previous CHAPTER 4

Chapters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Chapter 1
General Economic Concepts
Chapter 2
What is Policy?
Chapter 3
History of Macroeconomic Policies in the United States
Chapter 4
Policy Goals: Maximum Employment
Chapter 5
Policy Goals: Maximum Production
Chapter 6
Policy Goals: Price Stability
Chapter 7
Policy Goals: External Balance
Chapter 8
Subsidiary Policy Goals
Chapter 9
Conflicting Policy Goals
Chapter 10
The Policy Makers
Chapter 11
The Policy Instruments
Chapter 12
The Decision-Making Processes
Chapter 13
The Policy Indicators
Chapter 14
The General Economic Model
Chapter 15
Monetarist Monetary and Fiscal Policies
Chapter 16
Keynesian Monetary and Fiscal Policies
Chapter 17
Debt Management Policies
Chapter 18
Incomes Policies
Chapter 19
Supply Management Policies
Chapter 20
The Long Wave
Chapter 21
Contemporary Issues
"Whilst workers will usually resist a reduction of money-wages, it is not their practice to withdraw their labour whenever there is a rise in the price of wage-goods. It is sometimes said that it would be illogical for labour to resist a reduction of money-wages but not to resist a reduction of real wages. ... Moreover, the contention that the unemployment which characterises a depression is due to a refusal by labour to accept a reduction of money-wages is not clearly supported by the facts."

John Maynard Keynes
British Economist
The General Theory of Employment, Interest and Money (1936)

  1. Definitions and the Problems
  2. Measurement Problems
  3. Types of Unemployment
  4. Consequences of Unemployment
  5. Full Employment
  6. What Are We Measuring and Are We Using The Right Numbers?
  7. Summary
top    I. Definitions and the Problems

The Employment Act of 1946 mandates maximum employment as the first of three macroeconomic goals. Maximum employment starts with measurement of an economy's labor force.

Labor Force

Economic concept of the labor force The labor force is deduced from the population as a whole. Economists start with the whole population and delete those who are unable to work, e.g., the mentally ill, the institutionalized, babies, the elderly, and the infirm, and those who are unwilling to work, e.g., leisure class:

LT = population - unable individuals - unwilling individuals

Empirical concept of the labor force In the U.S., the civilian labor force (Lc) includes those individuals 16 years of age or older who are willing and able to work at the going wage rate. It includes the unemployed and the underemployed (U) as well as the employed (Le). It includes sole proprietors and the self-employed. It includes those who work part-time, full-time, and over-time. The labor force, however, excludes all persons engaged exclusively in housework in their homes and those who are attending school. Students are not members of the labor force unless they are working in addition to attending school. If they work or look for work during the summer vacation period, they become members of the labor force. Likewise, when they graduate, they generally become members of the labor force.

Lc = Le + U

The total labor force (LT) may include members of the armed forces (Lm):

LT = Le + Lm + U

but inclusion is discretionary, since all military personnel are employed, by definition. In 1983, the Census Bureau opted to include domestically-based military personnel in its measure of the labor force, thus moving the measure closer to the economic concept, but it discontinued this approach in 1994.

The labor force participation rate is the ratio of the labor force to the population. It is calculated as a percentage of either the total population or the non-institutionalized population:

LT/Population or LT/Non-institutionalized population


Employed labor includes those civilians who are willing and able to work at the going wage rate, and currently have jobs (Le).


Unemployed labor includes those individuals who are willing and able to work at the going wage rate, but who cannot find jobs (U).

Economic concept of unemployment To economists, unemployment is a residual, what is left over, after accounting for all persons (both civilian and military), who are employed:

U = LT - Le + Lm

Empirical concept of unemployment (Bureau of Labor Statistics (BLS) Method) The BLS is a division within the Department of Labor. It constructs the labor force statistics from data collected by the Census Bureau during a survey it conducts of randomly selected households for one week during the calendar week containing the 19th of the month to determine which individuals are employed, unemployed, or not in the labor force. The sampling covers 65,000 households in 754 areas of the country.

The randomly selected interviewees, chosen at headquarters by address rather than by name, are surveyed eight times before being replaced. They are interviewed in four consecutive months and then again in those same months a year later. The first and fifth interviews for each household are conducted face to face — what the Census Bureau calls "personals." The rest are done by telephone. The typical interview can last anywhere from five minutes for a retired couple to well over an hour for a large family in the March survey, in which many detailed income-related questions are added. The interviewer must get information on everyone at the address who is 14 years at age or older. According to the survey question responses, persons are classified as "employed", "unemployed", or "not in the labor force." The questions are designed in such a way that all respondents must fall into one of the three categories. The labor force is defined and calculated as the sum of those who are employed and unemployed:

L = Le + Lm + U

The employed labor force includes persons who did not work at all during the census week because of illness, bad weather, vacation, or labor disputes, but who had jobs or businesses.

The unemployment rate is the percentage of unemployed labor in the labor force.

u = U/LT = 1 - Le/LT - Lm/LT

Any change in the total labor force with stable employment (Le + Lm) will cause unemployment (U) to change: e.g., wars, famines, birth rates, changing attitudes (e.g., the Women's Liberation movement), compulsory education, changes in the minimum work age, etc.

