John Crudele

John Crudele

Business

Labor Dept. acts to hide older data in job surveys

The Labor Department has made it more difficult for the Census Bureau to screw around with the nation’s unemployment report.

More difficult — but not impossible.

In case you are just joining this ongoing drama, the Labor Department pays Census to conduct the monthly Household Survey that produces the national unemployment rate, which despite numerous failings is — inexplicably — still very important to the Federal Reserve and others.

One of the problems with the report is that Census field representatives — the folks who knock on doors to conduct the surveys — and their supervisors have, according to my sources, been shortcutting the interview process.

Rather than collect fresh data each month as they are supposed to do, Census workers have been filling in the blanks with past months’ data. This helps them meet the strict quota of successful interviews set by Labor.

That’s just one of the ways the surveys are falsified.

This month, Labor tried to outsmart the cheaters by denying them access to past interviews. So the only way the questionnaires could be filled out is by actually conducting interviews.

Government workers would have to do their job — now, that’s a first!

So now, the Census workers for the Current Employment Survey will have to interview the same household for four straight months.

Then these households are given an eight-month break. (Other families are on different schedules so there is never a shortage of people that Census is harassing in any given month.)

After the eight-month break, these households are interviewed for five months.

Before Labor made its surprise change, Census field reps and their supervisors had unfettered access to any of the family interviews. So, after the eight-month break, the field rep and the supervisor could look back and see whether someone in the family, say, had a job during the first interview cycle.

Then, if they were inclined to cheat on the latest survey, the Census field reps and/or supervisors could just copy the old responses.

There was little chance that anyone would catch on.

Now the field reps and supervisors can’t see the old results after a family’s eight-month hiatus.

This, of course, is going to make it much harder for Census to meet the 80 percent success rate for responses that Labor requires. But it’ll also get more accurate readings of the employment situation.

With these new restrictions, the Census Bureau is having an even more difficult time completing the surveys.

As of Monday, 6,500 surveys still needed to be completed in Denver, New York City and Philadelphia. Denver alone needed 1,300 surveys to be completed by Tuesday to be ready for the June jobless report to be released a week from Friday.

Of course, the problem is that Census field reps and supervisors may be lazy, but they aren’t stupid.

So they will probably get around the new obstacle by keeping a personal, handwritten diary on the families they have interviewed. Cheat notes — just like we used before computers.

Hey, but at least Labor is trying!

The spokesman for Labor did not return my phone call from Friday. And Labor still hasn’t done the right thing and make a statement that the unemployment data has been corrupted by lax standards at Census.

Census workers were apparently surprised that they no longer had access to all the interview data and they were assured by higher-ups that the change was intentional.

A June 17 email that went out from Joe Quartullo, the survey statistician at Census, for instance, explained that “the interviewing system for labor force was designed in a manner that broke all longitudinal links between MIS-4 and MIS-5 since BLS did not want any of the normal month-to-month longitudinal links to apply.”

Translation: Census workers can no longer see information from Month In Sample 4 (MIS-4) to MIS-5, which takes place after the eight-month break.

Quartullo’s email emphasizes that this was intentional — “by design,” he wrote in bold letters in the email’s first line.

This whole controversy began when a Philadelphia Census worker, Julius Buckmon, was caught falsifying surveys and — most important — his wrongdoing was covered up.

Worse, Buckmon alleged that supervisors told him to cheat.

Other Census sources have also told me that data is falsified all the time. And since Census polls for lots of different government agencies, including the Justice Department, the problem could be bigger than anyone can imagine.

The Commerce Department’s Inspector General completed an investigation in May and found that while Buckmon did falsify data, his accusations against others were unproven.

And it said that a larger conspiracy to rig the unemployment rate was nonsense. Apparently Labor isn’t so sure or it wouldn’t have changed its procedures.


Last week, the Financial Times reported that “central banks around the world, including China’s, have shifted decisively into investing in equities as low interest rates have hit their revenues.”

The paper quoted a study published by the Official Monetary and Financial Institutions Forum, a central bank research group.

So this is another one of those “I told you so” moments.

As I’ve been explaining for years, the stock markets are rigged and it’s being done by central banks.

Is the Fed involved? Probably. That was the plan proposed by a Fed member in 1989 when Washington was worried about a crash.

There is going to be hell to pay when this next central bank-made bubble pops.