How Unemployment Is Figured

During the third week in February, a pencil-wielding army of some 1,000 pollsters from the Bureau of the Census descended upon 50,000 households across the U.S. Their mission: to find out how many Americans were workIng and not working from Feb. 9 to Feb. 15. Information from the poll, called the Current Population Survey, was sent to Washington, fed into Census Bureau computers in Suitland, Md., and then turned over to the Labor Department's Bureau of Labor Statistics. The end result of this complex process was last week's announcement by the BLS that hi February the number of unemployed stood at 7.5 million, or 8.2% of a total labor force of 91.5 million.

The Government has been measuring U.S. employment in this way since 1940. Today the BLS spends some $5 million a year on its monthly surveys. The results that the bureau publishes are only extrapolations from polls, but the polls are remarkably thorough.

BLS statisticians are persuaded that there is only one chance in ten that their overall jobless figures can be off by more than 150,000. The current population survey is far broader than commercial polls that measure, say, public opinion or television audiences. The sample covers every state and the District of Columbia, and is designed to reflect urban, rural and industrial areas in proportion to their presence in the nation as a whole. For polling purposes, the survey is broken down into districts containing about 300 households. These, in turn, are divided into clusters of about four dwelling units each. One-fourth of the 50,000 households in the sample are replaced each month so that no single household is visited in more than four consecutive surveys.

The Census Bureau's trained pollsters all work from a standard questionnaire that is meticulously worded to avoid bias. Respondents are never asked directly if they consider themselves unemployed. That is determined by the statisticians and their computers. The survey rules out all of the 59 million Americans who are under the age of 16 as well as those who are confined to mental or penal institutions. The active labor force is considered the sum of all those who fit the BLS definitions of the employed and unemployed:

> Who is employed? Anyone 16 or over who did any work at all for pay or profit during the week before the survey is regarded as a jobholder; there is no upper age limit. Included in the employed are part-time workers and temporaries. Even some employees who are not paid are counted: a son or daughter who helps out for at least 15 hours a week in a family-operated enterprise--a mom-and-pop grocery, for example--is counted in the active labor force.

> Who is unemployed? Anyone 16 or over who actively sought work in the 30 days preceding the survey and is currently available for work is counted as unemployed. Seeking work means registering with a public or private employment agency, meeting with potential employers, writing letters of application, answering help-wanted ads. Two groups--those waiting to start a new job within 30 days and those on layoff waiting to be recalled to an existing job--do not have to be looking for work to be counted as unemployed.

When the unemployment figures come under fire--as they characteristically do when they are high--they are accused of both exaggerating and understating the "real" extent of joblessness. The overall unemployment rate is Indisputably swollen somewhat by the inclusion in the jobless totals of housewives, students and others who may be only marginally dependent upon regular paychecks. It can be argued, for in stance, that the jobless rates among heads of households (5.4% v. 3.0% a year ago) or adult males (6.2% v. 3.5%) are far better barometers of economic distress than unemployment among teenagers (19.9% v. 15.3%). But the BLS defends including marginal members of the work force in the overall figures on the grounds that they are counted as employed when they hold jobs and thus, for the sake of consistency, should be regarded as unemployed when they do not.

In some ways the figures understate unemployment. The jobless totals do not include the so-called hidden unemployed. They are Americans--most of them women and elderly men--who want jobs but have given up trying to find them. These "discouraged workers" are considered not unemployed but to have left the labor force altogether. As of the last quarter of 1974, the number of such workers stood at 796,000, but many more have joined that category since then. Most of some 580,000 Americans who left the labor force in February were discouraged workers.

Similarly, the published statistics do not reflect the fact that many of those who are counted as jobholders are only partially employed. According to the BLS, there are 3.7 million Americans who are working short hours or hold part-time jobs because they cannot find full-tune work. One little-noted BLS measure called "labor-force time lost" combines partial employment with unemployment on a man-hour basis to show the recession in terms of missing production time rather than missing jobs. The current rate: 8.9%.

Obviously, what should be counted In arriving at the overall unemployment figure is debatable. But the mam value of the monthly figures is in what they say about the direction in which unemployment is going. What those figures showed last week was that the recent sharp rise in the unemployment rate had come to a merciful, if temporary pause --but at a record high level.

This file is automatically generated by a robot program, so viewer discretion is required.