Any change in military employment (Lm),with a stable labor force (LT) and constant civilian employment (Lc),will cause unemployment (U) to change: e.g., U.S. draft policies, monetary and other incentives to encourage enlistment in the military, war or its de-escalation, etc.

Any change in civilian employment (Le), with a stable labor force (LT) and a constant military staff (Lm), will cause unemployment (U) to change: e.g., job satisfaction or dissatisfaction, quit rates, changes in geographical, social, and vocational mobilities, changing relative demands for different types of labor, the amount and duration of unemployment compensation payments, the flow of job information, etc.

The BLS releases the unemployment report on the first Friday of each month. (See also a graph of unemployment rates at the FRB Dallas.)

top    II. Measurement Problems

Historical Comparability of the BLS Surveys

The Census Bureau methodology for compiling and computing labor force statistics has been revised many times over the years. Originally, a person was unemployed only if he actually looked for work. In 1940, it was decided to classify non-lookers as unemployed, if they said they had not looked for work because they guessed no work was available. These figures were kept in a separate count until 1945, when it was decided to drop the distinction between "active" and "inactive" unemployed.

In 1954, the BLS revamped its field forces and increased the size of the sample area. That year, the old method found 2.4 million unemployed persons, while the new method uncovered 3.1 million unemployed persons — a difference of 22%!

In 1961, the BLS shifted from 14 year-olds to 16 year-olds as the basis for labor force definition. During the Reagan presidency, officials suggested changing that definition to 18 years of age or older. Since younger persons typically have higher unemployment rates, their elimination from the labor force would automatically reduce the number of unemployed and the overall unemployment rate. The suggestion was never implemented.

In 1983, the BLS added a new unemployment measure, called "total employment," which included armed services members on active duty in the U.S. as employed members of the labor force. Prior to 1983, all military personnel were excluded from the civilian unemployment rate. The BLS added the measure in response to a recommendation by the National Commission on Employment and Unemployment Statistics, an independent panel convened in 1978 to evaluate and recommend improvements to the nation's system of labor market information. The Commission determined that with the change to a volunteer military, military employment was not substantively different from civilian employment. Hence, the recommendation that the resident military be counted in employment and in the labor force totals and thus be reflected in an overall unemployment rate. The new unemployment rate for the total labor force was usually about one-tenth of one percentage point lower than the civilian rate (occasionally it was two-tenths lower). For instance, in 1983, the total unemployment rate (including the Armed Forces) was 9.5 percent, while the civilian unemployment rate was 9.6 percent. In 1993, the total unemployment rate was 6.7 percent, while the civilian unemployment rate was 6.8 percent.

The BLS discontinued its use of "total employment" in 1994 for three (3) main reasons:
  1. The Department of Defense data on resident armed forces were problematic. First of all, each branch of the armed services classified its troops differently. Second, many of the troops involved in Desert Storm were not in residence within the U.S., yet they were still classified as residents because their official stations were not changed from their U.S. posts.

  2. Only the New York Times reported the total rate rather than the civilian rate.

  3. The BLS did not have the same detailed demographic data on members of the armed forces as it had on the civilian labor force. This data deficiency made for difficult comparisons of particular demographic classifications, e.g., black teens, in the civilian labor force with the same demographic classifications in the total rate.
One of the most recent changes occurred in November 1993, after the Labor Department discovered a bias in its old survey. The questions in the old survey failed to capture many societal changes, such as the continuing growth in service-sector employment; the more prominent role of women, particularly mothers, in the labor force; and shifts in the nature of employment, including more part-time work and less permanent attachment of employees to their employers.

Under the old procedure, the interviewer asked the respondent to choose between alternatives, that could wrongly seem to be mutually exclusive. The interviewers often assumed that people who were keeping house were out of the labor force and did not ask if the respondent was on layoff or looking for work. This had the effect of dropping significant numbers of unemployed women from the labor force, lowering the unemployment rate. As a result, many women who were seeking jobs were erroneously described as homemakers and, therefore, not counted as being in the labor force. In addition, the old survey miscounted laid-off and so-called discouraged workers.

In 1993, the BLS changed the nature of the questions its interviewers were asking of the households. The questions in the new survey attempt to correct for these biases. Under the new procedures, the interviewer asks a series of "Yes" or "No" questions better designed to determine whether or not the respondent is part of the work force. Housework is not mentioned and the interviewer is not allowed the discretion to shape the question depending upon the sex or age of the respondent.

The BLS Survey Does Not Measure Hardship

A question often asked about the unemployment data is whether it should be used as a measure of idle, unused and/or underutilized resources or whether it should be used as a measure of hardship. The BLS Method does not do a good job of measuring hardship.

On the one hand, the BLS Method counts as "employed" persons who have jobs, whether or not they need one. It counts secondary wage earners, students, and pensioners with jobs as employed, despite the fact that they have other sources of income. It also counts as "employed" persons who have jobs, but the jobs are below their skill levels (and so are the incomes), e.g., a Ph.D in philosophy driving a taxi.

