On the demand side, many existing jobs are destroyed every year, and many new jobs are created. Economic growth causes some sectors of the economy to decline and others to expand. Jobs are lost in the sectors that are contracting. Jobs are created in the expanding sectors.

Additionally, even in stable industries, many firms die and many new firms are born. The net increase in employment is the difference between all the jobs that are lost and all those that are created. In most years, enough new jobs have been created both to replace the old jobs that are destroyed and to provide jobs for the increased numbers in the labor force. The result has been a net increase in employment in almost all years. During the early 1980's, worldwide unemployment rose to high levels.

The unemployment rate remained high in many advanced industrial countries and only began to come down, and then very slowly, during the latter half of the decade. For purposes of study, the unemployed are classified in various ways. They can be grouped by personal characteristics, such as age, sex, degree of skill or education, or ethnic group. They can also be classified by geographical location, by occupation, by the duration of unemployment, or by the reasons for their unemployment.

Interestingly enough, the recorded figures for unemployment may significantly understate or overstate the numbers who are actually willing to work at the existing set of wage rates. Normally, overstatement arises because the measured figure for unemployment includes people who are not interested in work but who say they are unemployed in order to collect unemployment benefits. Also, understatement arises because people who would like to work but have ceased to believe that suitable jobs are available voluntarily withdraw from the labor force (Snowdon et al, 1994). Although these people are not measured in the survey of unemployment (which requires that people actively look for work), they are unemployed in the sense that they would accept a job if one were available at going wage rates. People in this category are referred to as discouraged workers. They have voluntarily withdrawn from the labor market, not because they do not want to work, but because they believe that they cannot find a job given current labor market conditions.

Generally, economists distinguish three types of unemployment: cyclical, frictional and structural. It is important to mention that a certain combination of frictional and structural unemployment exists even when the national income is at its potential. Together, these two types of unemployment make up natural rate of unemployment. Cyclical unemployment can be measured as the number of persons currently employed minus the number of persons who would be employed at potential income. When cyclical unemployment is zero, the number of unfilled jobs currently available is equal to the number of persons unemployed.

In this situation, all existing unemployment is either structural or frictional, and the rate of unemployment is the NAURU or natural rate of unemployment (Snowdon et al, 1994). People who are cyclically unemployed are normally presumed to be involuntarily unemployed in the sense that they are willing to work at the going wage rate, but jobs are not available. The persistence of cyclical unemployment poses a challenge to economic theory. Frictional unemployment results from the normal turnover of labor.

An important source of frictional unemployment is young people who enter the labor force and look for jobs. Another source is people who leave their jobs. Some may quit because they are dissatisfied with the type of work or their working conditions; others may be fired. Whatever the reason, they must search for new jobs and that naturally takes time.

Persons who are unemployed while searching for jobs are said to be frictionally unemployed, or, alternatively, in search unemployment. The normal turnover of labor would cause frictional unemployment to persist, even if the economy were at potential income and the structure of jobs in terms of skills, industries, occupations, and location were unchanging. Some frictional unemployment is clearly voluntary. For example, a worker may know of an available job but may not accept it so he can search for a better one.

The new classical view regards all frictional unemployment as voluntary. Critics of that view argue that many of the frictionally unemployed workers are involuntarily unemployed because they lost their jobs through no fault of their own and because they have not yet located a specific case of jobs they believe to be available somewhere and for which they believe to be qualified. The other component of the natural rate of unemployment is structural unemployment. Structural adjustments can cause unemployment. When the pattern of demand for goods changes the pattern of the demand, labor structure changes as well. Until labor adjusts fully, structural unemployment develops.

Such unemployment may be defined as unemployment caused by a mismatch between the structure of the labor force in terms of skills, occupations, industries, or geographical locations and the structure of the demand for labor. The evolutionary shift toward service-sector employment and the restructuring within all industries in response to technological change has favored workers with more years of schooling. To meet changing demands, the structure of the labor force must change. Some existing workers can retrain and some new entrants can acquire fresh skills, but the transition is often difficult, especially for experienced workers whose skills become economically obsolete. Structural unemployment occurs when such adjustments are slow enough that severe pockets of unemployment develop in areas, industries, and occupations in which the demand for factors of production is falling faster than the supply. One of the more dramatic recent structural changes in the economy has been in the organization of the firm.

Manufacturing firms used to be organized much like an army, with a pyramidal command structure. Most key strategic decisions were made near the top and lesser ones concerning implementation at lower levels. This structure required an array of middle-level managers who passed information upward to the top level and downward to the production level and who made various secondary decisions themselves. Recent changes pioneered among Japanese firms have led to a much looser organization with much more local autonomy among the subsections of the firm. As a result, large numbers of middle managers' jobs have been made redundant. They find themselves on the labor market at middle age and with many of their skills rendered obsolete.

Government policies can influence the speed with which labor markets adapt to changes. Some countries, such as the United Kingdom and Canada, have adopted policies that discourage movement among regions, industries, and occupations. These policies tend to raise structural unemployment. Others, such as Sweden, have done the reverse and have encouraged workers to adapt to change. Partly for this reason, Sweden's unemployment rates were well below the European and North American norms during the 1980's. Policies that discourage firms from replacing labor with machines may protect employment over the short term.

If, however, such policies lead to the decline of an industry because it cannot compete effectively with innovative foreign competitors, serious structural unemployment can result in the long run. Minimum-wage laws cause structural unemployment by pricing low-skilled labor out of the market. Effective minimum-wage laws have two effects when they are imposed on competitive markets: They reduce employment of the unskilled, and they raise the wages of the unskilled who retain their jobs. In the United States today, however, minimum wages are so low in most states that they probably have little effect on wage rates and employment. The major characteristic of both frictional and structural unemployment which together make up natural unemployment is that there are as many unfilled vacancies as there are unemployed persons.

Over the years, unemployment has been regarded in many different ways. Harsh critics see it as proof that the market system is badly flawed. Reformers regard it as a necessary evil of the market system and a suitable object for government policy to reduce its incidence and its harmful effects. Others see it as overblown in importance and believe that it does not reflect any real inability of workers to obtain jobs if they really want to work. Most government policy has followed a middle route.

Fiscal and monetary policies have sought to reduce at least the most persistent of recessionary gaps, and a host of labor market policies have sought to reduce the incidence of natural unemployment. Such social policies as unemployment insurance have sought to reduce the sting of unemployment for the many who were thought to suffer from it for reasons beyond their control.