John Crudele at The New York Post is out with an explosive piece titled: Census ‘faked’ 2012 election jobs report. It alleges that workers conducting the Household Survey (the survey from which the Unemployment Rate is derived) fabricated phony responses at the behest of their supervisors. The article also hints in the title and in the opening paragraphs that this is connected to the jobs reports right before the 2012 election, including the infamous September jobs report, when the unemployment rate fell from 8.1 percent to 7.8 percent, prompting Jack Welch to make his well-known “Chicago Guys” comment.
The issue of data quality is an important one, and the extent to which people are fabricating survey responses is something that needs to be addressed and corrected. But here are a few issues with the report.
1. The meat of the article doesn’t actually say what the title and introduction claim it says. The alleged fakery was not a pre-2012 thing, but rather something that had been an issue for years.
In one document from the probe, Program Coordinator Joal Crosby was ask in 2010, “Why was the suspected … possible data falsification on all (underscored) other survey work for which data falsification was suspected not investigated by the region?”
2. The article also does not say that there was a conspiracy to depress the stated unemployment rate. The guy at the center of the story says there was pressure to produce responses (in order to hit the target for the number of households they surveyed) but that there was no pressure on how to report them:
But, Buckmon says, he was never told how to answer the questions about whether these nonexistent people were employed or not, looking for work, or have given up.
3. While it’s true that at the time the September Jobs Report looked very weird, in retrospect that drop-off in the unemployment rate looks totally on-trend.
4. Also, the article states that the problem was focused on one region.
Census currently has six regions from which surveys are conducted. The New York and Philadelphia regions, I’m told, had been coming up short of the 90 percent.
Philadelphia filled the gap with fake interviews.
One main problem with this, however, is that the unemployment rate in Pennsylvania barely dipped in September. (via Corey Bowers)
As for New York, it also saw a normal-looking decline in September.
So bottom line, there are some issues with the article itself (the meat doesn’t match the explosive headline) and the data also doesn’t quite line up. But we’re still looking into it, and are eager to hear what the Department of Labor and the Census Bureau are saying.