24729 00 front end sheets - University of...

$: 24729 00 front end sheets - University of Phoenixmyresource.phoenix.edu/secure/resource/XECO212r4/xeco212_r4_w… · A country that saves and invests a high fraction of its income,$
Unemployment

Losing a job can be the most distressing economic event in a person’s life. Mostpeople rely on their labor earnings to maintain their standard of living, andmany people get from their work not only income but also a sense of personalaccomplishment. A job loss means a lower living standard in the present, anxietyabout the future, and reduced self-esteem. It is not surprising, therefore, thatpoliticians campaigning for office often speak about how their proposed policieswill help create jobs.

In previous chapters, we have seen some of the forces that determine the leveland growth of a country’s standard of living. A country that saves and invests ahigh fraction of its income, for instance, enjoys more rapid growth in its capitalstock and its GDP than a similar country that saves and invests less. An evenmore obvious determinant of a country’s standard of living is the amount ofunemployment it typically experiences. People who would like to work but can-not find a job are not contributing to the economy’s production of goods andservices. Although some degree of unemployment is inevitable in a complexeconomy with thousands of firms and millions of workers, the amount of unem-ployment varies substantially over time and across countries. When a countrykeeps its workers as fully employed as possible, it achieves a higher level ofGDP than it would if it left many of its workers standing idle.

28

613

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 613

9780324832945, Principles of Economics, 4e, N. Gregory Mankiw - © Cengage Learning

This chapter begins our study of unemployment. The problem of unemploy-ment is usefully divided into two categories: the long-run problem and the short-run problem. The economy’s natural rate of unemployment refers to the amount ofunemployment that the economy normally experiences. Cyclical unemploymentrefers to the year-to-year fluctuations in unemployment around its natural rate,and it is closely associated with the short-run ups and downs of economic activ-ity. Cyclical unemployment has its own explanation, which we defer until westudy short-run economic fluctuations later in this book. In this chapter, we dis-cuss the determinants of an economy’s natural rate of unemployment. As wewill see, the designation natural does not imply that this rate of unemploymentis desirable. Nor does it imply that it is constant over time or impervious to eco-nomic policy. It merely means that this unemployment does not go away on itsown even in the long run.

We begin the chapter by looking at some of the relevant facts that describeunemployment. In particular, we examine three questions: How does the gov-ernment measure the economy’s rate of unemployment? What problems arise ininterpreting the unemployment data? How long are the unemployed typicallywithout work?

We then turn to the reasons economies always experience some unemploy-ment and the ways in which policymakers can help the unemployed. We discussfour explanations for the economy’s natural rate of unemployment: job search,minimum-wage laws, unions, and efficiency wages. As we will see, long-rununemployment does not arise from a single problem that has a single solution.Instead, it reflects a variety of related problems. As a result, there is no easy wayfor policymakers to reduce the economy’s natural rate of unemployment and, atthe same time, to alleviate the hardships experienced by the unemployed.

614 PART 9 THE REAL ECONOMY IN THE LONG RUN

IDENTIFYING UNEMPLOYMENTLet’s start by examining more precisely what the term unemployment means.

How Is Unemployment Measured?

Measuring unemployment is the job of the Bureau of Labor Statistics (BLS),which is part of the Department of Labor. Every month, the BLS produces dataon unemployment and on other aspects of the labor market, such as types ofemployment, length of the average workweek, and the duration of unemploy-ment. These data come from a regular survey of about 60,000 households, calledthe Current Population Survey.

Based on the answers to survey questions, the BLS places each adult (aged 16and older) in each surveyed household into one of three categories:

• Employed: This category includes those who worked as paid employees,worked in their own business, or worked as unpaid workers in a familymember’s business. It also includes those who were not working but who hadjobs from which they were temporarily absent because of, for example, vaca-tion, illness, or bad weather.

• Unemployed: This category includes those who were not employed, wereavailable for work, and had tried to find employment during the previous 4weeks. It also includes those waiting to be recalled to a job from which theyhad been laid off.

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 614


• Not in the labor force: This category includes those who fit neither of the firsttwo categories, such as a full-time student, homemaker, or retiree.

Figure 1 shows the breakdown into these categories for 2004.Once the BLS has placed all the individuals covered by the survey in a category,

it computes various statistics to summarize the state of the labor market. The BLSdefines the labor force as the sum of the employed and the unemployed:

Labor force � Number of employed � Number of unemployed.

The BLS defines the unemployment rate as the percentage of the labor force thatis unemployed:

Number of unemployedUnemployment rate � � 100.

Labor force

The BLS computes unemployment rates for the entire adult population and formore narrowly defined groups such as blacks, whites, men, women, and so on.

The BLS uses the same survey to produce data on labor-force participation.The labor-force participation rate measures the percentage of the total adultpopulation of the United States that is in the labor force:

Labor forceLabor-force participation rate � � 100.

Adult population

CHAPTER 28 UNEMPLOYMENT 615

11F I G U R E

The Breakdown of thePopulation in 2004The Bureau of Labor Statisticsdivides the adult populationinto three categories:employed, unemployed, and not in the labor force.

Source: Bureau of Labor Statistics.Adult

Population(223.4 million)

Labor Force(147.4 million)

Employed(139.3 million)

Not in labor force(76.0 million)

Unemployed (8.1 million)

labor forcethe total number ofworkers, including boththe employed and theunemployed

unemployment ratethe percentage of thelabor force that is unem-ployed

labor-force participationratethe percentage of theadult population that isin the labor force

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 615


This statistic tells us the fraction of the population that has chosen to participatein the labor market. The labor-force participation rate, like the unemploymentrate, is computed both for the entire adult population and for more specificgroups.

To see how these data are computed, consider the figures for 2004. In thatyear, 139.3 million people were employed, and 8.1 million people were unem-ployed. The labor force was

Labor force � 139.3 � 8.1 � 147.4 million.

The unemployment rate was

Unemployment rate � (8.1/147.4) � 100 � 5.5 percent.

Because the adult population was 223.4 million, the labor-force participation ratewas

Labor-force participation rate � (147.4/223.4) � 100 � 66.0 percent.

Hence, in 2004, two-thirds of the U.S. adult population were participating in thelabor market, and 5.5 percent of those labor-market participants were withoutwork.

Table 1 shows the statistics on unemployment and labor-force participation forvarious groups within the U.S. population. Three comparisons are most apparent.First, women ages 20 and older have lower rates of labor-force participation thanmen, but once in the labor force, men and women have similar rates of unem-ployment. Second, blacks ages 20 and older have similar rates of labor-force par-ticipation as whites, but they have much higher rates of unemployment. Third,teenagers have lower rates of labor-force participation and much higher rates ofunemployment than older workers. More generally, these data show that labor-market experiences vary widely among groups within the economy.

The BLS data on the labor market also allow economists and policymakers tomonitor changes in the economy over time. Figure 2 shows the unemploymentrate in the United States since 1960. The figure shows that the economy alwayshas some unemployment and that the amount changes from year to year. Thenormal rate of unemployment around which the unemployment rate fluctuatesis called the natural rate of unemployment, and the deviation of unemployment


11The Labor-MarketExperiences of VariousDemographic GroupsThis table shows the unemploy-ment rate and the labor-forceparticipation rate of variousgroups in the U.S. populationfor 2004.

Source: Bureau of Labor Statistics.

T A B L E Labor-ForceDemographic Group Unemployment Rate Participation Rate

Adults (ages 20 and older)White, male 4.4% 76.2%White, female 4.2 59.7Black, male 9.9 70.9Black, female 8.9 64.2

Teenagers (ages 16–19)White, male 16.3 47.4White, female 13.6 46.7Black, male 35.6 30.0Black, female 28.2 32.8

natural rate ofunemploymentthe normal rate ofunemployment aroundwhich the unemploy-ment rate fluctuates

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 616


from its natural rate is called cyclical unemployment. The natural rate of unem-ployment shown in the figure is a series estimated by economists at the Congres-sional Budget Office. For 2004, they estimated a natural rate of 5.2, only slightlylower than the actual unemployment rate of 5.5 percent. Later in this book, wediscuss short-run economic fluctuations, including the year-to-year fluctuationsin unemployment around its natural rate. In the rest of this chapter, however, weignore the short-run fluctuations and examine why there is always some unem-ployment in market economies.

