The Deceptive Unemployment Rate

I have long viewed the Unemployment Rate as an awfully fuzzy concept, and that because “unemployment” as such is measured by the Current Population Survey rather than some more appropriate physical count. Yesterday’s BLS announcements were greeted with pleasure because the unemployment rate dropped from 8.5 percent in December to 8.3 percent in January. Here is a tabulation worth pondering:

(Data in 1,000s or %)
	Dec 2011	Jan 2012	% Change
Civilian Work Force	153,887	154,395	0.3
Unemployed	13,097	12,758	-2.6
% Unemployed	8.5	8.3	-2.9
Not in Labor Force	87,212	88,784	1.8
Not in Force but want to work	6,135	6,495	5.9
If Want to Work is added:
Civilian Work Force	160,022	160,890	0.5
Actual Unemployed	19,232	19,253	0.1
% Actual Unemployed	12.02	11.97	-0.2

The civilian work force is defined as those actually employed and the “unemployed.” The unemployed are those actively seeking jobs—as reported to the CPS, and the survey results are then extrapolated to the nation. Active job-seeking means that you have applied for a job in the last four weeks. The same survey, however, also tracks other categories. One of these is called Not in the Work Force. A subset of that is a rather large number of people who “Want to Work” but do not fit the “unemployed” category because they have not been out beating on doors, keyboards, or telephones to get a job last month. That number, 6.5 million in January, is suspiciously close to the 5.6 million jobs lost in the Great Recession but not yet regained.

Now consider the December to January changes. The Work Force grew by 0.3 percent, the number of “unemployed” decreased by 2.6 percent, and the unemployment rate dropped 2.9 percent. But those defined as outside the workforce but still “wanting to work” grew 5.9 percent in that month.

Next let’s restate those data and include those who want to work both under the Civilian Work Force and the Unemployed. That produces a truer picture. It would show that the number of unemployed actually grew (0.1%) but the unemployment rate went down marginally, by 0.2 percent.

I am still be displeased by this number because it is based on a survey. Instead I prefer simply to look at total jobs over time—and the total population aged 16 and over. That would at least be counted properly every 10 years.