Commentary: It Happens Every Month

The March employment situation ""report"":http://dsnews.comarticles/economy-adds-88k-jobs-in-march-unemployment-rate-down-to-76-2013-04-05 was another in a series of labor reports that had analysts scratching their heads--on the left or right side, depending on their politics.

[IMAGE]

Just as there is no Democratic or Republican way to collect garbage (okay, there might be depending on how much government you want), there should be no Democratic or Republican economic data. The numbers are what they are, not what your political lens tells you they are.

That said, when data such as the March report are released--weak job growth, yet a drop in the unemployment rate--conspiracy theorists emerge from the woodwork. Imagine if this report had been issued just before Election Day. Oh, wait: The report for last September showed the economy added a weak 114,000 jobs (later revised to show a gain of 138,000 jobs), while the unemployment rate dropped 0.3 percentage points to 7.8 percent, below 8.0 percent for the first time since January 2009 and conveniently just five weeks before Election Day.

One month later, the Bureau of Labor Statistics reported the unemployment rate ticked up to 7.9 percent--still remaining under 8.0 percent--while the economy added 171,000 jobs (subsequently revised down to a gain of 160,000 jobs).

Curiously, while the report for September caused a firestorm, there was no similar reaction a month later, perhaps because in the interim, critics learned how the employment situation is compiled.

The monthly employment report is actually the result of two surveys: the Current Population Survey (CPS) and the Current Employment Statistics (CES). The CPS is often referred to as the household survey and the CES the establishment survey.

While they overlap--each is conducted for the week including the 12th calendar day of the month--they use different yardsticks, hence the apparent contradiction with one result. In the case of March, the unemployment rate appeared strong, while the other was weak.

Indeed, they have very little to do with each other.

While it's more apparent in local area data, the CES tracks where jobs are located, and the CPS is a measure of where employees live. (In an area such as New York City, the unemployment rate of suburban counties on Long Island or in New Jersey, for example, would be affected by jobs in New York City.)

The unemployment rate is a defined term, which is why it is so difficult to reconcile it with movement in payroll jobs.

The unemployment rate is the calculated by dividing the number of unemployed by the labor force, which is the sum of unemployed plus employed. To be considered unemployed, an individual has to be out of work, available for work, and looking for work. In a society with 11 individuals, if nine are employed, one is unemployed (as defined), and one isn't bothering to even look for work, the unemployment rate would be one divided by 10 or 10 percent. If the individual who is not considered [COLUMN_BREAK]

unemployed decides to start looking, the numerator and denominator of the equation would each increase by one, and the unemployment rate would be two divided by 11: 18 percent. This will often occur following months of strong job creation when individuals who aren't looking have renewed confidence of their ability to find work.

That's exactly the opposite of what happened in March: The total labor force dropped by 496,000 as employment fell 206,000 and unemployment dropped 290,000. The reduction in both the numerator and denominator of unemployment rate calculation caused the rate to drop to 7.6 percent from 7.7 percent.

If unemployment had not dropped--if confidence in getting a new job remained constant--the unemployment rate would have increased to 7.8 percent in March instead of falling.

There are a variety of qualifiers to the employment situation report (a 38-page report with 30 pages of small type data), including factors often overlooked by analysts such as the number of ""multiple jobholders"" each month. As that number fluctuates, it can either exaggerate or mute the change in payroll jobs. In March, the number of multiple job holders dropped, which means that even a relatively weak number of ""new jobs"" would have reduced unemployment and thus the unemployment rate.

An individual who holds two jobs but loses one of them would have no impact on the unemployment rate, but the number of payroll jobs would be affected.

Another qualifier is self-employment. In March, self-employment also declined, another factor that would have been masked in the household employment data and supported a reduction in the unemployment rate.

Self-employment is even trickier to capture in the employment report since the CES draws from the universe of companies/employers who file unemployment insurance tax returns, something a self-employed individual would not do. In those months in which self-employment increases, ""employment"" as tracked in the CPS would increase with no corresponding increase in payroll jobs. The reverse, though, is not necessarily true in months when self-employment declines since it is more the likely the result of a self-employed individual finding more traditional employment.

It's not a right or left issue. It's a matter of reading all the lines, not just between them.

-------------------------------------------------
Housing data will be scarce in the upcoming week, but there will be other key reports, including Tuesday's often overlooked Job Openings and Labor Turnover Survey, which--on a one month lag--tracks the ins and outs of the Employment Situation report. One key element matches industry job openings against industry unemployment as an indicator of whether certain sectors (notably construction) are expanding.

The National Federation of Independent Businesses reports its monthly optimism index Tuesday as well, another underlying indicator of employment since small business is an important job creator.

Friday, the Commerce Department will report on retail sales, which could be an interesting number against the data in the employment report, which showed the number of retail jobs contracted 24,000 in March, about two-thirds of the expansion of January and February combined.

_Hear Mark Lieberman on P.O.T.U.S. (Sirius 124) on Friday at 6:20 a.m. and again at 9:20 a.m. Eastern time._

*_Want to write an opinion piece for publication on our site? Send your submission to_* ""[email protected]"":mailto:[email protected].