LABOR-FORCE PARTICIPATION OF MEN AND WOMEN IN THE U.S. ECONOMY

Women’s role in American society has changed dramatically over the past cen-tury. Social commentators have pointed to many causes for this change. In part,it is attributable to new technologies, such as the washing machine, clothesdryer, refrigerator, freezer, and dishwasher, which have reduced the amount oftime required to complete routine household tasks. In part, it is attributable toimproved birth control, which has reduced the number of children born to thetypical family. This change in women’s role is also partly attributable to chang-ing political and social attitudes, which in turn may have been facilitated by the


22F I G U R E

Unemployment Ratesince 1960

10%

8

6

4

2

01970 19751960 1965 1980 1985 1990 2005

Percent ofLabor Force

1995 2000

Natural rate ofunemployment

Unemployment rate

This graph uses annual data on the U.S. unemployment rate to show thepercentage of the labor force without a job. The natural rate of unemploy-ment is the normal level of unemployment around which the unemploy-ment rate fluctuates.

Source: U.S. Department of Labor, Congressional Budget Office.

cyclical unemploymentthe deviation of unem-ployment from its naturalrate

C A S ESTUDY

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 617



advances in technology and birth control. Together these developments havehad a profound impact on society in general and on the economy in particular.

Nowhere is that impact more obvious than in data on labor-force participa-tion. Figure 3 shows the labor-force participation rates of men and women in theUnited States since 1950. Just after World War II, men and women had very dif-ferent roles in society. Only 33 percent of women were working or looking forwork, in contrast to 87 percent of men. Over the past several decades, the differ-ence between the participation rates of men and women has gradually dimin-ished, as growing numbers of women have entered the labor force and somemen have left it. Data for 2004 show that 59 percent of women were in the laborforce, in contrast to 73 percent of men. As measured by labor-force participation,men and women are now playing a more equal role in the economy.

The increase in women’s labor-force participation is easy to understand, butthe fall in men’s may seem puzzling. There are several reasons for this decline.First, young men now stay in school longer than their fathers and grandfathersdid. Second, older men now retire earlier and live longer. Third, with morewomen employed, more fathers now stay at home to raise their children. Full-time students, retirees, and stay-at-home fathers are all counted as out of thelabor force. •

Does the Unemployment Rate Measure What We Want It To?

Measuring the amount of unemployment in the economy might seem a straight-forward task, but it is not. While it is easy to distinguish between a person with

33Labor-Force Participation Rates forMen and Women since 1950

Source: U.S. Department of Labor.

F I G U R E

100

80

60

40

20

01950 1955 1960 1965 1970 1975 1980 1985 1990 2000 2005

Labor-ForceParticipation

Rate (in percent)

Women

Men

1995

This figure shows the percentage of adult men and women who aremembers of the labor force. It shows that over the past severaldecades, women have entered the labor force, and men have left it.

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 618


22T A B L E

Alternative Measures ofLabor UnderutilizationThe table shows various mea-sures of joblessness for theU.S. economy. The data arefor July 2005.

Source: U.S. Department of Labor.

Measure and Description Rate

U-1 Persons unemployed 15 weeks or longer, as a percent of the 1.6%civilian labor force (includes only very long-term unemployed)

U-2 Job losers and persons who have completed temporary jobs, 2.4as a percent of the civilian labor force (excludes job leavers)

U-3 Total unemployed, as a percent of the civilian labor force 5.0(official unemployment rate)

U-4 Total unemployed, plus discouraged workers, as a percent 5.3of the civilian labor force plus discouraged workers

U-5 Total unemployed plus all marginally attached workers, as a percent 6.0of the civilian labor force plus all marginally attached workers

U-6 Total unemployed, plus all marginally attached workers, plus total 8.9employed part-time for economic reasons, as a percent of the civilian labor force plus all marginally attached workers

Note: The Bureau of Labor Statistics defines terms as follows.• Marginally attached workers are persons who currently are neither working nor

looking for work but indicate that they want and are available for a job and havelooked for work sometime in the recent past.

• Discouraged workers are marginally attached workers who have given a job-marketrelated reason for not currently looking for a job.

• Persons employed part-time for economic reasons are those who want and areavailable for full-time work but have had to settle for a part-time schedule.

a full-time job and a person who is not working at all, it is much harder to dis-tinguish between a person who is unemployed and a person who is not in thelabor force.

Movements into and out of the labor force are, in fact, common. More thanone-third of the unemployed are recent entrants into the labor force. Theseentrants include young workers looking for their first jobs, such as recent collegegraduates. They also include, in greater numbers, older workers who had previ-ously left the labor force but have now returned to look for work. Moreover, notall unemployment ends with the job seeker finding a job. Almost half of all spellsof unemployment end when the unemployed person leaves the labor force.

Because people move into and out of the labor force so often, statistics onunemployment are difficult to interpret. On the one hand, some of those whoreport being unemployed may not, in fact, be trying hard to find a job. They maybe calling themselves unemployed because they want to qualify for a govern-ment program that financially assists the unemployed or because they are actu-ally working and paid “under the table.” It may be more realistic to view theseindividuals as out of the labor force or, in some cases, employed. On the otherhand, some of those who report being out of the labor force may want to work.These individuals may have tried to find a job and may have given up after anunsuccessful search. Such individuals, called discouraged workers, do not showup in unemployment statistics, even though they are truly workers without jobs.

Because of these and other problems, the BLS calculates several other mea-sures of labor underutilization, in addition to the official unemployment rate.These alternative measures are presented in Table 2. In the end, it is best to viewthe official unemployment rate as a useful but imperfect measure of joblessness.


discouraged workersindividuals who wouldlike to work but havegiven up looking for ajob

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 619


How Long Are the Unemployed without Work?

In judging how serious the problem of unemployment is, one question to con-sider is whether unemployment is typically a short-term or long-term condition.If unemployment is short-term, one might conclude that it is not a big problem.Workers may require a few weeks between jobs to find the openings that bestsuit their tastes and skills. Yet if unemployment is long-term, one might con-clude that it is a serious problem. Workers unemployed for many months aremore likely to suffer economic and psychological hardship.

Because the duration of unemployment can affect our view about how big aproblem unemployment is, economists have devoted much energy to studyingdata on the duration of unemployment spells. In this work, they have uncovered a


In The NewsThe Rise of Adult Male JoblessnessAn increasing number of men are neither working nor lookingfor work.

A Growing Number ofMen Are Not Working, SoWhat Are They Doing?By Alan B. Krueger

A growing number of men in their primeworking years are pursuing what mightbe called the Kramer lifestyle, after theenigmatic ‘’Seinfeld’’ character: neitherworking nor attending school. In 1967,2.2 percent of noninstitutionalized menage 25 to 54 spent the entire year with-out working for pay or attending school.That figure climbed to 8 percent in2002, the latest year available from theBureau of Labor Statistics.

This trend is partly related to the ris-ing disability rolls. More than half ofmale nonworkers reported themselvesas sick or disabled. But the number oflong-term jobless men who were able-

bodied—a diverse group includingyoung retirees, men who cannot findwork, and family care providers—grewat a faster rate than the number whowere disabled over the last 35 years.

The problem is much more severe forsome groups than others. Nearly one infive men age 25 to 54 with less than ahigh school degree did not work evenone week in 2002. The nonworking ratefor college graduates was only 3.3 per-cent. In central cities, 10.8 percent ofmen spent the year without work, com-pared with 7.1 percent elsewhere.

Joblessness is persistent over time,so it ends up being highly concentratedamong a small cadre of men who fre-quently spend long stretches withoutwork. Just 3 percent of men accountedfor more than two-thirds of the totalnumber of years that men spent not

working in the period from 1987 to1997, according to an analysis by JayStewart, an economist at the Bureau ofLabor Statistics.

