Wednesday, November 19, 2014

Conspiracy Theories About Jobs Numbers

Every month, the Labor Department releases figures on the current state of the economy, and every month, without fail, malcontents take to the media to say that the numbers are phony. For example, Paul Singer, a hedge fund billionaire and prominent donor to Republican candidates, recently wrote to his investors, “Nobody can predict how long governments can get away with fake growth, fake money, fake jobs, fake financial stability, fake inflation numbers and fake income growth.”

If you have read my books, you know that I rely heavily on data from the U.S. Department of Labor, so it’s obvious that I don’t share Singer’s skepticism. The data analysts at DOL are career civil servants without a stake in enhancing the reputation of whoever happens to be in power. More important, if there were a vast conspiracy to conceal the true state of the economy, surely someone would play the Edward Snowden role and blow the whistle.

It’s true that the unemployment figure does not indicate the actual percentage of people out of work, because it covers only those who are looking for work and ignores those who have given up. But how many times does the public need to be told this? The Labor Department never pretended otherwise. The numbers are what they are.

Last week, a friend of mine whom I have known since seventh grade pointed out an example of the absurdity of the conspiracy theorists. He wrote on Facebook,

You may have heard about the jobs report that came out last week. The mainstream media generally received it as good but not super—248,000 jobs added (200,000 is order-of-magnitude break-even). Now, this is seasonally adjusted, and October is a strong month for jobs, so it was adjusted down, by a lot, in fact—800,000 or so. If the adjustment formula from last year had been used, growth would have been around 100,000 more. So sure enough, the daily stock market letter I look at has rumblings about government fraud to report the low number. The logic is that acknowledging the true extent of expanding employment would have undercut the case for low interest rates.

He goes on to note another competing conspiracy theory, regarding the other number that was reported—the unemployment percentage:

On the other hand, following links from the newsletter, I find another deep thinker saying that the reported unemployment rate is too low because of fraudulent underestimation of the potential work force. The reason for that fraud would be to minimize the appearance of troubles. So is the current employment picture better or worse than the government is telling us?

As my friend observes in conclusion, these market-watchers accuse the government of “two separate conspiracies working at cross-purposes.” Surely the government can’t be lying to make the economy look both better and worse than it really is.

Writing in The New York Times, Floyd Norris acknowledges that sometimes the reported economic indicators are inaccurate, but this is not because of a conspiracy to make the administration in power look better. Rather, it’s because when Labor’s economic analysts adjust figures, they tend to extrapolate from current trends. If the economy reverses direction, this tendency can cause economists to get the numbers wrong. When the economy goes into free-fall, they may overstate performance, but when it bottoms out and starts to improve, they may understate the situation. He gives the example of a 2012 Twitter comment by Jack Welch, the former chief executive of General Electric, who said that the reported decline in unemployment at that time was “unbelievable.” (This was in October of a presidential election year.) In fact, the figures later had to be readjusted to reflect the fact that the emerging economic recovery was actually more rapid than the initial figures indicated. (Incidentally, following Welch’s comment on Twitter, several other messages pointed out Welch’s reputation for manipulating the numbers reported for GE Capital.)

Norris recommends skepticism in situations where there are “rapid changes in any indicator, particularly if other indicators do not show similar changes.” But he cautions that we should “separate reality from ideology.”

1 comment: