bureau of Labor Statistics added 4,000 jobs to the tally, 1,000 more than last July and 46,000 more than the 5-year average for the July data (January and July historically are two months when the amount of net business formation is negative).

The fact that the government is adding more jobs than in recent years seems odd given the downshifting in growth and in overall job creation. Most absurd is the assumption for net business formation in the construction sector, which resulted in a 1,000 boost in jobs to the sector compared to assumptions for slight losses on average over the past three years.

It is obvious that the construction sector is contracting, yet the government has not yet made this assumption in its estimation of net business formation for the sector.

One needn't be a statistician to realize the BLS is making assumptions regarding net business formation are wrong. Clear evidence of this is in most recent benchmark revision to the jobs data for the 12 months ended March 2007.

The BLS took about 30,000 jobs per month off the tally. Since that time, the disparity between the BLS data and the actual job tally has probably increased, such that the BLS is probably overestimating the monthly job count by at least 50,000 per month, although this would still leave job losses below that of past recessions.

The figures on hours worked were weak, which bodes ill for personal income. The average workweek fell a tenth of an hour to 33.6 hours, matching its lowest ever (the average workweek has been in a secular decline for decades and never went higher than 33.9 hours in the recent economic expansion).

The drop is the income-equivalent of nearly nearly 400,000 jobs, which will have a negative implications for the monthly figure on personal income. Offsetting this in the short run is the infusion of income from income tax rebate checks.

The employment diffusion index, which measures the percentage of industries that added workers, fell to 41.2 from 42.2 in June, 46.4 in May, and 45.6 in April. The current level is the lowest since June 2003. It was as low as 34.3 in October 2001 as well as April 1991.

Tony Crescenzi is the chief bond market strategist at Miller Tabak + Co., LLC, and advises many of the nation's top institutional investors on issues related to the bond market, the economy and other macro-related issues. At the request of the Federal Reserve, Crescenzi is a regular participant in the board's Livingston Survey of economic forecasters. He is also the author of the revised investment classic,

The Money Market

, first published in 1978 by Marcia Stigum, and

The Strategic Bond Investor

. At the time of publication, Crescenzi or Miller Tabak had no positions in the securities mentioned in this column, although holdings can change at any time. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. Crescenzi also is the founder of, a popular Web site covering the bond market and the economy. Crescenzi appreciates your feedback;

click here

to send him an email. has a revenue-sharing relationship with Trader's Library under which it receives a portion of the revenue from purchases by customers directed there from