On the other hand, the BLS Method counts persons who are employed in part-time capacities as employed, despite the fact that these people might desire and even need full-time jobs to support themselves. It also counts people who volunteer to work as "not in the labor force," when, in fact, these volunteers would very much like to have jobs, even if only on a part-time schedule.

Interviewee Problems

Very often, households are reluctant to give true responses, especially for fear that the answers obtained from the survey will really remain confidential. On this score, the respondents have little to fear. The Census Bureau has a tradition of vigilantly protecting the confidentiality of the information it collects. For example, after the attack on Pearl Harbor, the Census Bureau refused to remit to other government agencies the names of Japanese-Americans. To make sure that interviewees do not invent responses, other interviewers routinely call various households to check the answers. Additionally, interviewers are required to maintain their "interview" telephone in an isolated part of their homes, where not even their families can overhear the conversations. Interviewers must destroy all unnecessary documents and to turn face down any work papers left in a locked car.

All information is forwarded to a central processing operation in Jeffersonville, Indiana, where the documents are microfilmed and put through an optical scanning device that converts the data to magnetic tape. The magnetic tapes are forwarded to Census Bureau headquarters in Suitland, Maryland, outside Washington, D.C. At Suitland, the information is edited, tallied, and prepared for pickup by the Labor Department's Bureau of Labor, which analyses and disseminates the information to the public on the first Friday of each month.

The BLS Survey Does Not Capture Employment in the Underground Economy

The underground or "off-the-books" economy has been thriving since the early 1970s as higher rates of inflation pushed households and businesses into higher marginal income tax brackets. In addition to the illegal underground activities of organized crime, which employment is not recorded, many decent citizens switched to cash-and-carry transactions to avoid any record of income for tax purposes. Cabinet makers, plumbers, roofers, and other self-employed persons, including the local handyman, are gainfully employed, yet do not respond favorably to the BLS survey for fear of getting caught. For example, in Homestead, Pennsylvania, home of the infamous Homestead Strike of 1892, much of the labor force was idled by steel plant closings in nearby Pittsburgh. The official unemployment has been as high as 40% in recent years. Yet, many of the unemployed ex-steel workers are gainfully, but covertly employed as taxicab drivers and local handymen. When interviewed, most will admit that they hide their employment to avoid income taxes. None will reveal their names for publication.

In 1976, Peter M. Gutmann, a Professor at the City University of New York, estimated the size of the underground economy at $175 billion, or 10% of the reported GNP. Since then, he has revised upward his estimate to 15%. Other economists, who originally set out to disprove Gutmann's conclusions, now put the percentage substantially higher. If Gutmann and other economists are correct, a 7% unemployment rate translates into an overemployment rate of 8%!

One of the biggest problems with the household survey, however, is the difficulty of adjusting for seasonal variation rather than inadequate sampling or interviewing practices.

top    III. Types of Unemployment

Voluntary Unemployment

Voluntary unemployment consists of all individuals who prefer leisure to work at the going wage rate. These individuals may be keeping house, attending school, unable to work, or simply idle by choice. Since these persons prefer not to work they are not included in the labor force. In 1947, the first year of the census, 42,500,000 persons 16 years of age and older were considered to be "not in the labor force." As of December 2009, approximately 84,000,000 persons 16 years of age and older, who are not institutionalized, are considered to be "not in the labor force."

Voluntary unemployment includes induced unemployment. Induced unemployment results from public policies that make it easier for labor to become unemployed and to remain voluntarily unemployed for long periods of time. Public policies that induce unemployment include unemployment compensation as well as welfare, food stamps, rent subsidies, and health care programs.

The unemployment compensation program has the capacity to induce unemployment. The tax levied on an employer is not in direct proportion to the employer's layoffs. Thus, firms find it easier to layoff workers than it would be if they had to pay the full cost of the resulting unemployment benefits. Moreover, the tax is a cost of hiring workers over and above the wage rate. Employers may either offer lower wages for the same labor employment or offer less employment at the going wage.

The unemployment compensation program has the capacity to prolong unemployment. In many cases, the difference between unemployment compensation and the net income, which the person could earn from another job, is minimal. This situation is especially true where laid-off workers, such as those in the auto industry, receive supplementary unemployment benefits (SUBs) from their employers. In such cases, total unemployment benefits provide unemployed workers with an income equal to 90% - 95% of their take-home pay. Overall compensation prolongs the need for job search and increases the overall level of unemployment in the interim.

Other programs, such as welfare, food stamps, rent subsidies, and health care programs, cause a similar upward bias in the rate and the duration of voluntary unemployment. Sometimes the lack of low cost programs, like child care, impede workers, especially young mothers, from taking available, but low-paying jobs. As long as employers are not offering big salaries, substantially above the benefits of the dole, many workers will prefer leisure to work. The solutions to reducing induced unemployment are reduced benefit payments, reduced benefits periods, subsidized "workfare" programs, improved the quantity and quality of daycare programs and facilities for unemployed persons with young children, etc.

