Behind the Phony Jobs Numbers
Behind the Phony Jobs Numbers
By Arnold Ahlert
As most Americans are probably aware, president Barack Obama's re-election chances are largely dependent on the success of the economy between now and next November. Thus, last week's Bureau of Labor Statistics (BLS) report that the unemployment rate had dropped to 8.3 percent, and total non-farm payroll employment rose by 243,000 was great news for the administration. Yet to put it as charitably as possible, the BLS has done a rather remarkable job of "massaging" the numbers. In short, any unemployment rate can be lowered when the number of people counted as unemployed is dramatically reduced.
First, it should be noted that all BLS figure are estimates, extrapolated from a survey of only .05% of households. Then there is the BLS interpretation of the word "unemployed," itself. The agency defines it as people who "do not have a job, have actively looked for work in the prior 4 weeks, and are currently available for work." If a jobless American does not meet these particular criteria? The BLS contends that he or she is "not in the labor force."
This is a critical distinction because the unemployment rate is calculated only as a percentage of those who are in the labor force. If fewer Americans are counted as part of that labor force, the unemployment rate goes down, even as the "non-participants" remain jobless. In October 1988, the labor force participation rate hit 66 percent. It remained there through 1992, briefly topping 67 percent during the dot.com boom years of the Clinton administration, and stayed above 66 percent for almost all of George W. Bush's eight years.
Between Barack Obama's election and his first year in office, the rate dropped from 65.8 percent to 64.6 percent. At the end of January, when the BLS was conjuring up its latest jobless number, the labor participation rate dropped to 63.7 percent. That number represents a 30-year low. The raw numbers, however, tell the real story. Since 2008, approximately 8.8 million more people have been classified as "not in the labor force," including 2.6 million more than one year ago.
There is an ideological argument occurring as to what these numbers represent, especially with respect to the month of January. Those on the right contend that, based on the statistics, 1.2 million people dropped out of the labor force in a single month. Those on the left contend that the numbers represent an increase in the overall population reflected in the 2010 Census, and that years of changes in the Household Survey to measures that have taken place during that time period are incorporated into one month for simplicity's sake. Who's right? Time magazine trumpets the "good" news, citing BLS employee Mary Bowler who said that "the steep jump in the number of those not seeking work came entirely from the census adjustment, which added 1.25 million people to that group. If you take out the census adjustment, the labor force numbers stayed essentially the same, as reflected by the labor force participation rate of 63.7%."
The increase comes entirely from the census adjustment? Not a single one of those extra 1.25 million people "found" by the Census Bureau needs a job? Note the one figure that is not in dispute by either side: the 63.7 percent of people in the labor force. which once again begets the other undisputed figure, namely the 30-year low in workforce participation. Yet Ms. Bowler remains unperturbed, contending "the spike in the number of people no longer looking for work is entirely the result of some people at the Labor Department adding numbers to their spread sheets rather than an actual observed shift anywhere in the real economy."
Inadvertently, Ms. Bowler has put her finger on a serious problem with BLS calculations, namely they are in fact numbers added to spread sheets rather than actual observed shifts in the economy–which brings us to the figure of 243,000 jobs the BLS contends were created in the month of January. That number was based on something the BLS calls "seasonal adjustments." The BLS notes that "seasonal fluctuations in the number of employed and unemployed persons reflect not only the normal seasonal weather patterns that tend to be repeated year after year, but also the hiring (and layoff) patterns that accompany regular events such as the winter holiday season and the summer vacation season." Thus, the numbers are "adjusted" to smooth out the various up-and-downs in employment that occur throughout the entire year.
How are they calculated? The BLS "uses the past history of the series to identify the seasonal movements and to calculate the size and direction of these movements. A seasonal adjustment factor is then developed and applied to the estimates to eliminate the effects of regular seasonal fluctuations on the data." How large are these seasonal adjustment factors? The same BLS that estimated 243,000 jobs were gained in January also reported that nearly 2.7 million jobs were lost in January. The massive difference between the two figures? Seasonal adjustments.
The New York Post's John Crudele explains that while seasonal adjustments "are intended to smooth out holiday bumps" the "unusual nature of the Great Recession" skews the numbers. "Let's say there are rumors in your company that 300 people are going to be laid off," writes Crudele. "Instead, management decides to fire just 200. Two hundred people, of course, have lost their jobs. But, adjusting it for expectations, 100 people didn't get fired. Using this analogy, the government would say that, on an expectation-adjusted basis, 100 jobs were created." He also asks and answers the bottom-line question. "Does that mean there really are 243,000 new jobs out there? Absolutely not."
Charles Biderman, President & CEO of TrimTabs Investment Research, is even more skeptical. "Actual jobs outstanding, not seasonally adjusted, are down 2.9 million over the past two months," he contends. "It is only after seasonal adjustments–made at the sole discretion of the Bureau of Labor Statistics economists that 2.9 million less jobs gets translated into 446,000 new seasonally adjusted jobs for January and December." Yet he is far more critical of the BLS's use of past history to calculate their numbers. "BLS historic data is changed almost every month until the income tax returns for each year are available three years in arrears. In other words, the BLS currently has accurate data for 2008 and before."
Steven Leslie, lead analyst for financial services at the Economist Intelligence Unit, amplifies this argument, noting that BLS adjustments could be "based on patterns in the past [that] may no longer be correct," and that "BLS have this time made additional adjustment for recently received population estimates based on the 2010 census figures." Does this amount to some sort of conspiracy theory as some on the right are contending? "[T]here is no reason to doubt the BLS statisticians who are civil servants following a complex, but transparent approach to calculating and publishing the numbers," he adds. Complex, no doubt. Transparent? "No one I know has any idea as to how the BLS does this seasonal adjustment," Charles Biderman counters.
Again, one's ideology will color one's thinking in that regard, but there is far less to be understood looking at one month of jobs data than there is looking at the overall economic picture. If one finds the BLS data genuinely optimistic, data from the Congressional Budget Office is a sobering counter-weight. Their report reveals that the national deficit will top $1 trillion for the fourth year in a row, servicing that debt will soon cost as much as paying for all of Medicaid's expenses, economic growth will be 2 percent in 2012, falling to 1.1 percent in 2013, and unemployment could rise to 8.9 percent by this fall, and jump back up to 9.2 percent by the end of next year.
If one wishes to see only the positive aspects of an 8.3 percent unemployment rate–absent workforce participation–there are no arguments that will persuade otherwise. For the rest, a numerical reality: in January CBO projections expected workforce participation to be 65.3 percent. It was 64 percent. One month later it's 63.7 percent, 1.6 percent points below the original estimate. And even if we grant the BLS the benefit of the doubt and concede that 243,000 jobs were created, the difference in the labor participation rates is the equivalent of almost three million "missing" workers. Maybe most, or even all of them don't really want a job.
But I wouldn't bet the 2012 election on it.