Extension 11.6 Why there is always unemployment

None of these countries, and no capitalist economies, have ever had zero unemployment. Why is there always unemployment in capitalist economies? To answer this question, we first need to distinguish between two types of unemployment: search unemployment and involuntary unemployment.

Two types of unemployment

search unemployment

Since workers differ from each other, and so do jobs, unemployed workers and firms with vacancies spend time searching for an employment match that suits them both. Unemployment caused by the search and matching process is called search unemployment.

involuntary unemployment

A person is involuntarily unemployed if they are seeking work, and willing to accept a job at the going wage for people of their level of skill and experience, but unable to secure employment.

As we learned in Chapter 10, the labor market is a matching market. If a worker quits a job, it may take some time for that worker to find a new job. When a firm decides to hire a new worker, that process also takes time. The unemployment that results from the time it takes for firms and workers to make a match is search unemployment.

So long as the labor market remains a matching market, there will always be a certain amount of search unemployment. But even if we could eliminate search unemployment, capitalist economies would still have unemployed workers. Even if the matchmaking process could be completed instantaneously, there will always be some workers who want a job, are looking for a job, and are ready to work, but cannot find anything. This type of unemployment is involuntary unemployment.

Whereas search unemployment results from the matching process, involuntary unemployment presents more of a puzzle. It seems inefficient to prevent people who want to work from working. Why does involuntary unemployment happen?

Why employers need unemployment to make profits

So long as the unemployment rate is sufficiently high, workers will expect there to be a significant cost to losing their job. As we saw in this chapter, this expectation produces an employment rent, which incentivizes workers to put in the effort employers demand. Some unemployment is needed if employers want the power to command workers to exert work effort without sharing profits or ownership.

We can see why it is impossible to sustain zero involuntary unemployment in a capitalist economy by thinking through what would happen if there were no involuntary unemployment. Figure E11.4 walks us through each step.

To begin, assume there is zero involuntary unemployment. In this situation, everyone who wants a job right now has one. The only unemployment is search unemployment.

If everyone who wants a job immediately gets one, then losing your job will have basically zero cost. But if the net benefit of your outside option is equal to the net benefit of your current job, there will be zero employment rent. If the workers in a firm have zero employment rent, they have no reason to work. If workers don’t work, then firms cannot produce anything. And if firms cannot produce anything, they will shut down, leading to involuntary unemployment.

Thus, when you start in a situation of zero involuntary unemployment, workers necessarily become involuntarily unemployed. Zero involuntary unemployment is thus not a stable outcome in a capitalist economy.

In other words, involuntary unemployment is necessary in a capitalist economy because without it there is no employment rent, and without employment rent there is no reason for workers in a firm to put in any effort. You can’t have an economy if workers aren’t doing any work.

Data Extension 11.6 Who is unemployed?

In this Data Extension we look at the data on unemployment rates disaggregated by education and then by race. We will use what we’ve learned in this chapter to help us better understand the data.

Education and unemployment

Figure E11.5 shows the US unemployment rate from January 2001 to December 2024 for those over the age of 25 divided into four categories:

1. people who never graduated high school
2. people who graduated high school and did not take any college classes
3. people who attended some college or got an associate’s degree but did not complete a bachelor’s degree
4. people who earned at least a bachelor’s degree

As the figure shows, there is always some unemployment for each of the four categories, but the unemployment levels differ. Comparing the highest and lowest educational levels in the graph, the average unemployment rate over this period was more than three times higher for people who did not graduate high school than for those with at least a bachelor’s degree. Similarly, the average unemployment rate for people with at most a high school degree was about twice the unemployment rate of people with at least a bachelor’s degree.

Figure E11.5 also shows that the size of the unemployment gaps between these groups varies and tends to get smaller when overall unemployment is lower. Between 2009 and 2019, for example, the unemployment rate for people who never graduated college dropped substantially, such that all four lines started to converge around 2019. Lower overall unemployment tends to be especially beneficial for those who typically have the hardest time finding a job.