Involuntary Unemployment

Involuntary unemployment consists all individuals who are willing and able to work, but who cannot find jobs, at the going wage rate. It includes frictional unemployment, structural unemployment, and cyclical unemployment.



frictional structural
Frictional unemployment is a supply-side phenomenon. It changes as workers enter and reenter the labor force and/or quit one job to look for another. The number of new entrants and reentrants depends principally upon demographics, changing attitudes toward work, government draft policies, and military preparedness. The number of job-switchers (the "quit" rate) depends principally upon job (dis)satisfaction and the degree of perceived upward mobility from job-switching.

Given that new entrants and re-entrants continually flow into the labor market and given that some workers are always switching jobs, an element of frictional unemployment is always evident in the unemployment statistics.

Frictional unemployment is generally short-term (30-60 days), with little or no hardship, as job-switchers eventually find new jobs and entrants are slowly absorbed into new jobs.

The solution to reducing frictional unemployment is improved information flows to match job applicants with job vacancies in a faster, more efficient manner, e.g., a centralized job bank.
Structural unemployment is a demand-side phenomenon. It changes as demand for the final product changes, as labor's productivity changes relative to the productivity of capital, and as labor costs in a region change relative to those in other areas. First, as individuals' tastes and preferences change and buyers switch from one product to another, the demand for labor to produce some goods increases, while the demand for labor to produce other goods falls. Second, as technology changes, and capital becomes relatively more productive and less expensive, businesses tend to substitute capital for labor. This "automation" of the production process reduces the demand for certain types of labor. Finally, in a large country like the United States, regional wage differentials may make skills obsolete in some areas, while the demand is high in other areas. Most often, this change in relative competitiveness crosses country boundaries changing international competitiveness and relative job opportunities. Geographic immobilities — the inability or unwillingness of workers to migrate — aggravate this condition. Unemployment in various parts of the U.S. can remain high as jobs go begging in other parts of the country because the cost of living prevents the jobless from migrating to where the jobs are.

Given that relative demands for products are continuously changing, given that new technologies make capital more productive than labor in certain occupations, and given that relative competitiveness is changed through technological transfers and wage differentials, some element of structural unemployment is always evident in the unemployment statistics.

Structural unemployment is generally long-term, with much hardship, as workers either need to relocate and uproot families or need to acquire new skills to reenter the job market. Some quit looking for new jobs and are dropped from the labor force and the unemployment statistics. These unemployed are the discouraged workers.

The solutions to reducing structural unemployment are improved job-search techniques, outplacement counseling, retraining programs, tuition grants, low-cost education loans, subsidies or tax credits to businesses for retraining, relocation subsidies and loans, venture capital funds to help workers start new business, etc.


Cyclical unemployment results from a downturn in the general level of economic activity. A decrease in aggregate demand results in an overall decline in the level of spending in the economy and reduces the overall demand for labor. The decreased spending is due to an excess of saving over investment, taxation over government spending, or imports over exports. Cyclical unemployment is generally short-term, depending on the nature of the downturn. Some downturns or recessions are long, but not severe; others are more protracted, but short-lived. Only the depressions are both long and severe. The solution to reducing cyclical unemployment is to increase aggregate demand through demand management policies (monetary and fiscal policies).


Volunteers and part-time workers Many individuals who are working on a part-time basis would prefer to work full-time, if they could get a position, and many volunteers would prefer to work at least part-time. Computation of this underemployment may be based on lost man-hours of work. Assume that full-time employment averages 37.5 hours per week and that voluntary part-time employment averages 17.8 hours per week. If the average hours worked by a part-time employee who prefers full-time employment is 23.3, then underemployment of involuntary part-time workers is (37.5 - 23.3 =) 14.2 lost man-hours per person per week and underemployment of volunteers is 17.8 lost man-hours per person per week. If the total available labor force is 110 million and 90 million are employed full-time, 5 million are voluntarily employed part-time, 5 million are involuntarily employed part-time, and 10 million are volunteers looking for part-time employment. then total man-hours available are (90 + 5 =) 95 million times 37.5 man-hours for full-time employment, plus (5 + 10 =) 15 million times 17.8 man-hours for part-time employment. Potential full-time employment (3,562.5 million man-hours) plus potential part-time employment (267 million man-hours) equals 3,829.5 million man-hours. Underemployment of involuntary part-time workers equals (5 million workers times 14.2 man-hours =) 71 million lost man-hours and underemployment of volunteers equals (10 million volunteers times 17.8 man-hours =) 178.0 million lost man-hours for total underemployment of 249.0 million lost man-hours. The percentage of underemployment equals (249.0/3,829.5 =) 6.5%.

The method, however, has many weaknesses. First, no standard man-hour work week exists. What standard should be used? A 35-hour work week? A 37.5 hour-hour work week? A 40-hour work week? Even those who are working full-time, 40 hours per week, may desire to work more hours, if work were available. Technically speaking, these individuals are also underemployed. Second, no standard exists to account for overtime employment by those who are gainfully employed full-time. Should overtime employment be subtracted to arrive at a net underemployment rate? Unfortunately, netting out overtime employment does not solve the hardship problems for those who are, in fact, underemployed. The underemployed are not substitutable for the overemployed. Finally, no standard exists to account for moonlighting (multiple job holders)? Although moonlighters are primarily holders of part-time jobs in addition to regular employment, these jobs could alternatively be made available to those seeking part-time employment. Fortunately, for those who are so employed, no laws restrict choice of employment or hours worked. Unfortunately, for those who desire such part-time employment, no laws exist.

