Bureau of Labor Statistics admits data wrong, unemployment rate really 7.7 percent

Bookmark and Share

Throughout much of October, during the partial federal government shutdown, some 700,000 federal workers were furloughed.  Yet, according to data compiled by the Bureau of Labor Statistics, even though they were not working, they were not unemployed, either.

Sure enough, the government's data shows that the amount of people working in October did decrease by 735,000.

But the amount of people unemployed only increased by 17,000, creating a 718,000 person disparity.

Nor were they excluded from the Bureau's establishment survey of employers, according to agency's the press release: "Federal employees on furlough during the partial government shutdown were still considered employed in the payroll survey because they worked or received pay for the pay period that included the 12th of the month."

So, if the furloughed federal employees were not working, yet were not unemployed, where were they? Apparently, not even in the labor force. According to the Bureau's data, the civilian labor force "shrank" by 720,000 in October.

Curious. Particularly, if one considers that had they been included in the survey, the unemployment would have jumped to a whopping 7.7 percent, instead of the reported 7.3 percent.

And, according to the Bureau's press release, they should have been counted as unemployed, but were not: "Workers who indicate that they were not working during the entire survey reference week and expected to be recalled to their jobs should be classified in the household survey as unemployed on temporary layoff. In October 2013, there was an increase in the number of federal workers who were classified as unemployed on temporary layoff. However, there also was an increase in the number of federal workers who were classified as employed but absent from work."

Significantly, the release says, "BLS analysis of the data indicates that this group included federal workers affected by the shutdown who also should have been classified as unemployed on temporary layoff."

Read that again. Every single one of the 700,000 federal workers furloughed "should have been classified as unemployed on temporary layoff [emphasis added]" by the Bureau's own admission.

So, why weren't they counted as unemployed? According to the Bureau, "Such a misclassification is an example of nonsampling error and can occur when respondents misunderstand questions or interviewers record answers incorrectly. According to usual practice, the data from the household survey are accepted as recorded. To maintain data integrity, no ad hoc actions are taken to reassign survey responses."    
However, the only options the Bureau offers here are that either the furloughed workers were laid off, in which case they should have been listed as unemployed, and in fact appeared that way in the survey. Or they said they were working yet absent, in which case they should have been listed as employed.

None of the above should have knocked these employees out of the labor force, yet somehow nearly the entire 700,000 federal workers were in fact removed from the data set. Curiously, the Bureau offers no explanation as to why. If this was a non-sampling error, it seems to have comprised more than 95 percent of all the federal workers surveyed.

Putting it mildly, that's quite an error. In this case, the non-sampling error caused the unemployment rate to be underreported by 0.4 percent.

It is therefore reasonable to question whether or not the data actually was manipulated in order to keep the reported unemployment rate down. Reading through fairly the straightforward survey questions that Bureau interviewers ask, it is hard to see how they could have possibly gotten this so wrong. Were they directed to classify the furloughed workers as not in the labor force?

It is a question the House majority might do well to investigate in its oversight capacity.

Robert Romano is the senior editor of Americans for Limited Government.