Basic definitions
The Current Population Survey (CPS), which is a monthly survey covering about 60,000 households, is the basis for classification of the population into various labor force categories. The labor force consists of that portion of the noninstitutionalized population that is aged 16 and over and that is either employed or unemployed.
In order to be classified as employed, an individual must have a full-time or part-time job (of at least one hour's duration if paid; 15 hours for an unpaid family worker). Those with jobs who are ill, on vacation, on strike, or otherwise temporarily absent from work are counted as employed.
To be classified as unemployed, an individual must be available for work and have made some specific effort during the past four weeks to have found a job (except for individuals waiting to be recalled from layoff). Note, then, that the noninstitutionalized population aged 16 and over may be divided into three mutually exclusive and exhaustive categories: employed (N), unemployed (U), and out of the labor force (OLF).
This classification framework has been used for a long time, and is the basis for the data that we study concerning employment and unemployment in the economy. There are shortcomings to this scheme, however. For example, people who would like to be employed full time but can only find part-time work, and who logically might be classified as partly employed and partly unemployed, are counted as employed (involuntary part-time employment).
In the same vein, people who would like to work but feel no work is available and hence have stopped looking are counted as being out of the labor force, despite the fact that a stronger economy would induce them to search actively again (these are discouraged workers). Note the cyclical dynamics: improved economic activity increases the number of jobs, but also pulls previously discouraged workers into the market, meaning that the overall level of unemployment declines only with a lag.
Finally, people unable to find work in the field in which they are trained who end up accepting lower-level jobs (the lawyer driving a taxi, the college grad working behind the counter at Burger Thing) are classified as employed. But this fails to take account of the fact that they are in fact not utilizing the skills they have acquired (this is underemployment -- when you're working at a job that does not make full use of your skills).
These shortcomings lead some critics to argue that official unemployment figures do not always accurate reflect the extent of unemployment and underemployment in the economy. This criticism is most pertinent in time of recession, when involuntary part-time employment, discouraged workers, and underemployment will be most prevalent.
Labor market indicators
Three key indicators of what's going on in the labor market are the unemployment rate, the labor force participation rate, and the employment to population ratio. As we saw earlier, the unemployment rate equals the total number of unemployed as a percentage of the labor force (i.e., UR = 100 x U/LF = 100 x U/(N+U). As is clear from Parkin's Fig. 7.2, the unemployment rate is an inverse indicator of the business cycle (counter-cyclical), rising during times of recession and falling during periods of expansion.
The labor force participation rate measures the labor force as a percentage of the adult population: LFPR = 100 x LF/Pop = 100 x (N+U)/Pop. The LFPR exhibits a modest amount of procyclical behavior, largely because expansions tend to attract previously discouraged workers back into the labor force while recessions eventually "push" people into discouraged-worker status. In addition, as shown in Parkin's Fig. 7.2, there is a clear upward trend in the aggregate participation rate over the past few decades, reflecting the rapid growth in participation by women that has more than offset the gradual declines in participation by men.
The employment to population ratio gives the percentage of the adult population that is employed: EPR = 100 x N/Pop. Like the aggregate LFPR, it has an upward trend and behaves pro-cyclically, but in a more pronounced manner than the LFPR (it mirrors the UR). With a little reflection, it should be clear that these three measures are linked, in that
.Wages and hours - a first look
The real wage rate is a key economic indicator. It is defined as the money (nominal) wage rate divided by the price level (W/P), and hence it measures the purchasing power of labor time (i.e., the amount of goods and services that an hour's work can buy).
Over time, we experience inflation, but we also observe increases in money wages. If the money wage increases exceed the price increases, then real wages are rising. Economic growth in per capita terms is equivalent to a phenomenon of rising real wage rates, and for most of American history there has been a clear upward trend in real wages.
However, the experience of the past 20-25 years has been taken by many people to indicate that the long-term trend has been halted. This can be seen in the red line in Parkin's Fig. 7.5, which shows declining average real hourly wage rates for production workers in manufacturing from the early/mid-1970s to the present. There is a good deal of newspaper ink devoted to the idea that since the 1970s real wages have been either declining or at least stagnant.
But as shown nicely by the green line in Parkin's Fig. 7.5, if we go beyond manufacturing production workers and include all wages and salaries earned in the economy, the decline disappears. Instead, we see stagnation throughout the 1970s and into the early 1980s, and then resumption of growth beginning in the mid-1980s (although at a slower pace than in the 1960s).
If we broaden further our consideration of compensation, it gives an even brighter picture of real wage changes since the 1970s. That is, if we include fringe benefits such as employers' payments for health insurance and pension contributions (which amount to roughly 25-30% of wages and salaries and hence 20+% of total compensation), it appears that there was only a brief slow-down in the mid-1970s and that in fact even since 1975 real compensation has been increasing (the blue line in Fig. 7.5).
In brief, then, Parkin's analysis suggests that since the 1970s there has not been a decline nor even much stagnation in the average real wage rate. Rather, he finds that average real wages have been rising.
This conclusion is reinforced even further if we take into consideration the notion presented earlier that our measures of price changes (CPI, GDP deflator) overstate annual inflation by 1-2 percentage points (again, see the "Marx for Dummies" article included in the notes to ch. 6).
Aggregate hours of work show an upward trend and a clear procyclical component. In fact, hours are one of several leading indicators of the business cycle (note leading, lagging, and coincident indicators). The growth in total hours of work is smaller than the growth in employment. This is partly due to modest declines in the average full-time work week, but primarily it reflects the fact that an increasing proportion of the labor force is engaged in part-time employment.
Unemployment - a brief first look
There are three basic ways by which people become unemployed: by losing a job (being fired or laid off, which represents an involuntary separation); by leaving a job without already having or quickly finding another job (quits, or voluntary separation); or by entering or reentering the labor force without finding a job right away (cf., students who finish school, mothers who return to the labor force after having children, previously discouraged workers).
Reflecting the importance of job destruction in the economy, job losers are the largest of these three categories. Further, the number of job losers varies countercyclically, rising sharply in time of recession and falling in time of expansion.
The incidence of unemployment varies considerably by demographic group. For example, for the past 40 years or so blacks have consistently had unemployment rates that were twice as high or more than those of whites. Teenagers also tend to have especially high unemployment rates.
These differences across groups stem from a variety of causes. For example, the racial difference in unemployment appears to be clearly related to discrimination in labor markets, among other factors; while high teenage unemployment rates reflect job shopping and weaker labor force attachment by younger workers.
© 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.