Underused Talents, Skills, and Education Individuals who are specially trained and educated may not find jobs in their chosen professions. Subsequently, they end up taking jobs for which they are overqualified. Some examples are the Ph.D. in philosophy who wants to be a college professor, but ends up driving a cab or selling insurance; the concert pianist who mops floors for a living; and the aspiring young actress who waits tables. These individuals are obviously gainfully employed, but they are not fully or productively employed. Unfortunately, since cross-occupational qualifications are difficult to quantify, this type of underemployment cannot be measured. In fact, a concert pianist who is driving a cab may actually work more hours as a cab driver than he would as a concert pianist and he may also be earning a higher income.


Multiple job holders and moonlighters Some individuals hold down more than one job. These workers, however, are counted only once in the employment statsistics.

Underqualified, overemployed persons Some individuals rise to their highest level of incompetence. They are underqualified for the jobs they are holding, but for one reason or another they are never let go. Sometimes these individuals are given lateral transfers within the organization. If a measure of underemployment is adjusted for occupational over-qualification, it should likewise be adjusted for overemployment by occupational underqualification. Since neither type of qualification is readily quantifiable, statisticians simply acknowledge their existence and let it go at that.

top    IV. Consequences of Unemployment

Economic Waste

One consequence of unemployment is that output for the economy as a whole is reduced. Output in the present time is less then what could be enjoyed if labor were fully and gainfully employed. This loss of output is cumulative. The loss of capital formation due to unemployment in 2010 means that the capital resource base in 2011 is not so large as it would have been if labor had been fully and gainfully employed. Output lost from unemployment in 2010 cannot be made up in 2011.

Human Waste

The other consequence of unemployment is that unemployed individuals suffer much hardship. First of all, there is the financial hardship — not being able to make ends meet. If the unemployed person was the only bread-winner in the family, others in the household may be requested to take up the slack. Mothers with small children may be forced to seek employment, and older children may be forced to work and forego higher education. Second, many unemployed persons suffer psychological and social deterioration. Incidences of crime, violence, drug abuse, alcoholism, and mental illness increase as unemployment increases. Finally, families of the unemployed are negatively affected in non-financial ways. Child abuse, wife-battering, and divorce occur more frequently among the unemployed.

top    V. Full Employment

Conceptual Definition

Full employment exists when everybody who is willing and able to work at the going wage (the labor force) is fully and gainfully employed and no involuntary unemployment exists. However, some involuntary unemployment always exists. At minimum, this "permanent" amount of unemployment will be partly frictional and partly structural. Frictional unemployment can be reduced with a better flow of job information, increased job satisfaction, and improved placement of job applicants in job vacancies. Structural unemployment, by its very nature, can only be reduced with increased programs for retraining and relocating displaced workers. Any definition of full employment must recognize that some element of unemployment will always exist and that frictional and structural unemployment can be reduced only at an increased cost.

NOTE: Employment can be increased even further, if the labor force itself is enlarged by minimizing voluntary unemployment. Voluntary unemployment can be reduced by increasing the going wage rate and improving the quality of jobs. Induced unemployment can be reduced by changing the quantity and timing of unemployment benefits. Note further that any attempt to increase the labor force also involves increased costs.

Natural Rate of Unemployment

The natural rate of unemployment (NRU) is the minimum sustainable rate of unemployment without an increase in the rate of inflation. NRU is also called the non accelerating inflation rate of unemployment (NAIRU). Both definitions include all of the non-cyclical types of unemployment (frictional and structural unemployment). Whenever the actual rate of unemployment goes below NRU or NAIRU, inflation increases, because the attempt to employ workers from the pools of frictionally and structurally unemployed increases costs and businesses pass along these cost increases as price increases. Whenever the actual rate of unemployment goes above NRU or NAIRU, the additional unemployment is called cyclical unemployment. Cyclical unemployment is considered to be unnatural, because it does not have to occur.

Empirical Definition

Theory does not dictate how high or low the natural unemployment rate is. At the time of its enactment, the Employment Act of 1946 allowed for a 3.5% to 4.5% range of unemployment. A 4.0% rate of unemployment was subsequently adopted as the measurement for full employment. However, this measurement was not necessarily scientific and, in all probability, quite arbitrary. Many years later, in 1986, Herbert Stein, a prominent economist, revealed that he was around in 1946 when the full employment figure was chosen and that the 4.0% figure was literally drawn out of a hat!

Hitting A Moving Target

The original 4.0% guideline has changed substantially over the years, especially since labor force changes lag population changes by approximately 18-20 years.

An upward shift A study by George Perry, another prominent economist, shows that a structural shift probably occurred in the mid-1960s, when the first of the baby boom children entered the labor force, and that the natural unemployment rate rose to 6.0% as a result of this demographic change. De-escalation of the Vietnamese War, changing social attitudes toward women in the workplace, and increased global competition from other countries probably added another percentage point to the natural unemployment rate.