PH

OTO

:©N

BC

TV

/TH

E K

OB

AL

CO

LLE

CT

ION

“YOU WOULDN’T CATCH ME IN THE

LABOR FORCE.”

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 620


result that is important, subtle, and seemingly contradictory: Most spells of unem-ployment are short, and most unemployment observed at any given time is long-term.

To see how this statement can be true, consider an example. Suppose that youvisited the government’s unemployment office every week for a year to surveythe unemployed. Each week you find that there are four unemployed workers.Three of these workers are the same individuals for the whole year, while thefourth person changes every week. Based on this experience, would you say thatunemployment is typically short-term or long-term?

Some simple calculations help answer this question. In this example, you meeta total of 55 unemployed people; 52 of them are unemployed for 1 week, and 3are unemployed for the full year. This means that 52/55, or 95 percent, of unem-ployment spells end in 1 week. Yet whenever you walk into the unemployment


Long-term joblessness among ma-ture men has become a much more im-portant phenomenon than unemploy-ment. Many jobless men do not activelysearch for work, so they are not countedas unemployed. Yet they still represent asignificant loss of productive human re-sources for the economy.

The conventional wisdom is that job-lessness has grown since the early1980’s because the demand for less-skilled workers has dropped, causingtheir pay to fall. The decline in unionsand erosion of the real value of the min-imum wage have also caused their payto fall. Rather than toil at low pay, moreand more men have withdrawn from thejob market.

How are these men spending theirtime and getting by?

A new working paper by Mr. Stewartof the Labor Bureau provides the mostcomprehensive answers to date. Thestudy, ‘’What Do Male NonworkersDo?’’, draws on information from severalnational data sets on the time alloca-tion, living arrangements and incomesources of male nonworkers in theirprime earning years. . . .

In short, the average day of a malenonworker looks very much like the av-

Source: The New York Times, April 29, 2004.

erage day of a worker—on his day off.Nonworkers devoted 8.4 hours a day toleisure and recreation and 3.3 hours tohousework. On their days off, workersdevoted almost the same amount oftime—8 and 3.4 hours, respectively—tothese activities.

On workdays, the average full-timeworker devoted only 3.5 hours to leisureand recreation and one hour to house-work. Men worked an average of 8.6hours on days when they performedsome work for pay.

Comparing workers and nonworkersover a full week, nonworkers spentabout a quarter of their extra time in‘’home production,’’ which includeshousehold chores, cleaning and repairs.The bulk of their extra time went intoleisure and recreation, particularlywatching television, socializing andplaying sports and games. Nonworkersalso slept 10 percent more (44 minutes)a night than workers. Both groups de-voted relatively little time to child care,at least as a primary activity.

By contrast, nonworking womenspend half their extra time engaged inhousehold work and child care.

Supporting a Kramer lifestyle is noteasy, especially if your neighbors are

less magnanimous than Jerry Seinfeld.Nearly two-thirds of nonworking menage 25 to 54 received income fromsome source in 2002. Among those withunearned income, the average amountwas $11,551, with the largest sumscoming from Social Security and disabil-ity payments. . . .

Not surprisingly, wives are also an im-portant source of financial support fornonworking men, but only 42 percent ofmale nonworkers between age 25 and 54are married, compared with 68 percent oftheir employed counterparts. Twenty-ninepercent of nonworkers live with theirparents or other relatives, substantiallyhigher than the 9 percent of workers insuch a living arrangement. . . .

The experiences of nonworking adultmen are quite varied, and many havesevere disabilities. Although these statis-tics paint a picture of nonworking menstruggling to get by financially, manymanage to live as if every day were Sun-day. As one man from Brooklyn who hasnot worked since 1998 told me thisweek, he thinks of the Off-Track Bettingparlor in Midtown Manhattan as his‘’club,’’ and he sees many of the samemen there day after day.

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 621


office, three of the four people you meet will be unemployed for the entire year.So, even though 95 percent of unemployment spells end in 1 week, 75 percent ofunemployment is attributable to those individuals who are unemployed for a fullyear. In this example, as in the world, most spells of unemployment are short,and most unemployment observed at any given time is long-term.

This subtle conclusion implies that economists and policymakers must becareful when interpreting data on unemployment and when designing policiesto help the unemployed. Most people who become unemployed will soon findjobs. Yet most of the economy’s unemployment problem is attributable to the rel-atively few workers who are jobless for long periods of time.

Why Are There Always Some People Unemployed?

We have discussed how the government measures the amount of unemploy-ment, the problems that arise in interpreting unemployment statistics, and thefindings of labor economists on the duration of unemployment. You should nowhave a good idea about what unemployment is.

This discussion, however, has not explained why economies experienceunemployment. In most markets in the economy, prices adjust to bring quantitysupplied and quantity demanded into balance. In an ideal labor market, wageswould adjust to balance the quantity of labor supplied and the quantity of labordemanded. This adjustment of wages would ensure that all workers are alwaysfully employed.

Of course, reality does not resemble this ideal. There are always some work-ers without jobs, even when the overall economy is doing well. In other words,the unemployment rate never falls to zero; instead, it fluctuates around the natu-ral rate of unemployment. To understand this natural rate, the remaining sec-tions of this chapter examine the reasons actual labor markets depart from theideal of full employment.

To preview our conclusions, we will find that there are four ways to explainunemployment in the long run. The first explanation is that it takes time forworkers to search for the jobs that are best suited for them. The unemploymentthat results from the process of matching workers and jobs is sometimes calledfrictional unemployment, and it is often thought to explain relatively shortspells of unemployment.

The next three explanations for unemployment suggest that the number ofjobs available in some labor markets may be insufficient to give a job to every-one who wants one. This occurs when the quantity of labor supplied exceeds thequantity demanded. Unemployment of this sort is sometimes called structuralunemployment, and it is often thought to explain longer spells of unemploy-ment. As we will see, this kind of unemployment results when wages are, forsome reason, set above the level that brings supply and demand into equilib-rium. We will examine three possible reasons for an above-equilibrium wage:minimum-wage laws, unions, and efficiency wages.

How is the unemployment rate measured? • How might theunemployment rate overstate the amount of joblessness? How might it understate theamount of joblessness?


frictionalunemploymentunemployment thatresults because it takestime for workers tosearch for the jobs thatbest suit their tastes andskills

structuralunemploymentunemployment thatresults because the num-ber of jobs available insome labor markets isinsufficient to provide ajob for everyone whowants one

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 622



JOB SEARCH One reason economies always experience some unemployment is job search. Jobsearch is the process of matching workers with appropriate jobs. If all workersand all jobs were the same, so that all workers were equally well suited for alljobs, job search would not be a problem. Laid-off workers would quickly findnew jobs that were well suited for them. But in fact, workers differ in their tastesand skills, jobs differ in their attributes, and information about job candidatesand job vacancies is disseminated slowly among the many firms and householdsin the economy.

Why Some Frictional Unemployment Is Inevitable

Frictional unemployment is often the result of changes in the demand for laboramong different firms. When consumers decide that they prefer Dell to Gatewaycomputers, Dell increases employment, and Gateway lays off workers. The for-mer Gateway workers must now search for new jobs, and Dell must decidewhich new workers to hire for the various jobs that have opened up. The resultof this transition is a period of unemployment.

Similarly, because different regions of the country produce different goods,employment can rise in one region while it falls in another. Consider, forinstance, what happens when the world price of oil falls. Oil-producing firms inTexas respond to the lower price by cutting back on production and employ-ment. At the same time, cheaper gasoline stimulates car sales, so auto-producingfirms in Michigan raise production and employment. Changes in the composi-tion of demand among industries or regions are called sectoral shifts. Because ittakes time for workers to search for jobs in the new sectors, sectoral shifts tem-porarily cause unemployment.

