Bad feelings grow about employment report

John Crudele

TAKING STOCK

When the Labor Department released its employment figures for January 2009, it subtracted an astounding 356,000 jobs from the overall count because it felt that small companies were quietly going out of business.

It couldn't prove this was really happening.

But the department's computers made that adjustment anyway -- and much more optimistic ones over the 11 other months of 2009 -- because something called the birth/death model instructed them to do so.

The model is a guess -- and a bad one, it turns out -- at how many companies too small to be reached in the official monthly survey are creating jobs (the "birth" part) or eliminating them ("death").

Prepare yourself. It's déjà vu time.

Just like it did last January, the Labor Department will be making pessimistic assumptions in tomorrow's employment report.

Worse, it will also be correcting some of the errors this adjustment made in the past, resulting in an even gloomier job picture (if you can believe it) for the last year.

The Obama administration isn't supposed to have the January figures yet. So it must just be an incredible coincidence that the president began his full-court press for a jobs program last week.

The 356,000 downward adjustment for the birth/death model contributed to Jan. 2009's overall employment figure being frighteningly bad -- the loss of 741,000 jobs for that month.

And that started a string of monthly job losses -- 681,000 in February, 652,000 in March, 519,000 in April, etc. -- that turned 2009 into one of the most dreadful labor markets of all time.

I warned readers the day before last January's jobs report that the figures were likely to be worse than anyone expected. That column was headlined: Why You Will See Huge Drop in January Jobs (2/5/09).

Consider yourself warned again.

Just a week ago, the experts thought there would be an increase of 50,000 jobs in January. As of yesterday, that optimism had been pared down to a miniscule rise of just 13,000 jobs.

Either would be a miracle considering the statistical hoops that the Labor Department puts its numbers through. The experts expected job growth in December too, but the government reported a loss of 85,000 jobs.

For the record, an outfit called TrimTabs yesterday said its independent calculations showed that the economy lost 104,000 jobs in January. It bases that estimate on daily income tax deposits at the US Treasury.

Another job tabulator that tracks payrolls, Automatic Data Processing (ADP), thinks the January job loss was 22,000. That's the smallest monthly decline in two years.

Neither TrimTabs nor ADP is trying to guess the Labor Department's number. Both, instead, track company payrolls or tax receipts and forego all the guesstimates, like seasonal adjustments and the birth/death models, used by the Labor Department.

A few months ago the Labor Department looked as if it was going to correct the mistakes caused by its birth/death model.

But insaner heads prevailed, and it now seems as if no corrective measures are being taken.

So, expect the DoL to reduce the overall job count, before seasonal adjustments, by the same magnitude as last year's 356,000. In short, Labor thinks 356,000 jobs were cut by small companies that it couldn't reach by phone because -- well, naturally -- their phones were cut off.

And expect the DoL to also report a so-called "benchmark revision" to its figures that show the 2009 economy had roughly 820,000 to 840,000 fewer jobs than were originally thought.

Bottom line: I hope I'm wrong, but tomorrow's number has a good chance of being disappointing to those who think job totals will increase.

I saved the good news for last.

I'm going to make the same prediction that I (accurately) made last year.

While I don't know if companies will actually be hiring in the months ahead, the stats should start looking better. That will be caused both by the full Census Bureau hiring and an optimistic turn in the birth/death model assumptions that always occurs in spring.

On that latter point, the Labor Department's computers figure lots of small companies suddenly appear, like crocuses.

*

Here's one more matter that will be of interest to everyone.

Will the unemployment rate stay at 10 percent, as the experts are predicting?

As the recession gets longer, more and more unemployed people get disqualified from being included in the U-3 calculation -- through a separate telephone survey -- that produces that 10 per cent figure. That's because they stop looking for work.

So, the unemployment rate could actually im prove even as the job market gets worse.

The stat to watch is the U- 6 figure that mea sures underemployed peo ple, which last month was at 17.3 percent. That's where people formerly on the U-3 jobless list end up before they finally quit looking for work and disappear altogether. john.crudele@nypost.com