The increasing rate of inflation in the U.S. may have contributed to an increase in the natural unemployment rate. In the late 1960s, 1970s, and early 1980s, as the U.S. inflation rate increased and U.S. workers earned higher wages, many companies found it more profitable to locate operations abroad in lower-wage countries. This transfer of production abroad left behind a pool of highly skilled American job applicants without offsetting job opportunities in the U.S.

U.S. foreign direct investment abroad also may have helped to increase the natural unemployment rate. U.S. foreign direct investment abroad quickened the pace of technological transfer. New technologies were adapted at a high rate by businesses in foreign countries, making foreign goods and services more competitive in international markets. Changing relative prices caused an increase in U.S. imports of foreign goods and services, even as U.S. exports to foreign markets were declining. With the decline in exports went valuable job opportunities for American workers.

Finally, rising interest rates may have played a role. As U.S. interest rates rose, foreigners bought dollars to take advantage of the higher relative rates of return in the U.S. This increase in the demand for dollars to buy U.S. assets pulled up the value of the dollar and changed U.S. competitiveness in international markets. U.S. exports became more expensive for foreigners, while foreign imports became cheaper for Americans. This change in global competitiveness contributed to the transfer of jobs abroad, leaving fewer jobs for Americans back home.

A downward shift A subsequent downward shift occurred, starting in the mid-1980s, such that the current natural rate of unemployment is probably about 4.0% to 4.5%. Some of the reasons for this shift are as follow:
  1. Falling birth rates, from 1965 through the mid-1970s, the lowest birth rate since the 1930s, meant an overall tightening of the job market approximately 18-20 years later. Fewer new entrants meant less frictional unemployment and, therefore, less overall unemployment starting in about 1985.

  2. The proportion of teenagers in the labor force has continually declined from 7.6% in 1965 to 6.5% in 1985 to 3.9% in 2010. Since inexperienced and unskilled teenagers tend to have the highest unemployment rates, taking this element of structural unemployment out of the labor force reduced the natural rate of unemployment.

  3. Women, who previously attained only nominal (high school) skills, began entering the labor market with the appropriate educational qualifications (college degrees) and experience to make them more employable. As female unemployment declined, so did the structural component of unemployment.

  4. More balanced growth eased structural unemployment by halting the absolute decline in manufacturing and farm jobs. The U.S. went through a period of structural realignment of resources and resource usage during the last quarter of the 20th century. Labor markets do adjust, but much more slowly than do the markets for capital or consumer goods. Retraining and relocation of labor take time. This structural realignment was completed in or about 2000.
Offsets Offsetting some of the above demographic trends is likely to be an increasing mismatch between job opportunities and job applicants.
  1. Since 1985, jobs in traditional low-skilled, entry-level positions, such as those in the manufacturing, mining, and construction industries, decreased. The job growth shifted to information processing, telecommunications, computers, financial services, biogenetics, and other areas where most of the jobs require abilities beyond just reading and writing. By the year 2000, people with less than a high school education filled fewer job vacancies than they did in 1985. In 2010, makers of innovative products like advanced medical devices and wind turbines are among those growing quickly and looking to hire, but they need workers with higher skills than they have.

  2. Not all cyclically unemployed workers are rehired when the economy improves. During a downturn, manufacturers often accelerate automation to cut costs. When the economy improves, they need to hire people who can operate sophisticated computerized machinery, follow complex blueprints, and demonstrate higher math proficiency than was previously required of the typical assembly line worker. The people who are out of work just do not match the types of jobs that are here, open and growing. Initially, these cyclicallt unemployed workers become part of the pool of structurally unemployed persons, but, eventually, they end up as part of the pool of long-term unemployed (discouraged workers).

  3. Currently, women make up about 46.7% of the work force as opposed to 33.4% approximately 50 years ago. Women continue to enter and remain in the labor force, either delaying childbirth or foregoing children altogether. Women can be expected to supply an increasing proportion of the net addition to the labor force both now and in the future.

  4. Blacks and Hispanics are expected to fare worse than whites, since the white birth rate dropped more precipitously than did black or Hispanic rates. Young whites will be a smaller percentage than ever of new entrants, while blacks and Hispanics will become a larger percentage. Since the unemployment rates for blacks and Hispanics tend to be higher than they are for whites, the natural unemployment rate will have an upward bias.

  5. Similarly, a geographic imbalance is expected to exist. A huge deficit of entry-level jobs is expected to exist in cities, where the bulk of the labor pool resides, and a huge deficit of entry-level workers is expected to exist in the outlying suburbs, exurbs, and small towns where the jobs will be. Jobs in suburban malls and fast-food chains do not do inner-city drop-outs much good.
top    VI. What Are We Measuring and Are We Using The Right Numbers?