Frictional unemployment is inevitable simply because the economy is alwayschanging. A century ago, the four industries with the largest employment in theUnited States were cotton goods, woolen goods, men’s clothing, and lumber.Today, the four largest industries are autos, aircraft, communications, and elec-trical components. As this transition took place, jobs were created in some firmsand destroyed in others. The end result of this process has been higher produc-tivity and higher living standards. But along the way, workers in decliningindustries found themselves out of work and searching for new jobs.

Data show that at least 10 percent of U.S. manufacturing jobs are destroyedevery year. In addition, more than 3 percent of workers leave their jobs in a typ-ical month, sometimes because they realize that the jobs are not a good matchfor their tastes and skills. Many of these workers, especially younger ones, findnew jobs at higher wages. This churning of the labor force is normal in a well-functioning and dynamic market economy, but the result is some amount of fric-tional unemployment.

Public Policy and Job Search

Even if some frictional unemployment is inevitable, the precise amount is not.The faster information spreads about job openings and worker availability, themore rapidly the economy can match workers and firms. The Internet, forinstance, may help facilitate job search and reduce frictional unemployment. In

job searchthe process by whichworkers find appropriatejobs given their tastesand skills

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 623


addition, public policy may play a role. If policy can reduce the time it takesunemployed workers to find new jobs, it can reduce the economy’s natural rateof unemployment.

Government programs try to facilitate job search in various ways. One way isthrough government-run employment agencies, which give out informationabout job vacancies. Another way is through public training programs, whichaim to ease the transition of workers from declining to growing industries andto help disadvantaged groups escape poverty. Advocates of these programsbelieve that they make the economy operate more efficiently by keeping thelabor force more fully employed and that they reduce the inequities inherent in aconstantly changing market economy.


In The NewsGerman UnemploymentTraditionally, many European countries have had unemploymentinsurance that is far more generous than that offered to U.S.workers. But some are starting to rethink that policy.

In a Deep Crisis, GermanyBegins to Revamp Its VastWelfare StateBy Christopher Rhoads

Essen, Germany—Four years ago, theGerman social-welfare system rescuedRenate Franke.

Soon after her mother died of cancerthat year, her son was diagnosed withthe same disease. Emotionally drained,Ms. Franke, then 48 years old, quit herjob at an electronics company. The statestepped in, sending her to a spa forthree weeks and paying her jobless ben-efits that were close to 60% of her for-mer wage.

But last year, the state got toughwith Ms. Franke. It cut her unemploy-ment aid after she refused to take full-time jobs. It told her to sell her car, as acondition for receiving any further socialassistance. Sitting one recent morning in

her one-bedroom apartment on the out-skirts of this industrial city, she said: “Ibegan to fear for my future.”

So, too, does Germany. Faced with itsworst economic slump since World WarII, Germany is beginning to broach somelong-held taboos as it comes to termswith a cold reality: The country’s eco-nomic system doesn’t work anymore.The world’s third-largest economy afterthe U.S. and Japan has slid into its sec-ond recession within the past threeyears—making it the weakest of theworld’s major economies. Unemploy-ment is hovering at nearly 11%.

The dire conditions are prompting anunprecedented rethinking of the pater-nalistic role of the state in the country’seconomic life. In the decades since thewar, West Germany and then unitedGermany had to deal with a cata-strophic legacy of military and moral de-feat. One source of pride, however, re-

mained constant: the country’s eco-nomic power tethered to a strong social-welfare system.

Now, the country’s downward spiralhas made this model no longer afford-able. Social spending has reached closeto 30% of gross domestic product, themost of any country in the world exceptfor Sweden and more than twice that ofthe U.S. . . .

Within the past few months, Berlin’scenter-left government has proposed re-ducing unemployment benefits, openingthe public health-care system to privateinsurers, cutting hundreds of millions ofdollars of subsidies and easing laws thatprotect workers from being fired. Latelast month, the government brought for-ward by a year a planned tax cut. Someministers even want to cut back on thecountry’s famously large amount of freetime—30 vacation days on average,compared with 12 in the U.S. . . .

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 624


Critics of these programs question whether the government should getinvolved with the process of job search. They argue that it is better to let the pri-vate market match workers and jobs. In fact, most job search in our economytakes place without intervention by the government. Newspaper ads, Internetjob sites, college placement offices, headhunters, and word of mouth all helpspread information about job openings and job candidates. Similarly, muchworker education is done privately, either through schools or through on-the-jobtraining. These critics contend that the government is no better—and most likelyworse—at disseminating the right information to the right workers and decidingwhat kinds of worker training would be most valuable. They claim that thesedecisions are best made privately by workers and employers.


Declining birth rates, longer life-spans, earlier retirement ages, less work-ing time and steadily higher unemploy-ment mean that those paying into thesystem can no longer support those liv-ing off of it. Since 1970, the total num-ber of pensioners and jobless increasedby 80% to 16.3 million in western Ger-many. The number of workers, who to-gether with employers finance the sys-tem through payroll taxes, grew by just4% in that time, to 30.7 million.

In the early part of the 20th centurymost social needs, such as care for chil-dren and the elderly, were handled bythe family across Europe. After WorldWar II, much of the responsibility shiftedto the state, and the social-welfare sys-tem began to grow exponentially. Dri-ving the expansion were Europeangrowth rates of 5% or more, spurred bythe rebuilding after the destruction ofthe war. Jobs were so plentiful thatcountries had to import workers fromTurkey and elsewhere. Birth ratessoared, which meant that there wereplenty of young workers to finance theretirement needs of the relatively fewelderly.

Europe could afford to be generous.In the 1960s and ‘70s, the social role of

Source: The Wall Street Journal, July 10, 2003. Copyright 2003 by DOW JONES & CO INC. Reproduced with permission of DOW JONES &CO INC. in the format Textbook via Copyright Clearance Center.

governments grew well beyond caringfor the needy to providing health care,higher-level education, child and old-agecare—for all citizens. “The instrumentsof redistribution got hijacked by themiddle class,” says Dennis Snower, aprofessor of international economics atBirkbeck College, at the University ofLondon. That meant that reversing direc-tion became increasingly difficult, sincean ever-larger number of citizens—andvoters—benefited from the system.

While at the beginning of the 1960ssocial spending in Europe was onlyslightly higher than in the U.S., by theend of the 1990s, it was twice as much.“Americans are not as obsessed with so-cial insurance because they think if theywork hard they will get rich,” saysRobert MacCulloch, a professor of eco-nomics at Princeton University. And theythink that once they get rich, they won’twant to be burdened with high taxes tocover welfare costs. In Europe, many feeltheir chances of improving their lot arelower, increasing their appetite for assis-tance from the state, he says. Europeansalso favor income equality more thanAmericans, surveys show.

But in some ways, the massivegrowth of the welfare state has led to

just the opposite of what was intended:a less-integrated society through thecreation of a growing underclass oflong-term unemployed. More than 50%of Germany’s 4.5 million unemployedhave been without a job for a year ormore, making them less employable andmore wedded to welfare. The equivalentfigure in the U.S. is 6%. . . .

In an effort to prevent the newly job-less from becoming permanently unem-ployed, the government is becoming lessforgiving. As of this year, singles mustmove anywhere in the country to fillopen positions or lose benefits. Othermeasures include cutting benefits whena jobless person refuses to accept anopening and enforcing participation instate-sponsored temp agencies.

For Ms. Franke, the former worker inthe electronics firm, the tightening ofthe screws made a difference. She hadworked several off-the-books part-timesecretarial jobs to augment her welfarepayments—a common practice. Butwith her state assistance dwindling, inrecent months she began looking moreactively for full-time work. In May, shelanded a job in charge of bookkeepingfor another electronics company nearby.

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 625


Unemployment Insurance

One government program that increases the amount of frictional unemploy-ment, without intending to do so, is unemployment insurance. This program isdesigned to offer workers partial protection against job loss. The unemployedwho quit their jobs, were fired for cause, or just entered the labor force are noteligible. Benefits are paid only to the unemployed who were laid off becausetheir previous employers no longer needed their skills. Although the terms ofthe program vary over time and across states, a typical American worker cov-ered by unemployment insurance receives 50 percent of his or her former wagesfor 26 weeks.

