Parkin discusses three types of unemployment: frictional, structural, and cyclical. Frictional unemployment is an outcome of normal labor turnover -- that is, the movement of people into and out of the labor force and the creation and destruction of jobs in a dynamic economy are both sources of frictional unemployment. While no one enjoys being unemployed, frictional unemployment -- associated with workers and employers searching for new jobs and new employees, respectively -- is in fact desirable as a sign of dynamism in the labor market.
Structural unemployment reflects underlying changes in the structure of the economy. More specifically, structural unemployment occurs when there is a mismatch between the skills required in jobs that are destroyed as compared to the skills required in jobs that are created. This type of mismatch occurs most frequently as a consequence of technological change or foreign competition. Similarly, structural unemployment also occurs when new jobs are created in regions different from those where old jobs have been destroyed (cf., Rust Belt/Sun Belt in the 1970s and early 1980s).
Cyclical unemployment is the unemployment that occurs as part of the business cycle. In particular, periods of recession normally entail increases in the volume of layoffs (job destruction exceeds job creation), and this represents cyclical unemployment. By its nature, then, such unemployment tends to be a relatively short-term phenomenon, like frictional unemployment and in contrast to structural unemployment. However, the duration of unemployment also tends to rise near the trough of a business cycle and decline during expansions (i.e., more of it is longer-term during the worst of a recession).
In addition to the three types of unemployment described by Parkin, economists also distinguish seasonal unemployment -- the unemployment that results from seasonal changes in the demand for labor (as occurs in agriculture or tourism) or in the supply of labor (as with students in the summertime). Seasonal fluctuations are present not only in unemployment, but also in numerous other economic indicators, and monthly and quarterly government statistics typically are seasonally adjusted to remove these short-term influences. These seasonal adjustments cancel each other out over the course of a year, and hence are not relevant when we consider annual data.
The natural rate of unemployment and full employment
Economists define the natural rate of unemployment as the level of unemployment existing in the absence of cyclical (and seasonal) unemployment. That is, the economy is at the natural rate of unemployment when there is only frictional and structural unemployment. This is a theoretical and somewhat fuzzy concept, and consequently there is no consensus as to the precise level of the natural rate of unemployment. Another name for the natural rate of unemployment is the full-employment rate of unemployment.
The extent to which the actual unemployment rate differs from the natural unemployment rate thus reflects the presence of cyclical unemployment. Economists use the term full employment to refer to the unemployment rate that exists in the absence of cyclical unemployment -- i.e., when there is only frictional and structural unemployment (ignoring seasonal unemployment). That is, full employment occurs when the measured unemployment rate is equal to the natural rate of unemployment.
Although there is no consensus on the precise level of the natural rate of unemployment, economists agree that the natural rate can change over time in response to various factors. For example, changes in the demographic composition of the labor force will influence the natural rate of unemployment. In particular, the rapid growth in the youth labor force that took place in the late 1960s and 1970s as a consequence of the entry of large baby boom cohorts into the labor market entailed an increase in the natural rate of unemployment, because of the high frictional unemployment levels of young workers.
Reflecting the changing demographic composition of the labor force, the full-employment unemployment rate was generally considered to be about 4 percent during the 1960s, while more recently it has been estimated to be somewhere in the neighbor-hood of 5 to 6 percent.
Similarly, rapid technological change or increased foreign competition may bring about an increased level of structural unemployment in the economy. This occurred in the U.S. during the late 1970s and early 1980s, when the natural rate was believed to be distinctly higher than its present level (5-6%).
As Parkin notes, the concept of the natural rate of unemployment is a good example of a technical economic term that does not correspond to everyday language. But zero unemployment -- what one might initially think of as full employment -- is essentially impossible to attain in a dynamic economy where frictions and structural change create unemployment.
Explaining employment and wages
The basic model of demand and supply that we looked at earlier in the course provides a framework for understanding the behavior of employment and wages. We need to apply the model to the labor market.
The demand curve for labor shows the quantity of labor that firms plan to hire at alternative real wage rates. Its downward slope reflects the profit-maximizing behavior of firms: when labor's real cost is lower, it is profitable for firms to hire additional labor and hence produce additional output (underneath this is the diminishing marginal product of labor, reflecting the law of diminishing marginal returns).
The supply curve of labor shows the quantity of labor that households are willing to supply to the market at various wage rates. It has an upward slope, reflecting the fact that as the real wage rate rises the opportunity cost of not being in the labor market rises correspondingly -- i.e., the incentive to work in the market rises.
As Parkin notes in his section on trends in employment and wages, over time there has been an outward shift in the supply curve of labor, largely reflecting population growth. There has also been an even larger increase in the demand for labor, as a consequence of increased productivity of labor stemming from increases in education of workers (human capital), increased capital per worker, and improvements in technology.
The combination of moderately increased supply and more substantially increased demand thus can account for the increased volume of employment (or aggregate hours of work) in conjunction with increases in the real wage rate. This is shown in Parkin's Fig. 7.12. (Note that this is a double shift, and one in which quantity must increase but price -- the real wage -- could in theory rise, fall, or stay the same.)
The discussion above focuses on the labor market as a whole. However, we can identify and characterize distinct segments of the aggregate labor market -- e.g., by industry, occupation, or education. The presence of structural change in the economy means that what happens in the overall labor market may not be what happens in each market segment (i.e., in each industrial, occupational, or educational labor market).
More specifically, Parkin provides an example focused on industry (manufacturing versus services), in which the structural change that has taken place over the past 15 years or so has led to increased real wages and employment in services and decreases in real wages and employment in manufacturing.
A similar perspective may be obtained by thinking in terms of labor markets for workers of different education (skill) levels. The structural changes of the past 15 years and more appear to have pushed out demand for well-educated workers while reducing demand for less-educated workers.
The consequences of these shifts, as shown by the demand and supply graphs below, have been an increase in real wages and employment for workers with high levels of schooling, and de-creases in real wages for workers with lower levels of schooling.
These changes, in turn, have resulted in increased wage or earnings inequality in the U.S. economy. The overall average is increasing, but those at the bottom of the wage distribution are doing worse now as compared to those at the bottom 20 or 25 years ago (see transparency on changes in wage distributions).
© 1996 David Shapiro
Return to the Econ 4 index page.
Return to David Shapiro's homepage
For questions, comments, and suggestions, write to: dshapiro@psu.edu.