Underutilization of Resources

A rate of unemployment supposedly indicates the extent of utilization of available rather then potential labor resources. If we wish to measure this level of utilization, then we must include underemployment. After all, the loss of output from such unemployment and underemployment is a real cost to society, not only in the present period, but cumulatively over time. By the same token, underemployment must be adjusted for overtime and secondary (moonlighting) employment to assess the true impact on real output. Unfortunately, most secondary employment is accomplished on a cash-and-carry basis to avoid income taxation. Therefore, it is difficult to measure.

Perhaps, a better measure of utilization is the labor force participation rate or the rate of employment (employment-population ratio). In 1948, when the Census Bureau first collected employment data, 58.8% of the civilian working-age population was in the labor force and over 56.6% of the working-age population was employed. Fifty years later, at the end of the 20th century, over 67.0% of the civilian working-age population was in the labor force and approximately 64.5% of the working-age population was employed. Although these measures have dropped to approximately 65.0% and 59.0%, respectively, in 2010, the U.S. is better utilizing its labor resources today than it did 60 years ago.

One disturbing trend is the number of adult males leaving the work force, either through voluntary early retirement or through involuntary forced retirement. In 1948, 90% of American males aged 55-64 were in the labor force. That ratio slipped to 83% in 1970 and accelerated downward to 72% in 1980. As of 2010, only 70% of this group is still in the labor force. Almost one of every three males (5.0 million out of 16.7 million) aged 55-64 is no longer working. While this may come as a welcome respite for those desiring more leisure, economists contend it is unhealthy. These workers could still be contributing their talents to society rather than drawing from it. For those who have been involuntarily forced into retirement, the vacant time can present a hardship, both psychologically and financially.

The biggest force behind the change is the restructuring of the United States economy. This older group of workers never participated in the recovery from the 1980-83 back-to-back recessions and, after a spurt of leveraged buyouts in the 1980s and the subsequent restructuring and downsizing of these companies, many middle-aged, middle-level managers found themselves out on the street, pounding the pavements, looking for jobs in other companies which were also restructuring and downsizing. Many more mid-level executives were let go during the 1991 downturn. Many of these middle-aged, middle-level managers were replaced by younger workers earning $20,000 per year less. Some were never replaced. Some became self-employed management consultants, working out of their homes. However, most have no hope of finding another job. Most are no longer looking for a job. These ex-workers become part of the structurally unemployed not counted in the unemployment rate.

Personal Hardship

The unemployment rate is also used to help assess the hardship experienced by families with workers who are willing and available for work, but who are unable to find the right jobs. Whether the unemployment rate indicates their hardship or need as precisely as one would like is questionable. The measure may be too low, if it excludes "discouraged" workers, who drop out of the labor force because they have lost all hope of finding a job, and if it includes part-time workers who want to work full-time. The measure may be too high, if it includes secondary wage earners (like housewives, househusbands, and students), and if it includes those unemployed persons receiving unemployment benefits, food stamps, and, perhaps, union benefits as well. In an attempt to uncover the amount of hardship from unemployment, the Census Bureau calculates and publishes several other data series on particular classifications of workers and the duration of unemployment:

Get the numbers:
top     Summary

Over the last 60 years, since the government has been keeping track of labor and employment data, labor force participation increased by approximately 92,000,000 persons, from 62,000,000 in 1950 to 154,000,000 in 2010. During these years participation increased an average of 2.5% per year. Most of this increase occurred during the years 1964-1986 as the "baby boom" children matured and entered the labor force. However, changing social, political, economic, and demographic trends also contributed to this increase.

Since the mid-1980s, the rate of entry has slowed significantly. The "baby boomers" had far fewer children than their parents, which slowed the rate of increase in labor force participation when these children matured and entered the labor force. Since 1990, the annual rate of increase in labor force participation has been only about 1.1%.

During the period of "baby boom" absorption, the unemployment rate remained at a relatively high rate of 7.0%. Since the mid-1980s, the average unemployment rate has dropped by more three whole percentage points to less than 4.0%. By 2000, with the unemployment rate under 4.0%, many economists and public officials feared that the economy was at or beyond its full-employment threshold and that inflation would begin to accelerate sharply. However, by 2003, the unemployment rate had once again risen to over 6.0%. From 2005 to 2008, the unemployment rate hovered between 4.5% and 5.0%, but it rose rapidly in 2008 and reached 10.1% in 2009 in response to the current financial crisis.

In the 1980s, economists believed that the full-employment or the natural unemployment rate was 7.0, but demographic shifts, the end of the baby boom, the integration of women in the work force, competitive pressures from overseas, and structural changes in the economy have clearly lowered the this rate. Also, wiht a fall in the value of the U.S. dollar, the rates of technological transfer and change in foreign countries has once again equalized competitiveness between the U.S. and the rest of the world. At this point, in 2010, the natural unemployment rate is probably about 5.0% with another 5.0% in cyclical unemployment.

top    Readings

Structural/Frictional vs. Deficient Demand Unemployment: Some New Evidence, Katherine G. Abraham, American Economic Review, LXXIII(4) Sep 1983: 708-23

Factory Jobs Return, but Employers Find Skills Shortage, Motoko Rich , New York Times, 2 Jul 2010

Okun's Law Revisited: Should We Worry About Low Unemployment? David Altig, Terry Fitzgerald, and Peter Rupert, Economic Commentary (FRBClev), 15 May 1997

