Counting the Jobless

Measuring unemployment is not as simple as it sounds

Few economic indicators are as politically charged as unemployment rates. Britain’s Conservative government is often accused of having changed its definitions of unemployment to reduce the published rate. Unemployment rates are also used to judge the tightness of labour markets: declining rates signal a tight market for labour and hence, eventually, pressure for higher wages. And they are used to assess the degree of social pressure that is caused by joblessness. To many politicians, particularly in continental Europe, persistently high unemployment rates are a constant reminder of the failure of their economic policies. Yet official measures of unemployment, based on surveys of the labour market, may not be as useful as many think.

To qualify as unemployed in industrialized countries, a person must be without a job, taking active measures to find one and available to start work (usually immediately). Some maintain that this definition of unemployment is too loose. They point out that some of those counted as unemployed by official number crunchers may actually be working in the black economy. So high official unemployment rates may not necessarily mean that there is little upward pressure on wages. Others complain that the definition is too restrictive. Many people, such as those who are no longer actively looking for work, are classified as being out of the labour market, In fact, they may still be a big part of the jobless problem. One solution to this would be to use "non-employment" instead. This measure includes not just the unemployed but other economically inactive people as well. It would capture those who had simply “dropped out” of the labour force, and might be a better guide to the impact of joblessness than standard rates.

The snag with such a measure is that it is too broad to be useful. The “non-employed" include retired people, housewives, students and others who may not be able to work, or may not wish to. Thus, an increase in “non-employment” could even be a good thing — if, for example, it meant that more people were staying on longer in education.

An alternative approach is to try to supplement the coverage of the standard unemployment rate. Many economists reckon that two groups of people, in particular, should be added. These are "discouraged workers” (those who would like to work, but are not actively seeking a job because they do not think any suitable positions are available), and "involuntary part-time workers" (those who are working fewer hours than they would like, or who can find only a part-time job when they want a full-time one). Once such discouraged workers are added in, the argument goes, the unemployment rate will provide a better measure of any excess supply of labour.

In its latest Employment Outlook, the Organization for Economic Cooperation and Development (OECD) examines how useful these supplementary measures are. They certainly change the picture of unemployment substantially. In 1993, there were 4 million discouraged workers in OECD countries and 15 million involuntary part-time ones, compared with 34 million officially unemployed. Adding the two categories to the standard definition of unemployment would increase the total number of unemployed people in OECD countries by over 50%.

In some countries, the difference would be even more dramatic. Total unemployment would more than double in Japan, primarily because of discouraged workers. In Sweden, the number of involuntary part-time workers is higher than the number of officially unemployed. Even America’s unemployment rate would suddenly look much less rosy (see chart). The effect would be to reduce the variation between OECD countries, though the overall ranking would not change much. Spain would still have the highest rates, Japan the lowest.

Cross-border comparisons of these broader indicators are not straightforward, however. Measures of discouraged and involuntary part-time workers are not standardized in the same way as the official unemployment rate. In Germany, for instance, the category of “discouraged worker” does not even exist.

Measurement problems aside, the OECD raises a more serious concern about the whole concept of discouraged workers. The rationale for including them, rather than other categories of non-employed, is that they are supposedly closer to entering the labour market. Were they not discouraged, they could rapidly find a job.

But after looking at the statistical evidence, the OECD pours some cold water on this notion. For example, it points out that most discouraged workers have been jobless for over three years. Evidence from America, Australia and Norway suggests that they may actually find it harder than other non-employed people to get a job. This makes the measure less reliable as an indicator of labour market pressures.

In fact, the OECD reckons that standard unemployment rates are still the best way of gauging pressures in the world’s labour markets. If the goal is to assess the broader social consequences of unemployment, then it seems odd to single out discouraged workers as opposed to some of the other non-employed. What is really needed is a far better understanding of why some people are not employed in the first place.