The more disconnected from reality Bureau of Labor Statistics (BLS) reports become, the more scrutiny they receive. This has now become a monthly ritual for bloggers and economists. An astonishing amount of ink was spilled analyzing the effects of January's cold, snowy weather on the latest numbers, but that wasn't the end of it.
The BLS changed virtually every important number they define—the population, the labor force, those working, those not working, etc. The economy "lost" 483,000 jobs in December once the benchmark revisions spanning the last 5 years were incorporated. I covered this latest fiasco in Government Revisions To "Reality".
If you wanted to find out what the unemployment rate really was, it was easy to do. All you had to do was go over to the Gallup website and look at the current job polling.
Comparing Gallup's unemployment and underemployment rates so far in 2011 with those for the same periods in 2010 provides something of a seasonally adjusted view of Gallup's jobs data. Unemployment and underemployment are now at least one point below the rates of a year ago, reflecting modest improvement over the past year.
Still, Gallup's measures paint a real-time picture of the current job realities on the ground: nearly 1 in 10 Americans in the U.S. workforce are unemployed, nearly one out of five are underemployed, and the nation's overall hiring situation has not improved over the past four to six months.
Gallup's rolling average shows that the unemployment rate was 9.8% at the end of January, not 9.0% as reported by the BLS. And here is where the jobs fiasco lies. The BLS has so obscured and complicated the process by which unemployment is measured, that the monthly numbers, especially during the Great Recession and its aftermath, have become next to meaningless. These complications and obscurations include but are certainly not limited to—
- The use of two surveys, one of businesses (establishment data) and one of households (household data). The former yields the number of jobs gained or lost, while the latter gives us the unemployment rate. Needless to say, the two surveys are at odds with each other.
- The seasonal adjustment, which is applied to the unemployment rate and other unofficial measures
- The Birth/Death (of businesses) model, which is applied to the unseasonally adjusted establishment data. This model has (apparently) added hundreds of thousands of jobs over the last few years.
- The yearly January benchmark revisions, which have caused about 1.3 million jobs to disappear in the last two years. This must have something to do with the Birth/Death model adding jobs in, but who can really say?
By the time the "official" numbers are reported in the press, the BLS has messed around with the data so much that it is virtually impossible to figure out what's really going on. The BLS does not publish its secret sauce. It is those distortions or revisions of the data that Mike Shedlock, James Hamilton, Zerohedge, Calculated Risk, and many, many others analyze to death. Without the actual algorithm which determines the official numbers, I don't see any point in this kind of spotty analysis of short-term results.
The situation has now gotten beyond absurd. I refuse to play this monthly BLS game. It's time to blow off the BLS' monthly numbers. I no longer want my jobs data smoothed, crunched, massaged and otherwise tampered with in any way. I want the data straight, no chaser. It is a matter of confidence and trust—I don't have either. What is the unemployment rate? It stood at 9.8% at the end of January. How do I know this? Gallup polling told me so.
I would say toss the BLS out the window entirely if it weren't for the fact that the historical data they keep has considerable value. But the monthly data is now mostly noise. And given the positive bias of government institutions which I discussed in Government Revisions To "Reality", short-term data (i.e. the last 3 months) is all but useless. Only longer term trends matter in this game. The year-over-year comparison is the mininum meaningful measure of how America is doing jobs-wise. Using that comparison, Gallup found a "modest" improvement in the jobs situation. In short, there is nothing to write home about.
It is very easy to get caught up in the hubbub accompanying the BLS monthly jobs report, especially if the results are particularly outrageous as in this latest report. Dangerous, potentially lethal human qualities—short-term thinking, standard psychological defense mechanisms, delusional optimism, and the like—have been highlighted during the Great Recession and its aftermath. Given two "good" BLS reports in a row, deluded people start talking about "green shoots" and similar nonsense.
In more "normal" times, the human tendency to flee from Reality is less prominent, and not as risky as it is during the hard times we're experiencing now. Throw in the propaganda with which we're bombarded daily, and the situation becomes very dangerous indeed.
Having gotten this off my chest, I will turn to the longer term job trends tomorrow. A recent article by David Stockman demonstrates beyond any reasonable doubt that the situation is alarming and getting worse.
Thank you for the David Stockman article, it certainly is an eye opener.
I'm just wondering when the health care system will implode. Will Obamacare push the date out, as a result of forcibly injecting another trillion dollars, or will it still implode before 2020? Look at the graphs on health care spending, now consider the graphs in David's article. Where is the money coming from? It is quite clear there aren't enough young healthy employed people, paying for "benefits" they aren't personally collecting on, to sustain this level of spending.
Yes, yes, bring on, I mean sustain, the debt bubbles. Yes, your country needs you. Buy Obama Bonds!
Posted by: BJ | 02/09/2011 at 11:04 AM