In Search of the NAIRU: What do dating and Brad Pitt tell us about unemployment? David Altig and Paul Gomme, Economic Commentary (FRBClev), 1 May 1998

The Measurement and Determination of Okun's Law: Evidence from the State Economies, P. R. Blackley, Journal of Macroeconomics, 13, Fall 1991, 641-656

Is Low Unemployment Inflationary? Roberto Chang, Economic Review (FRBAtl), 82(1) 1Q1997

Low Unemployment: Old Dogs or New Tricks? Abbigail J. Chiodo and Michael T. Owyang, Regional Economist (FRB StL) Oct 2001

The Erosion of Okun's Law, Bradford J. DeLong, 3 Sep 2003 04:22 PM

History and Theory of the NAIRU: A Critical Review Marco A. Espinosa-Vega and Steven Russell, Economic Review (FRBAtl), 82(2) 2Q1997

Labor and Output Over the Business Cycle: Some Direct Evidence, Jon A. Fay and James L. Medoff. American Economic Review, LXXV(4) Sep 1985: 638-55

The Underemployment of Minorities Alan Greenspan, Wall Street Project Anniversary Conference of the Rainbow/PUSH Coalition, New York, New York, 16 Jan 1998

Has Structural Change Contributed to a Jobless Recovery? Erica L. Groshen and Simon Potter, Current Issues in Economics and Finance, 9(8) Aug 2003

Is Unemployment a Macroeconomic Problem? Robert E. Hall, American Economic Review, LXXIII(2) May 1983: 219-22

Unemployment and Monetary Policy Friedrich A. Hayek, San Francisco: Cato Institute, 1979

Counting the Jobless A Job in Itself, Robert D., Hershey, Jr., New York Times, 4 Nov 1988: D(1, 4)

There are two Okun's relationships between output and unemployment, B. Humberto and F. Howland, 1993

Growth and Unemployment - Okun's Law, John S. Irons, 7 Dec 1999 02:34 PM

Jobs Creation and Government Policy, Jerry L. Jordan, Economic Commentary (FRBClev), 1 Oct 1996

NAIRU: Is It Useful for Monetary Policy? John P. Judd, Economic Letter (FRBSF), 97(35) 21 Nov 1997

An International Comparison of Okun's Laws, R.T. Kaufman, Journal of Comparative Economics, 12, June 1988, 182-203

A Loanable Funds Theory of Unemployment and Monetary Disequilibrium, Meir Kohn, American Economic Review, LXXI(5) Dec 1981: 859-79

Macroeconomics of Stagflation under Flexible Exchange Rates, Pentti J. Kouri, American Economic Review, LXXII(2) May 1982: 390-95

Explaining Unemployment: Sectoral vs. Aggregate Shocks Prakash Loungani and Bharat Trehan, Economic Review (FRBSF), 97(1) 1997

Job Creation and Destruction Prakash Loungani and Bharat Trehan, Economic Letter (FRBSF), 97(13) 2 May 1997

Bugaboo of Joblessness: The Unemployment Rate is a Poor Guide to Public Policy, Frank Maurer, Barron's, 26 Aug 1974: 7, 8

Potential GNP: Its measurement and significance, Arthur M. Okun, Proceedings of the Business and Economics Section of the American Statistical Association, 1962, 98-103

The Political Economy of Prosperity, Arthur M. Okun, Brookings Institution, Washington, DC, 1970

Prices and Quantities: A Macroeconomic Analysis, Arthur M. Okun, Blackwell, Oxford, 1981, 228

Unemployment, the Allocation of Labor, and Optimal Government Intervention, Donald O. Parsons, American Economic Review, LXX(4) Sep 1980: 626-34

Potential GNP: Its Measurement and Significance -- A Dissenting Opinion, Charles I. Plosser and G. William Schwert, Carnegie-Rochester Conference Series on Public Policy (supplement to Journal of Monetary Economics), 10, Spring 1979, 179-186

Okun's law: theoretical foundations and revised estimates, M.F.J. Prachowny, Review of Economics and Statistics, 75, 1993, 331-336

The Empirical Okun Model, Kevin Quinn, Prentice-Hall, Inc., Upper Saddle River, NJ, 2002

Employment Surveys Are Telling the Same (Sad) Story, Mark Schweitzer and Guhan Venkatu, Economic Commentary (FRBStL), 15 May 2004

Unemployment Duration and Incidence, Hal Sider, American Economic Review, LXXV(3) Jun 1985: 461-72

On Theories of Unemployment, Robert M. Solow, American Economic Review, LXX(1) Mar 1980: 1-11

America's Army of Non-Workers? Louis Uchitelle, New York Times, 27 Sep 1987: (3)1

The Effects of Industry Employment Shifts on the U.S. Wage Structure, 1979-1995, Robert G. Valletta, Economic Review (FRBSF). 97(1) 1997

top    Websites

Overview of BLS Statistics on Employment and Unemployment
Employment Situation Summary, BLS, current report
Employment Situation Explanatory Note, BLS, current report
Handbook of Methods, BLS survey methodology