Our new co-blogger, Luigi Zingales, in his first post, has done an excellent job of dispelling the conspiracy theory that political operatives in the Obama administration “got to” the professionals in the Bureau of Labor Statistics who gather and report unemployment data. By the way, welcome, Luigi.
One disagreement, though. I do think that Luigi exaggerated somewhat with his rhetorical question: “How many millions of dollars would a newspaper pay to the first employee of the Bureau of Labor Statistics to blow the whistle?” The actual answer would be zero. In a later post I will tell my own experience in the Reagan Labor Department when, in 1982, I approached a Chicago Tribune reporter with a scandal story about unemployment data regarding the governor of Illinois. She wouldn’t do a thing about it. It wasn’t because of her political views. Her views were close to mine. But I couldn’t give her the whole story and she didn’t want to make the effort of digging.
With that taken care of, it’s important to look at the latest employment data themselves.
And rather than do that from scratch, I’ll refer to two excellent posts that dig into the data. The first is Greg Mankiw’s.
The big thing that caught many people by surprise was the divergence between establishment data that measure employment from the employers’ viewpoint and the household data that measure employment from the individuals’ viewpoint. The establishment data show a net increase in nonfarm payroll employment in September by 114,000. The household data show a whopping 873,000 increase in employment.
Which is right? The one thing we can be sure of is that neither is right. These are estimates. Here’s Mankiw’s exposition:
One might expect these two measures of employment to be identical, but that is not the case. Although they are positively correlated, the two measures can diverge, especially over short periods of time. A particularly large divergence occurred in the early 2000s, as the economy recovered from the recession of 2001. From November 2001 to August 2003, the establishment survey showed a decline in employment of 1.0 million, while the household survey showed an increase of 1.4 million. Some commentators said the economy was experiencing a “jobless recovery,” but this description applied only to the establishment data, not to the household data.
Why might these two measures of employment diverge? Part of the explanation is that the surveys measure different things. For example, a person who runs his or her own business is self-employed. The household survey counts that person as working, whereas the establishment survey does not because that person does not show up on any firm’s payroll. As another example, a person who holds two jobs is counted as one employed person in the household survey but is counted twice in the establishment survey because that person would show up on the payroll of two firms.
Another part of the explanation for the divergence is that surveys are imperfect. For example, when new firms start up, it may take some time before those firms are included in the establishment survey. The BLS tries to estimate employment at start-ups, but the model it uses to produce these estimates is one possible source of error. A different problem arises from how the household survey extrapolates employment among the surveyed households to the entire population. If the BLS uses incorrect estimates of the size of the population, these errors will be reflected in its estimates of household employment. One possible source of incorrect population estimates is changes in the rate of immigration, both legal and illegal.
Paul Krugman digs into the data further and finds some plausible grounds for hope about the recovery. He starts by pointing out that instead of measuring unemployment, which, as economists know, doesn’t capture the people who quit looking for work and leave the labor force, it makes sense to measure the employment to population ratio. That has been falling until recently and that looks bad. But, notes Krugman, the population has been changing. In particular, the leading edge of the baby boomers (typically defined as people born between 1946 and 1964) are retiring. So the raw employment to population ratio is not ideal either.
What to do? Here’s what Krugman does:
So here’s an arguably better measure: constant-demography employment, which shows what would have happened to the employment-population ratio if the age structure of the population had stayed constant.
For my calculation, I’ve divided the population into three age groups, 16-24, 25-54, and 55 plus, for which employment-population ratios are available in the BLS databases. (Scroll down and use the one-screen data search). I’ve then taken a weighted average of these ratios, where the weights are the 2007 shares of each group in the civilian noninstitutional population.
Then he shows a graph that has the constant-demography employment to population ratio improving and sums it up:
So there is real if modest improvement over the past year. Also, the September numbers looks not like an aberration but like a return to trend from what looks like noise in the data over the previous couple of months.
There is one element of the household employment data, though, that shows the 873,000 increase to be less good than it appears. Of those new jobs, about two thirds are part-time jobs for what the Bureau of Labor Statistics calls “economic” reasons: because the workers’ hours had been cut back or because they were unable to find a full-time job.
READER COMMENTS
R Richard Schweitzer
Oct 6 2012 at 1:41pm
How does it seem that the Federal Reserve Board analyzed these trends in its recent policy decisions?
MG
Oct 6 2012 at 5:03pm
I am pretty sure that Street economists never forecast “household survey payrolls”, they seem to rely more on the information content of the “establishment survey”. They consider a difference between their forecast and the print of 100,000 (absent one-offs) as a big miss. They reluctantly look at the household survey for its impact on the unemployment rate.
andy weintraub
Oct 6 2012 at 5:18pm
Along with “constant demography employment”, why doesn’t Krugman use a measure of unemployment that assumes a constant labor force circa 2009?
Thucydides
Oct 6 2012 at 9:53pm
The household survey numbers given are seasonally adjusted. The BLS does not give us the unadjusted numbers, nor the amount of the adjustment, which they could easily do. Neither do they tell us the methodology of their adjustments. Further, the adjustment is not the same from year to year, as they now use monthly contemporaneously updated adjustments.
We are dealing with a black box, and no way of comparing one set of numbers to any other.
Is it any wonder there is skepticism as to the validity of the numbers?
Instead of all the commentary about how dependable and honest these bureaucrats are, how about just plain transparency? Absent that, why have any opinion as to the good faith of the BLS practitioners?
John Carney
Oct 8 2012 at 3:56pm
With all due respect to David, who I’m certain knows far more about economic theory and statistical models than I do, I came across a web post that cuts right to the heart of the matter with dazzling common sense. I know no one who has moved up the corporate ladder since November of 2007, but I do know a number of people who are either unemployed or working for significantly less than they were earning before. And I’m sure I’m not the only person whose base pay has gone up by a single-digit percentage, but whose actual income has decreased 25% due to a freeze on overtime. The article was posted back in February, but I think we all know that the underlying economic landscape hasn’t really changed since then. The link feature isn’t working so you’ll have to copy and paste this into your browser:
http://www.infowars.com/10-reasons-why-the-latest-unemployment-numbers-are-no-reason-to-cheer
I had never heard of Alex Jones before coming across this post. He sounds like a confirmed libertarian which I am not. But I do know common sense when I come across it.
Comments are closed.