The New Yorker: Οι χειρότεροι δείκτες απασχόλησης των τελευταίων ετών στις ΗΠΑ, 9,7% η πραγματικη ανεργια αντι 7,4%..

The official jobs figures for August have focussed attention on an issue that labor-market experts have been puzzling over for years: the decline in the proportion of Americans who are working or actively looking for work.

Last month, the labor force participation rate dropped to 63.2 per cent, its lowest level in thirty-five years. That’s pretty remarkable, especially for an economy that is supposedly in its fifth year of recovery. (According to the National Bureau of Economic Research, the Great Recession, ended inthe spring of 2009.)

In the usual scheme of things, many people drop out of the labor force during recessions because jobs are scarce and they lose hope of finding employment. But once the economy picks up, they start sending out résumés, and the participation rate gradually picks up. (For the Labor Department to count someone as being in the labor force, he or she has to have been working or actively looking for work during the month prior to the date of the government’s monthly employment survey.) This recovery is different. In August, 2008, just before Lehman Brothers blew up, the participation rate was 66.1 per cent. Five years later, it’s still almost three percentage points lower than it was then.

Assuming the participation rate stayed constant over the past five years, the unemployment rate would now be considerably higher than the official 7.3 per cent. How much higher? A bit of grade-school arithmetic provides the answer. In August, 2008, the participation rate was 66.1 per cent. Applying that figure to a working-age population that has grown by about ten million in the past five years, there would be about 162.6 million people in the labor force, rather than the actual figure of 155.5 million. With 144.2 million Americans currently employed, 18.4 million would then be classified as unemployed. (The actual figure is 11.3 million.) And the unemployment rate would be roughly 11.3 per cent—yes, well into double digits.

To be sure, this is just a mechanical calculation that ignores quite a few things. If more people were looking for work—i.e. the participation rate was higher—the number of people in employment might also be higher, and thus the unemployment rate would be lower. How come? It’s a supply-side argument. With more people searching for jobs, some of them might be willing to take low-wage or part-time openings that are currently unfilled. And the enhanced competition would bring down wages, which would encourage firms to do more hiring.

These things are possible, but I doubt the effects would be large enough to make much of a difference to the unemployment rate. The job-vacancy figures don’t suggest that large numbers of low-wage and part-time jobs are going unfilled. And the level of over-all demand has a much bigger impact on month-to-month variations in hiring than modest changes in wages do.

A stronger argument is that the decline in the participation rate isn’t completely cyclical, and that it’s unrealistic to expect it to rebound to the levels of 2007 or 2008. The structure of the American work force is changing. Many baby boomers are approaching retirement age, or have already passed it. The rise in the number of women going out to work, which had a big impact in the period from 1960 to 2000, has levelled out. On the basis of these two factors alone, you would expect a fall in the proportion of the population in the work force, and that’s what we’ve seen. As the accompanying graph from the Bureau of Labor Statistics shows, the participation rate peaked in early 2000, at 67.3 per cent, and since then, generally, it’s been falling. At the end of 2005, for example, it stood at sixty-six per cent.

Labor Force Participation Rate, 2000-2013. Credit: Bureau of Labor Statistics Data.

But demographic factors aren’t the entire story. They aren’t even the main story. In the past five years, the participation rate hasn’t just edged down; it’s fallen precipitously. Most analysts agree that this was primarily a result of a soft economy, which led many people to stop looking for work and drop out of the official labor force. Last year, Heidi Shierholz, an economist with the Economic Policy Institute, published a paper in which she concluded that about two-thirds of the fall in the participation rate was due to a weak job market, and about a third was due to structural factors. A few months ago, the economics team at Goldman Sachs examined the figures and decided that structural changes account for less than half of the decline.

There are a few dissidents who question this conclusion, but it seems correct to me. Assuming it is, we can do another quick counterfactual calculation. For the sake of argument, let’s assume that in the past five years an aging population and other structural factors have caused the participation rate to decline by 1.2 percentage points, from 66.1 per cent to 64.9 per cent. (In picking this figure, I am assuming that about forty per cent of the over-all decline of 2.9 percentage points was structural, which jibes with the findings from the E.P.I. and Goldman.) If this is right, the rest of the fall in the participation rate—another 1.7 points—was a product of people losing heart and giving up on their job searches.

Now, let’s figure out what the rate of unemployment would be if the Labor Department could locate all these discouraged workers and categorize them as members of the labor force. If this was done, the labor force would be 159.7 million and the number of people counted as unemployed would be 15.5 million. Hence, the unemployment rate would be 9.7 per cent.

Again, this is only an illustrative calculation. You can argue with the assumptions I made and jiggle the numbers in lots of other ways. (For example, you could argue that August, 2008, marked the end of a bubble and that a labor force participation rate of 66.1 per cent wasn’t sustainable. If you bring it down a percentage point, say, and use that as your benchmark before adjusting for structural changes, you will end up with a “real” unemployment rate of about nine per cent.) But unless you believe that the fall in the participation rate is almost entirely a structural phenomenon, you can’t argue with the proposition that the labor market is considerably weaker than it appears to be, and the official unemployment rate is sending out some very misleading signals.

http://www.newyorker.com/online/blogs/johncassidy/2013/09/why-the-real-unemployment-rate-is-nine-or-ten-per-cent.html