Not only can the nation’s unemployment rate be manipulated, it is easy to do. And it wouldn’t take a team of conspirators, as a federal investigator alleged, to accomplish that feat.
In its report last week, the Commerce Department Inspector General’s office said it found no attempt to manipulate the unemployment statistics during the last presidential election, when the jobless rate suddenly fell sharply.
The IG’s most important point: the jobless rate couldn’t be rigged even if someone wanted to because too many people would have to be involved.
Let me give you the exact statement from the report: “Addressing allegation raised in the media (that would be me, folks) we found no evidence that the national unemployment rate was manipulated by staff in the Philadelphia regional office in the months leading up to the 2012 presidential election.”
“To accomplish this, our analysis concluded that it would have taken 78 Census Bureau field representatives working together, in a coordinated way, to report each and every unemployed person included in the sample as ‘employed’ or ‘not in the labor force’ during Sept. 2012.”
A field rep goes to peoples’ houses and asks them questions. One of them, Julius Buckmon, got caught filling out hundreds of fake reports in the Philadelphia region in 2010 and 2011. That fact was hidden by Census until I started reporting about it last November.
Buckmon told Census back then — and he repeated it to the IG — that one of his bosses told him to cheat and he’d be covered if he got caught. Sources of mine — you would call them whistleblowers — say this sort of cheating is rampant.
But even if cheating wasn’t widespread, it wouldn’t take an army of Julius Buckmons to move the unemployment rate. Not even close.
And if the IG doesn’t understand how the system works and how it can be abused, perhaps he should read the rest of this column.
After field representatives like Buckmon conduct interviews with households, they go into several categories: 201 for completed interviews; 202 for interviews that were never opened for whatever reason; 203 and 204 for partial interviews. There are also Types A, B and C.
Type Bs are when the interviewer finds the house vacant. Type Cs are for vacant lots or demolished houses.
The key category is Type A, which is when nobody is home, or people refuse to be interviewed or the occupant simply can’t be reached. A lazy field rep could also enter Type As simply because she didn’t feel like doing an interview.
Type As are considered bad by higher-ups because they don’t count toward the 90 percent success quota set by the Labor Department. Supervisors are rewarded for the percentage of interviews completed by their workers.
There are hundreds, if not thousands, of these Type A non-interviews every month. One source in the Denver regional office explained it best by saying that on the next-to-last day of the survey period, there could be 800 Type A non-interviews. But suddenly only 100 Type As will be outstanding when the results are turned over to the Labor Department.
Did the field reps suddenly get a second wind and miraculously reach all those Type As? Of course not. Management has access to any interview that isn’t complete. So, without leaving their offices, anyone higher than the field reps can change hundreds of Type As.
In most cases, they just want the interview completed so it counts toward the quota. But if managers at Census desired to effect the unemployment rate all they’d have to do is complete the Type As and make the phantom subjects either employed or “not in the workforce.”
Since Census has been consolidated into just six regions, very few people would need to be involved. In fact, a single manager could change the unemployment rate since each interview in the scientific jobless survey counts for 5,000 households.
Buckmon, by himself, was affecting the results for some 500,000 households.
Is there proof that cheating was occurring? There sure is. I started writing about Buckmon and the rest of these shenanigans in November. Before then, each Census district easily made the 90 percent quota required by the Labor Department.
Last month, only two of the six regions hit 90 percent. It’s reasonable to assume that the shortfall is being caused by a sudden burst of honesty and fear among the falsifiers.
And if nobody was fooling with the numbers in the Philly Census office, then why is Census acting so suspicious?
The government is refusing to turn over e-mails between a supervisor in the Philly office who was questioned (and exonerated) in the IG probe and another supervisor (who was recently transferred) from Chicago. The Post is appealing.
And after two months, I still haven’t received a second batch of 1,600 e-mails between bosses in the Philly office from around the time of the presidential election.
I got a call Tuesday from the Commerce Department’s Freedom of Information Act office saying that I will start getting those 1,600 pages in about three weeks. They are apparently going through them carefully to see what they’d like to release.
How about this idea: Release everything and then maybe I won’t believe there is a sinister cover-up.