John Crudele

John Crudele

Business

Labor Dept.’s P.E.E. distorts jobs numbers

The Labor Department will be P.E.E.’ing all over the May employment report that will be released this Friday.

So, as I correctly predicted would happen last month, the number of new jobs created in May should also come in deceptively strong.

First, let’s define the technical term P.E.E. It is an acronym I made up a few days ago for Performance Enhancing Estimates.

You’ve heard about the effects of performance-enhancing drugs in sports. Well, P.E.E.s aren’t illegal in statistics, and they make economic numbers look stronger than they really are.

But two scholars at the nonpartisan, highly rated and well-thought-of Brookings Institute recently published a report that seems to cast doubt on a lot of the job growth that has sprouted up because of the Labor Department’s persistent P.E.E.’ing.

The guessing is that Friday’s number will show that 235,000 new jobs were created in May. This follows a much stronger-than-expected 288,000 new jobs that were created in April.

To people who don’t understand what’s really going on, the April number brought cheer — and a little concern that the Federal Reserve might tighten interest rates.

The April numbers are part of the reason economists are predicting that the contraction in the gross domestic product in the first quarter was a fluke that will be strongly reversed in the April-through-June period.

The know-nothings don’t understand P.E.E.

Seasonal adjustments, of course, are one Performance Enhancing Estimate — especially since the Labor Department changes its seasonals each month rather than once a year.

But the P.E.E. at work this Friday is one of my favorites — the Birth/Death Model in the Labor Department’s Current Employment Statistics (CES).

In the CES, the Labor Department asks thousands of companies how many new jobs they’ve created in the last month. The department then takes the number and puts P.E.E. all over it.

For instance, those 288,000 new jobs allegedly created in April included 234,000 jobs that the Labor Department believes were created (the Birth part of the model) by companies just coming into existence. But it’s really P.E.E. — Washington can’t prove those jobs are real, and when a correction is made to the number next year, few people will notice or care.

In May 2013, Labor added 210,000 jobs from companies that were alleged to have formed during the month. So Friday’s number should have a similar stream of P.E.E. to last year’s.

So I don’t mislead you, there’s one thing I want to clarify. That 210,000 P.E.E. is added to the job figures before seasonal adjustments. So you can’t just do the math, say for April (288,000 new jobs minus the 234,000 guesstimate), and conclude that there would have been little growth without the B/D job guesses.

I’ve kept the scholars waiting long enough. They are Ian Hathaway and Robert E. Litan.

The two recently published a short report called “Declining Business Dynamism: It’s For Real.” Remember that I said the Labor Department adds hundreds of thousands of jobs because it believes companies are quietly coming into business and creating positions.

That’s what we all want to believe. We want to think that there are a zillion Steve Jobses out there creating the next big company in their garage. But Hathaway and Litan say this is not happening.

In their paper, Hathaway and Litan state that “business deaths now exceed business births for the first time in the 30-year history of our data.” And, the two add, that trend is “not isolated to a few regions. In fact, the data show that it is a pervasive force evident in nearly all corners of the country.”

If more companies are going out of business than are being “born,” why does the Labor Department continue to add hundreds of thousands of jobs a year to its totals? Why does the department’s modeling still think so many phantom jobs are being created when the Brookings people’s numbers show the opposite?

The Brookings data only go up to 2011. So there could have been a magical turnaround in the most recent three years. But with the economy still limping along, I wouldn’t count on that.

And remember also that a company that has just been “born” could be nothing more than a guy who was laid off and then set up shop in his extra bedroom. In business — as in life — not everything that’s born thrives.

I hate to pee all over the government’s Performance Enhancing Estimates, but good data help government officials make correct decisions. And bad data make people look foolish in their predictions and government officials look clueless.