While unemployment insurance reduces the hardship of unemployment, italso increases the amount of unemployment. The explanation is based on one ofthe Ten Principles of Economics in Chapter 1: People respond to incentives.Because unemployment benefits stop when a worker takes a new job, the unem-ployed devote less effort to job search and are more likely to turn down unat-tractive job offers. In addition, because unemployment insurance makes unem-ployment less onerous, workers are less likely to seek guarantees of job securitywhen they negotiate with employers over the terms of employment.

Many studies by labor economists have examined the incentive effects of unem-ployment insurance. One study examined an experiment run by the state of Illi-nois in 1985. When unemployed workers applied to collect unemployment insur-ance benefits, the state randomly selected some of them and offered each a $500bonus if they found new jobs within 11 weeks. This group was then compared to acontrol group not offered the incentive. The average spell of unemployment forthe group offered the bonus was 7 percent shorter than the average spell for thecontrol group. This experiment shows that the design of the unemployment insur-ance system influences the effort that the unemployed devote to job search.

Several other studies examined the search effort by following a group ofworkers over time. Unemployment insurance benefits, rather than lasting for-ever, usually run out after 6 months or 1 year. These studies found that when theunemployed become ineligible for benefits, the probability of their finding a newjob rises markedly. Thus, receiving unemployment insurance benefits doesreduce the search effort of the unemployed.

Even though unemployment insurance reduces search effort and raises unem-ployment, we should not necessarily conclude that the policy is a bad one. Theprogram does achieve its primary goal of reducing the income uncertainty thatworkers face. In addition, when workers turn down unattractive job offers, theyhave the opportunity to look for jobs that better suit their tastes and skills. Someeconomists have argued that unemployment insurance improves the ability ofthe economy to match each worker with the most appropriate job.

The study of unemployment insurance shows that the unemployment rate isan imperfect measure of a nation’s overall level of economic well-being. Mosteconomists agree that eliminating unemployment insurance would reduce theamount of unemployment in the economy. Yet economists disagree on whethereconomic well-being would be enhanced or diminished by this change in policy.

How would an increase in the world price of oil affect the amount offrictional unemployment? Is this unemployment undesirable? What public policiesmight affect the amount of unemployment caused by this price change?


unemploymentinsurancea government programthat partially protectsworkers’ incomes when they becomeunemployed

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 626



MINIMUM-WAGE LAWSHaving seen how frictional unemployment results from the process of matchingworkers and jobs, let’s now examine how structural unemployment resultswhen the number of jobs is insufficient for the number of workers.

To understand structural unemployment, we begin by reviewing how unem-ployment arises from minimum-wage laws. Although minimum wages are notthe predominant reason for unemployment in our economy, they have an impor-tant effect on certain groups with particularly high unemployment rates. More-over, the analysis of minimum wages is a natural place to start because, as wewill see, it can be used to understand some of the other reasons for structuralunemployment.

Figure 4 reviews the basic economics of a minimum wage. When a minimum-wage law forces the wage to remain above the level that balances supply anddemand, it raises the quantity of labor supplied and reduces the quantity oflabor demanded compared to the equilibrium level. There is a surplus of labor.Because there are more workers willing to work than there are jobs, some work-ers are unemployed.

While minimum-wage laws are one reason there is always some unemploy-ment in the U.S. economy, they do not affect everyone. Most workers have wageswell above the legal minimum, so the law does not prevent the wage from adjust-ing to balance supply and demand. Minimum-wage laws matter most for theleast skilled and least experienced members of the labor force, such as teenagers.Their equilibrium wages tend to be low and, therefore, are more likely to fallbelow the legal minimum. It is only among these workers that minimum-wagelaws explain the existence of unemployment.

Although Figure 4 is drawn to show the effects of a minimum-wage law, italso illustrates a more general lesson: If the wage is kept above the equilibrium level

44F I G U R E

Unemployment from a Wageabove the Equilibrium LevelIn this labor market, the wage atwhich supply and demand balanceis WE. At this equilibrium wage,the quantity of labor supplied andthe quantity of labor demandedboth equal LE. By contrast, if thewage is forced to remain abovethe equilibrium level, perhapsbecause of a minimum-wage law,the quantity of labor suppliedrises to LS, and the quantity oflabor demanded falls to LD. Theresulting surplus of labor, LS – LD,represents unemployment.

WE

Quantity ofLabor

LE0

Surplus of labor � Unemployment

Laborsupply

Labordemand

Wage

Minimumwage

LD LS

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 627


for any reason, the result is unemployment. Minimum-wage laws are just one rea-son wages may be “too high.” In the remaining two sections of this chapter, weconsider two other reasons wages may be kept above the equilibrium level:unions and efficiency wages. The basic economics of unemployment in thesecases is the same as that shown in Figure 4, but these explanations of unemploy-ment can apply to many more of the economy’s workers.

At this point, however, we should stop and notice that the structural unem-ployment that arises from an above-equilibrium wage is, in an important sense,different from the frictional unemployment that arises from the process of jobsearch. The need for job search is not due to the failure of wages to balance laborsupply and labor demand. When job search is the explanation for unemploy-ment, workers are searching for the jobs that best suit their tastes and skills. Bycontrast, when the wage is above the equilibrium level, the quantity of laborsupplied exceeds the quantity of labor demanded, and workers are unemployedbecause they are waiting for jobs to open up.

Draw the supply curve and the demand curve for a labor market inwhich the wage is fixed above the equilibrium level. Show the quantity of labor sup-plied, the quantity demanded, and the amount of unemployment.


FYIWho Earns the Minimum Wage?

In 2005, the De-partment of Laborreleased a study

of what workers reported earnings at or below the minimumwage, which at the time was $5.15 per hour. (A reported wagebelow the minimum is possible because some workers areexempt from the statute, because enforcement is imperfect, andbecause some workers round down to $5.00 when reportingtheir wages on surveys.) Here is a summary of the findings:

• Of those workers paid an hourly rate, about 2 percent ofmen and 4 percent of women reported wages at or belowthe prevailing federal minimum.

• Minimum-wage workers tend to be young. About half of allhourly paid workers earning $5.15 or less were under age25, and about one-fourth were age 16–19. Amongteenagers, about 9 percent earned $5.15 or less, comparedwith about 2 percent of workers age 25 and older.

• Minimum-wage workers tend to be less educated. Amonghourly paid workers age 16 and older, about 2 percent of

those who had only a high school diploma earned $5.15 orless, compared with about 1 percent for those who hadobtained a college degree.

• Minimum-wage workers are more likely to be working parttime. Among part-time workers (those who usually work lessthan 35 hours per week), 7 percent were paid $5.15 or less,compared to 1 percent of full-time workers.

• The industry with the highest proportion of workers withreported hourly wages at or below $5.15 was leisure andhospitality (about 15 percent). About three-fifths of all work-ers paid at or below the minimum wage were employed inthis industry, primarily in food services and drinking places.For many of these workers, tips supplement the hourlywages received.

• The proportion of hourly paid workers earning the prevailingfederal minimum wage or less has trended downward since1979, when data collection first began on a regular basis.

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 628



UNIONS AND COLLECTIVE BARGAININGA union is a worker association that bargains with employers over wages, bene-fits, and working conditions. Whereas only 13 percent of U.S. workers nowbelong to unions, unions played a much larger role in the U.S. labor market inthe past. In the 1940s and 1950s, when unions were at their peak, about a thirdof the U.S. labor force was unionized. Moreover, unions continue to play a largerole in many European countries. In Sweden and Denmark, for instance, morethan three-fourths of workers belong to unions.

The Economics of Unions

A union is a type of cartel. Like any cartel, a union is a group of sellers actingtogether in the hope of exerting their joint market power. Most workers in theU.S. economy discuss their wages, benefits, and working conditions with theiremployers as individuals. By contrast, workers in a union do so as a group. Theprocess by which unions and firms agree on the terms of employment is calledcollective bargaining.

