Making Sense of April Payroll Numbers

U.S. nonfarm payrolls were up 274,000. Bulls were crowing all morning but I wonder why. Treasury yields spiked with the 5 yr yield moving from 3.80 to 3.97. That is a pretty huge reaction. Likewise yields on 10-yr treasuries went from 4.16 to 4.28.

A sustained spike in interest rates will not be good for housing, period. Write this down and don't forget it: Housing is the economy, stupid.

Cheerleaders would have you believe this data is good for stocks because it shows that the economy is getting better. Likewise a poor jobs number would have been considered good for stocks because the FED would pause and that would support the economy. Day in and day out, there's never better a time to buy stocks.

Back to the real world.... In our drug ridden economy with US consumers keeping the entire world barely afloat, the only thing holding up this mess is 1)cheap credit 2)loose lending practices 3)housing. All that remains to be seen is whether or not housing dies of simple exhaustion or Greenspan hikes one too many times. Either way, there will be MASSIVE job losses when it happens.

As for today’s job numbers a mere 257k of the 274k jobs we added this month were assumed. Let's take a look:


Birth-Death-Adjustments April 2005

Hmmm 90,000 of the assumed jobs were in Leisure and Hospitality. This category includes food services. Yeah, that is going to support a lot of purchasing power.

How does the BLS go about guessing these numbers? Well it looks at the economy and says well historically we add such and such number of jobs in a "typical recovery" blah blah blah as if this train wreck has anything to do with a typical recovery.

What is interesting about the assumed jobs this month is that the GDP for the 1st quarter of 2004 was +4.1% while the GDP for the 1st quarter of 2005 was +3.1%. Last April we added 225,000 jobs and this April in spite of a lower GDP we assumed a "mere" 257,000 jobs. Thanks to John Succo at Minyanville for pointing that out.

Also interesting is that the BLS Household Survey showed an increase of 598,000 jobs. Does anyone really believe that? What did we do, add another 500,000 people selling stuff on EBAY as their job?

In the Establishment Survey we see that leisure and hospitality gained 58,000 jobs in April, including 35,000 in food services and drinking places. Since June 2002, employment in leisure and hospitality has expanded by 823,000 with four-fifths of the gain occurring in food services. This trend is clear, we are outsourcing manufacturing and adding bartenders, waiters, and greeters at Walmart.

Also note that according to BLS table A-12 (about the only place one has any inkling of what the employment situation might be, alternative measures of unemployment are still at 9%. That number is probably much closer to reality but with all the obvious shenanigans, who really knows for sure what the real rate is.

Finally I would be remiss if I failed to point out the BLS reported that for the 31st month in a row, more than 20% unemployed had been out of work longer than six months. Is that consistent with creating 589,000 jobs?

Here is the complete BLS pdf Report for April. If you can make any reasonable sense out of this bizarre combination of numbers please enlighten me.

BTW it is not technically correct to say 257k of the 274k jobs we added this month were assumed. The reason being the 257k jobs were not seasonally adjusted but the headline numbers are. Here is a Q&A on that from the BLS.

Q: Can I subtract the birth/death adjustment from the seasonally adjusted over-the-month change to determine what it is adding to employment?

A: No. Birth/death factors are a component of the not seasonally adjusted estimate and therefore are not directly comparable to the seasonally adjusted monthly changes. Instead, the birth/death factor should be assessed in the context of its effect on the not seasonally adjusted estimate.

Q: Can BLS provide an estimate of the contribution of the birth/death adjustment to the seasonally adjusted monthly payroll change? Can BLS independently seasonally adjust the net birth/death component?

A: There is not an estimate of the seasonally adjusted contribution of the birth/death model. The sample collected on a monthly basis, the imputation of business births using deaths, and the net birth/death model are all necessary components for obtaining an accurate estimate. The components are not seasonally adjusted separately because they do not have particular economic meaning in and of themselves.

Of course all this does is make an even bigger mess out of the reported figures. Bear in mind that at the turn, their model of how many jobs were "created" will likely be horribly out of whack since the assumptions they are making now are based on the false premise that the economy is headed for the "proverbial soft landing" and the recovery will continue. At some point they will be massively taking back a lot of these "assumed jobs".

In the meantime I will guarantee you this. Today's numbers are nowhere near as strong as the headline numbers indicate, especially in light of a slowing economy and a dramatically slowing GDP. I look for huge downside revisions in July if this keeps up.

Mike Shedlock / Mish
http://globaleconomicanalysis.blogspot.com/