Cross posted at Christian Science Monitor and the Mises Blog:
The people over at the Calculated Risk blog had a helpful post today that links to the BLS's primer on the differences between the "Establishment" employment data and the "Household" employment data.
Economist John Williams has often been referenced on these issues, and his site, Shadowstats.com is indeed helpful. However, if you want something straight from the BLS's mouth, the primer is very useful.
This will help you understand a bit more about the establishment survey's somewhat infamous "birth/death" model that assumes new businesses (and jobs) are created every time an existing business goes out of business. This sort of modeling, of course, makes employment look better than it is in the current environment.
Calculated Risk also credits a recent blog post by Nancy Folbre that looks at the futility of examining the unemployment rate as a measure of real-world job market health.
While one should always view government employment data with a critical eye, one need not debate the sampling models to see immediately that using the unemployment rate to discuss the job markets rather misses the point.
The unemployment rate is a function of both total employment and the labor force. Indeed, in a sense, it's just the delta between the total size of the labor force and the total number of jobs available.
But how big is the labor force? The labor force size depends on whether a person is actively looking for work or not. So, if a discouraged worker, or say, a recent college grad who hasn't even bothered to look for work given the economic conditions, reveals his or her lack of job seeking, that person will not be counted as a member of the labor force.
So, when we subtract all the older "retired" people who'd rather not be retired, the young people who are ecstatic with an unpaid internship or worse, and all the discouraged workers, we end up with a big decline in the labor force, and this is why labor force participation is at historic lows. As long as that's the case, that delta between labor force size and total employment will be smaller than if those people thought they could find actual full time employment, looked for work, and were counted as members of the labor force.
Folbre's point is that we should look to total employment for a real gut check on how the economy is doing, and not to unemployment. Taking Folbre's advice, we see that there aren't new jobs being created, and all the while high schools and colleges are graduating more and more people every December and every May, and as wages fall, spouses and non-wage earners in single-wage households may have to go out and look for work.
All of this in the face of extremely poor job creation.