When a union bargains with a firm, it asks for higher wages, better benefits,and better working conditions than the firm would offer in the absence of aunion. If the union and the firm do not reach agreement, the union can organizea withdrawal of labor from the firm, called a strike. Because a strike reducesproduction, sales, and profit, a firm facing a strike threat is likely to agree to payhigher wages than it otherwise would. Economists who study the effects ofunions typically find that union workers earn about 10 to 20 percent more thansimilar workers who do not belong to unions.

When a union raises the wage above the equilibrium level, it raises the quan-tity of labor supplied and reduces the quantity of labor demanded, resulting inunemployment. Workers who remain employed are better off, but those whowere previously employed and are now unemployed at the higher wage areworse off. Indeed, unions are often thought to cause conflict between differentgroups of workers—between the insiders who benefit from high union wagesand the outsiders who do not get the union jobs.

The outsiders can respond to their status in one of two ways. Some of themremain unemployed and wait for the chance to become insiders and earn the highunion wage. Others take jobs in firms that are not unionized. Thus, when unionsraise wages in one part of the economy, the supply of labor increases in other partsof the economy. This increase in labor supply, in turn, reduces wages in industriesthat are not unionized. In other words, workers in unions reap the benefit of col-lective bargaining, while workers not in unions bear some of the cost.

The role of unions in the economy depends in part on the laws that governunion organization and collective bargaining. Normally, explicit agreementsamong members of a cartel are illegal. If firms that sell a common product wereto agree to set a high price for that product, the agreement would be a “conspir-acy in restraint of trade.” The government would prosecute these firms in civiland criminal court for violating the antitrust laws. By contrast, unions are exemptfrom these laws. The policymakers who wrote the antitrust laws believed thatworkers needed greater market power as they bargained with employers. Indeed,various laws are designed to encourage the formation of unions. In particular,the Wagner Act of 1935 prevents employers from interfering when workers try

“GENTLEMEN, NOTHING

STANDS IN THE WAY OF A

FINAL ACCORD EXCEPT THAT

MANAGEMENT WANTS PROFIT

MAXIMIZATION AND THE

UNION WANTS MORE

MOOLA.”CA

RTO

ON

:©20

02 T

HE

NE

W Y

OR

KE

R C

OLL

EC

TIO

NF

RO

M C

AR

TOO

NB

AN

K.C

OM

.ALL

RIG

HT

S R

ES

ER

VE

D.

uniona worker association thatbargains with employersover wages, benefits,and working conditions

collective bargainingthe process by whichunions and firms agreeon the terms of employ-ment

strikethe organized withdrawalof labor from a firm by aunion

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 629


to organize unions and requires employers to bargain with unions in good faith.The National Labor Relations Board (NLRB) is the government agency thatenforces workers’ right to unionize.

Legislation affecting the market power of unions is a perennial topic of politi-cal debate. State lawmakers sometimes debate right-to-work laws, which giveworkers in a unionized firm the right to choose whether to join the union. In theabsence of such laws, unions can insist during collective bargaining that firmsmake union membership a requirement for employment. Lawmakers in Wash-ington have also at times debated a proposed law that would prevent firms fromhiring permanent replacements for workers who are on strike. This law wouldmake strikes more costly for firms and, thereby, would increase the marketpower of unions. These and similar policy decisions will help determine thefuture of the union movement.

Are Unions Good or Bad for the Economy?

Economists disagree about whether unions are good or bad for the economy as awhole. Let’s consider both sides of the debate.

Critics argue that unions are merely a type of cartel. When unions raise wagesabove the level that would prevail in competitive markets, they reduce the quan-tity of labor demanded, cause some workers to be unemployed, and reduce thewages in the rest of the economy. The resulting allocation of labor is, criticsargue, both inefficient and inequitable. It is inefficient because high union wagesreduce employment in unionized firms below the efficient, competitive level. Itis inequitable because some workers benefit at the expense of other workers.

Advocates contend that unions are a necessary antidote to the market powerof the firms that hire workers. The extreme case of this market power is the“company town,” where a single firm does most of the hiring in a geographicalregion. In a company town, if workers do not accept the wages and workingconditions that the firm offers, they have little choice but to move or stop work-ing. In the absence of a union, therefore, the firm could use its market power topay lower wages and offer worse working conditions than would prevail if ithad to compete with other firms for the same workers. In this case, a union maybalance the firm’s market power and protect the workers from being at themercy of the firm owners.

Advocates of unions also claim that unions are important for helping firmsrespond efficiently to workers’ concerns. Whenever a worker takes a job, theworker and the firm must agree on many attributes of the job in addition to thewage: hours of work, overtime, vacations, sick leave, health benefits, promotionschedules, job security, and so on. By representing workers’ views on theseissues, unions allow firms to provide the right mix of job attributes. Even ifunions have the adverse effect of pushing wages above the equilibrium leveland causing unemployment, they have the benefit of helping firms keep a happyand productive work force.

In the end, there is no consensus among economists about whether unions aregood or bad for the economy. Like many institutions, their influence is probablybeneficial in some circumstances and adverse in others.

How does a union in the auto industry affect wages and employ-ment at General Motors and Ford? How does it affect wages and employment in otherindustries?


24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 630



In The NewsShould You Join a Union?Someday you may face the decision about whether to vote foror against a union in your workplace. The following article dis-cusses some issues you might consider.

On Payday, Union JobsStack Up Very WellBy David Cay Johnston

With the teamsters’ success in theirtwo-week strike against United ParcelService, and with the A.F.L.-C.I.O. train-ing thousands of union organizers in adrive to reverse a quarter-century of de-clining membership, millions of workerswill be asked over the next few yearswhether they want a union to representthem.

It is a complicated question, the an-swer to which rests on a jumble of deter-minations: Do you favor collective actionor individual initiative? Do you trust theunion’s leaders? Do you want somebodyelse speaking for you in dealings withyour employer? Do you think you will bedismissed if you sign a union card—orthat the company will send your joboverseas if a union is organized?

But in one regard, the choice is sim-ple—and it is not the choice that mostworkers have made during the labormovement’s recent decades in the eco-nomic wilderness.

From a pocketbook perspective,workers are absolutely better off joininga union. Economists across the politicalspectrum agree. Turning a nonunion jobinto a union job very likely will have abigger effect on lifetime finances thanall the advice employees will ever readabout investing their 401(k) plans, buy-

Source: The New York Times Money & Business section, August 31, 1997, p. 1. Copyright © 1997 by the New York Times Co. Reprinted bypermission.

ing a home or otherwise making moreof what they earn.

Here is how the equation works, saidProf. Richard B. Freeman of Harvard Uni-versity: “For an existing worker in afirm, if you can carry out an organizingdrive, it is all to your benefit. If there aregoing to be losers, they are people whomight have gotten a job in the future,the shareholders whose profits will godown, the managers because there willbe less profit to distribute to them inpay and, maybe, consumers will pay alittle more for the product. But as aworker, it is awfully hard to see why youwouldn’t want a union.”

Over all, union workers are paidabout 20 percent more than nonunionworkers, and their fringe benefits aretypically worth two to four times asmuch, economists with a wide array ofviews have found. The financial advan-tage is even greater for workers with lit-tle formal education and training and forwomen, blacks, and Hispanic workers.

Moreover, 85 percent of union mem-bers have health insurance, comparedwith 57 percent of nonunion workers,said Barry Bluestone, a labor-friendlyeconomics professor at the University ofMassachusetts.

The conclusion draws no argumenteven from Prof. Leo Troy of Rutgers Uni-versity, who is widely known in aca-demic circles and among union leadersfor his hostility to organized labor.

“From a standpoint of wages and fringebenefits,” Professor Troy said, “the an-swer is yes, you are better off in aunion.”

His objections to unions concernhow they reduce profits for owners anddistort investment decisions in ways thatslow the overall growth of the econ-omy—not how they affect workers whobargain collectively. Professor Troy pointsout that he belongs to a union him-self—the American Association of Uni-versity Professors.

Donald R. Deere, an economist at theBush School of Government and PublicService at Texas A & M University, stud-ied the wage differential for comparableunion and nonunion workers between1974 and 1996, a period when unionmembership fell to 15 percent of Ameri-can workers from 22 percent.

In every educational and age cate-gory that he studied, Professor Deerefound that union members increasedtheir wage advantage over nonunionworkers during those years. Last year, heestimates, unionized workers with lessthan a high school education earned 22percent more than their nonunion coun-terparts. The differential declined as edu-cation levels rose, reaching 10 percentfor college graduates.

“It makes sense to belong to aunion,” Professor Deere said, “so longas you don’t lose your job in the longterm.”

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 631



THE THEORY OF EFFICIENCY WAGESA fourth reason economies always experience some unemployment—in additionto job search, minimum-wage laws, and unions—is suggested by the theory ofefficiency wages. According to this theory, firms operate more efficiently ifwages are above the equilibrium level. Therefore, it may be profitable for firmsto keep wages high even in the presence of a surplus of labor.

In some ways, the unemployment that arises from efficiency wages is similarto the unemployment that arises from minimum-wage laws and unions. In allthree cases, unemployment is the result of wages above the level that balancesthe quantity of labor supplied and the quantity of labor demanded. Yet there isalso an important difference. Minimum-wage laws and unions prevent firmsfrom lowering wages in the presence of a surplus of workers. Efficiency-wagetheory states that such a constraint on firms is unnecessary in many casesbecause firms may be better off keeping wages above the equilibrium level.

Why should firms want to keep wages high? This decision may seem oddat first, for wages are a large part of firms’ costs. Normally, we expect profit-maximizing firms to want to keep costs—and therefore wages—as low as pos-sible. The novel insight of efficiency-wage theory is that paying high wagesmight be profitable because they might raise the efficiency of a firm’s workers.

There are several types of efficiency-wage theory. Each type suggests a differ-ent explanation for why firms may want to pay high wages. Let’s now considerfour of these types.

Worker Health

The first and simplest type of efficiency-wage theory emphasizes the linkbetween wages and worker health. Better paid workers eat a more nutritiousdiet, and workers who eat a better diet are healthier and more productive. A firmmay find it more profitable to pay high wages and have healthy, productiveworkers than to pay lower wages and have less healthy, less productive workers.

This type of efficiency-wage theory is not relevant for firms in rich countriessuch as the United States. In these countries, the equilibrium wages for mostworkers are well above the level needed for an adequate diet. Firms are notconcerned that paying equilibrium wages would place their workers’ health injeopardy.

This type of efficiency-wage theory is more relevant for firms in less devel-oped countries where inadequate nutrition is a more common problem. Unem-ployment is high in the cities of many poor African countries, for example. Inthese countries, firms may fear that cutting wages would, in fact, adverselyinfluence their workers’ health and productivity. In other words, concern overnutrition may explain why firms do not cut wages despite a surplus of labor.

Worker Turnover

A second type of efficiency-wage theory emphasizes the link between wages andworker turnover. Workers quit jobs for many reasons: to take jobs in other firms,to move to other parts of the country, to leave the labor force, and so on. The fre-quency with which they quit depends on the entire set of incentives they face,including the benefits of leaving and the benefits of staying. The more a firmpays its workers, the less often its workers will choose to leave. Thus, a firm canreduce turnover among its workers by paying them a high wage.

efficiency wagesabove-equilibrium wagespaid by firms to increaseworker productivity

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 632


Why do firms care about turnover? The reason is that it is costly for firms tohire and train new workers. Moreover, even after they are trained, newly hiredworkers are not as productive as experienced workers. Firms with higherturnover, therefore, will tend to have higher production costs. Firms may find itprofitable to pay wages above the equilibrium level to reduce worker turnover.

Worker Quality

A third type of efficiency-wage theory emphasizes the link between wages andworker quality. All firms want workers who are talented, and they try to pickthe best applicants to fill job openings. But because firms cannot perfectly gaugethe quality of applicants, hiring has a degree of randomness to it. When a firmpays a high wage, it attracts a better pool of workers to apply for its jobs andthereby increases the quality of its work force. If the firm responded to a surplusof labor by reducing the wage, the most competent applicants—who are morelikely to have better alternative opportunities than less competent applicants—may choose not to apply. If this influence of the wage on worker quality isstrong enough, it may be profitable for the firm to pay a wage above the levelthat balances supply and demand.

Worker Effort

A fourth and final type of efficiency-wage theory emphasizes the link betweenwages and worker effort. In many jobs, workers have some discretion over howhard to work. As a result, firms monitor the efforts of their workers, and work-ers caught shirking their responsibilities are fired. But not all shirkers are caughtimmediately because monitoring workers is costly and imperfect. A firm canrespond to this problem by paying wages above the equilibrium level. Highwages make workers more eager to keep their jobs and, thereby, give workers anincentive to put forward their best effort.

This particular type of efficiency-wage theory is similar to the old Marxistidea of the “reserve army of the unemployed.” Marx thought that employersbenefited from unemployment because the threat of unemployment helped todiscipline those workers who had jobs. In the worker-effort variant of efficiency-wage theory, unemployment fills a similar role. If the wage were at the level thatbalanced supply and demand, workers would have less reason to work hardbecause if they were fired, they could quickly find new jobs at the same wage.Therefore, firms raise wages above the equilibrium level, causing unemploy-ment and providing an incentive for workers not to shirk their responsibilities.


©D

ILB

ER

T R

EP

RIN

TE

D B

Y P

ER

MIS

SIO

NO

F U

NIT

ED

FE

ATU

RE

SY

ND

ICAT

E.

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 633


HENRY FORD AND THE VERY GENEROUS $5-A-DAY WAGE

Henry Ford was an industrial visionary. As founder of the Ford Motor Com-pany, he was responsible for introducing modern techniques of production.Rather than building cars with small teams of skilled craftsmen, Ford built carson assembly lines in which unskilled workers were taught to perform the samesimple tasks over and over again. The output of this assembly process was theModel T Ford, one of the most famous early automobiles.

In 1914, Ford introduced another innovation: the $5 workday. This might notseem like much today, but back then $5 was about twice the going wage. It wasalso far above the wage that balanced supply and demand. When the new $5-a-day wage was announced, long lines of job seekers formed outside the Ford fac-tories. The number of workers willing to work at this wage far exceeded thenumber of workers Ford needed.

Ford’s high-wage policy had many of the effects predicted by efficiency-wagetheory. Turnover fell, absenteeism fell, and productivity rose. Workers were somuch more efficient that Ford’s production costs were lower even though wageswere higher. Thus, paying a wage above the equilibrium level was profitable forthe firm. Henry Ford himself called the $5-a-day wage “one of the finest cost-cutting moves we ever made.”

Historical accounts of this episode are also consistent with efficiency-wagetheory. An historian of the early Ford Motor Company wrote, “Ford and hisassociates freely declared on many occasions that the high-wage policy turnedout to be good business. By this they meant that it had improved the disciplineof the workers, given them a more loyal interest in the institution, and raisedtheir personal efficiency.”

Why did it take Henry Ford to introduce this efficiency wage? Why wereother firms not already taking advantage of this seemingly profitable businessstrategy? According to some analysts, Ford’s decision was closely linked to hisuse of the assembly line. Workers organized in an assembly line are highly inter-dependent. If one worker is absent or works slowly, other workers are less ableto complete their own tasks. Thus, while assembly lines made production moreefficient, they also raised the importance of low worker turnover, high workereffort, and high worker quality. As a result, paying efficiency wages may havebeen a better strategy for the Ford Motor Company than for other businesses atthe time. •

Give four explanations for why firms might find it profitable to paywages above the level that balances quantity of labor supplied and quantity of labordemanded.


C A S ESTUDY

CONCLUSIONIn this chapter, we discussed the measurement of unemployment and the rea-sons economies always experience some degree of unemployment. We have seenhow job search, minimum-wage laws, unions, and efficiency wages can all help

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 634


explain why some workers do not have jobs. Which of these four explanationsfor the natural rate of unemployment are the most important for the U.S. econ-omy and other economies around the world? Unfortunately, there is no easyway to tell. Economists differ in which of these explanations of unemploymentthey consider most important.

The analysis of this chapter yields an important lesson: Although the econ-omy will always have some unemployment, its natural rate does change overtime. Many events and policies can alter the amount of unemployment the econ-omy typically experiences. As the information revolution changes the process ofjob search, as Congress adjusts the minimum wage, as workers form or quitunions, and as firms change their reliance on efficiency wages, the natural rate ofunemployment evolves. Unemployment is not a simple problem with a simplesolution. But how we choose to organize our society can profoundly influencehow prevalent a problem it is.


• The unemployment rate is the percentage ofthose who would like to work who do not havejobs. The Bureau of Labor Statistics calculates thisstatistic monthly based on a survey of thousandsof households.

• The unemployment rate is an imperfect measureof joblessness. Some people who call themselvesunemployed may actually not want to work, andsome people who would like to work have leftthe labor force after an unsuccessful search.

• In the U.S. economy, most people who becomeunemployed find work within a short period oftime. Nonetheless, most unemployment observedat any given time is attributable to the few peoplewho are unemployed for long periods of time.

• One reason for unemployment is the time it takesworkers to search for jobs that best suit theirtastes and skills. Unemployment insurance is agovernment policy that, while protecting work-

ers’ incomes, increases the amount of frictionalunemployment.

• A second reason our economy always has someunemployment is minimum-wage laws. By rais-ing the wage of unskilled and inexperiencedworkers above the equilibrium level, minimum-wage laws raise the quantity of labor suppliedand reduce the quantity demanded. The resultingsurplus of labor represents unemployment.

• A third reason for unemployment is the marketpower of unions. When unions push the wages inunionized industries above the equilibrium level,they create a surplus of labor.

• A fourth reason for unemployment is suggestedby the theory of efficiency wages. According tothis theory, firms find it profitable to pay wagesabove the equilibrium level. High wages canimprove worker health, lower worker turnover,raise worker quality, and increase worker effort.

SUMMARY

labor force, p. 615unemployment rate, p. 615labor-force participation rate,

p. 615natural rate of unemployment,

p. 616

unemployment insurance, p. 626union, p. 629collective bargaining, p. 629strike, p. 629efficiency wages, p. 632

KEY CONCEPTS

cyclical unemployment, p. 617discouraged workers, p. 619frictional unemployment, p. 622structural unemployment, p. 622job search, p. 623

KEY CONCEPTSKEY CONCEPTSKEY CONCEPTS

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 635



1. The Bureau of Labor Statistics announced that inJuly 2005, of all adult Americans, 142,076,000were employed, 7,497,000 were unemployed,and 76,580,000 were not in the labor force. Usethis information to calculate:a. the adult populationb. the labor forcec. the labor-force participation rated. the unemployment rate

2. Go to the website of the Bureau of Labor Statis-tics (http://www.bls.gov). What is the nationalunemployment rate right now? Find the unem-ployment rate for the demographic group thatbest fits a description of you (for example,based on age, sex, and race). Is it higher orlower than the national average? Why do youthink this is so?

3. As shown in Figure 3, the overall labor-forceparticipation rate of men declined between 1970and 1990. This overall decline reflects differentpatterns for different age groups, however, asshown in the following table.

Men Men MenAll Men 16–24 25–54 55 and older

1970 80% 69% 96% 56%1990 76 72 93 40

Which group experienced the largest decline?Given this information, what factor may haveplayed an important role in the decline in overallmale labor-force participation over this period?

4. The labor-force participation rate of womenincreased sharply between 1970 and 1990, asshown in Figure 3. As with men, however, therewere different patterns for different age groups,as shown in this table.

Women Women Women WomenAll Women 25–54 25–34 35–44 45–54

1970 43% 50% 45% 51% 54%1990 58 74 74 77 71

Why do you think younger women experienceda bigger increase in labor-force participationthan older women?

5. Between 2003 and 2004, total U.S. employmentincreased by 1.5 million workers, but the num-ber of unemployed workers declined by only 0.6million. How are these numbers consistent witheach other? Why might one expect a reductionin the number of people counted as unemployedto be smaller than the increase in the number ofpeople employed?

6. Are the following workers more likely to experi-ence short-term or long-term unemployment?Explain.a. A construction worker laid off because of bad

weatherb. A manufacturing worker who loses her job at

a plant in an isolated areac. A stagecoach-industry worker laid off because

of competition from railroads

PROBLEMS AND APPLICATIONS

1. What are the three categories into which theBureau of Labor Statistics divides everyone? Howdoes the BLS compute the labor force, the unem-ployment rate, and the labor-force participationrate?

2. Is unemployment typically short-term or long-term? Explain.

3. Why is frictional unemployment inevitable? Howmight the government reduce the amount of fric-tional unemployment?

4. Are minimum-wage laws a better explanation forstructural unemployment among teenagers oramong college graduates? Why?

5. How do unions affect the natural rate of unem-ployment?

6. What claims do advocates of unions make toargue that unions are good for the economy?

7. Explain four ways in which a firm might increaseits profits by raising the wages it pays.

QUESTIONS FOR REVIEW

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 636



d. A short-order cook who loses his job when anew restaurant opens across the street

e. An expert welder with little formal educationwho loses her job when the company installsautomatic welding machinery

7. Using a diagram of the labor market, show theeffect of an increase in the minimum wage onthe wage paid to workers, the number of work-ers supplied, the number of workers demanded,and the amount of unemployment.

8. Consider an economy with two labor markets—one for manufacturing workers and one for ser-vice workers. Suppose initially that neither isunionized.a. If manufacturing workers formed a union,

what impact on the wages and employmentin manufacturing would you predict?

b. How would these changes in the manufactur-ing labor market affect the supply of labor inthe market for service workers? What wouldhappen to the equilibrium wage and employ-ment in this labor market?

9. It can be shown that an industry’s demand forlabor will become more elastic when thedemand for the industry’s product becomesmore elastic. Let’s consider the implications ofthis fact for the U.S. automobile industry andthe United Autoworkers Union (UAW).a. What happened to the elasticity of demand

for American cars when the Japanese devel-oped a strong auto industry? What happenedto the elasticity of demand for American autoworkers? Explain.

b. As the chapter explains, a union generallyfaces a trade-off in deciding how much toraise wages because a bigger increase is bet-

ter for workers who remain employed butalso results in a greater reduction in employ-ment. How did the rise in auto imports fromJapan affect the wage-employment trade-offfaced by the UAW?

c. Do you think the growth of the Japanese autoindustry increased or decreased the gapbetween the competitive wage and the wagechosen by the UAW? Explain.

10. Suppose that Congress passes a law requiringemployers to provide employees some benefit(such as healthcare) that raises the cost of anemployee by $4 per hour.a. What effect does this employer mandate have

on the demand for labor? (In answering thisand the following questions, be quantitativewhen you can.)

b. If employees place a value on this benefitexactly equal to its cost, what effect does thisemployer mandate have on the supply oflabor?

c. If the wage is free to balance supply anddemand, how does this law affect the wageand the level of employment? Are employersbetter or worse off? Are employees better orworse off?

d. Suppose that, before the mandate, the wagein this market was $3 above the minimumwage. In this case, how does the employermandate affect the wage, the level of employ-ment, and the level of unemployment?

e. Now suppose that workers do not value themandated benefit at all. How does this alter-native assumption change your answers toparts (b) and (c)?

For further information on topics in this chapter, additional problems, examples, applications, online quizzes, and more, please visit our website at

24729_28_c28_p613-638.qxd 12/7/05 2:37 PM Page 637

academic.cengage.com/economics/mankiw.


24729 00 front end sheets - University of...

Documents

Transcript of 24729 00 front end sheets - University of...