Why do such gaps exist in the first place? Why do unemployment rates go down as education levels go up? Some key factors include the following:

• On-the-job training: More-educated workers are more likely to take jobs involving more varied and complex tasks. Such jobs tend to require more on-the-job training, because the worker needs to learn about the job to perform it well. More-educated workers thus tend to have more firm-specific assets thanks to the increased on-the-job training they are likely to get. More firm-specific assets lead to higher net benefits of a job, leading to higher average employment rents for more-educated workers, which in turn incentivizes these workers to work harder to keep their job. Because the extra training is costly to the firm, it too has a strong incentive not to fire workers unless it’s absolutely necessary.
• Job search and job offers: Because they often have desk jobs using a computer, more-educated workers are more likely to search for other jobs while they are still employed at their current job, as opposed to waiting until they lose their job to search for another one. More-educated workers are also more likely to receive job offers from other firms while they are employed. Both factors mean that more-educated workers are less likely to have a spell of unemployment between jobs.
• Specialization: Workers who are more highly educated tend to be more specialized, meaning there is a relatively small pool of workers with a given specialization for firms to choose from. Firms will then have to compete more for these workers, who are harder to replace and less likely to be fired.
• Preference for more-educated workers: All else being equal, many employers prefer to hire workers with more education. This is true even for many jobs where a college or high school degree is not necessary, suggesting this preference may have nothing to do with skills. When unemployment is low, some jobs that are usually advertised as requiring a college degree remove that requirement.

Those with higher education also tend to get paid more. In 2024, for example, the median earnings were $2,278 per week for a worker with a PhD, $1,543 per week for a worker with at most a bachelor’s degree, and $930 for a worker with at most a high school diploma. The higher unemployment rates for less-educated workers can help explain these differences. As we saw earlier in this chapter, a higher unemployment rate means a weaker outside option. If the outside option of less-educated workers tends to be weaker, then those workers are going to have lower reservation wages and thus lower no-shirking wages.

Specialization and firm preferences for more-educated workers also tend to push up the wages of more-educated workers, thereby giving them stronger outside options.

Unemployment disaggregated by race

Figure E11.6 shows the unemployment rate for White people, Black people, and Hispanic people from January 2001 to December 2024. There are consistent differences in unemployment rates across groups. The unemployment rate for White people is always the lowest, and it is always the highest for Black people. A general rule over this time period is that the unemployment rate for Black people is about twice that of White people. The figure also shows that the unemployment gap between these three racial groups fluctuates with the state of the economy. In 2019 and 2023, for example, unemployment generally was low, and the unemployment rate for the three groups became much closer. By comparison, throughout much of the 1980s, Black unemployment was about 2.5 times that of White unemployment.

One reason for these gaps is education. In 2022, 41.8% of White people over the age of 25 had at least a bachelor’s degree, whereas that number was 27.6% for the Black population, and 20.9% for the Hispanic population. Because more education typically leads to lower unemployment rates, racial groups with more education are likely to have lower unemployment, which helps explain why White unemployment is lower than that for Hispanic people and Black people.

However, differing educational outcomes cannot fully explain the gap in unemployment rates. For one thing, if education alone explained the difference, we would expect Hispanics, who have lower average years of education, to have the highest unemployment rate, which they do not. In addition, if education primarily caused the Black/White unemployment gap, we would expect to see similar unemployment rates for Black workers and White workers who have the same levels of education. Figure E11.7 shows the unemployment rate in 2017 for Black workers and White workers with different levels of education. On the far left, for example, the figure shows the unemployment rate for Black workers and White workers who did not graduate high school. Although both groups have higher unemployment rates than those with more education, the unemployment rate for Black workers with no high school diploma is nearly twice that of similar White workers. A similar pattern is true for every level of education. Education alone cannot therefore explain the overall differences in unemployment between Black workers and White workers.

In Chapter 10 we saw that Black workers and Hispanic workers are more likely to experience labor market discrimination. Figure E11.7 is further evidence of this discrimination: a Black worker with the same education as an equivalent White worker has a harder time finding or keeping employment.

Discrimination affects both unemployment and wages. If a worker expects to face discrimination in the hiring process, then the expected net benefit of their outside option will be lower, because it will take longer to find a job. Their reservation wage and no-shirking wage therefore decrease. Employers may know Black or Hispanic workers have a lower reservation wage, and therefore offer lower wages than they would to similar White workers. The employers’ actions reinforce Black workers’ and Hispanic workers’ expectations of lower pay.

Another cause of the racial gaps in unemployment is segregation. Segregation reduces Black and Hispanic access to employment in two ways: spatial isolation and social isolation. Spatial segregation can impact a worker’s ability to get a job in several ways:

• If a person lives far from areas with high job growth and opportunities, they are less likely to know about job openings in those areas.
• Even if a worker knows about a job opening in the suburbs or a different part of the city, they may not be able to get to that location easily, either because they don’t have a car or because public transit doesn’t exist, doesn’t go to the job’s location, or takes too long.
• Employers located far from Black or Hispanic neighborhoods may be more likely to discriminate against Black or Hispanic workers.
• Black or Hispanic workers may fear venturing into majority White areas.

Compounding the effects of spatial mismatch is the social isolation that results from segregation. As we saw in Chapter 10, when looking for a job, workers use their personal networks—friends, family, friends of friends—to learn about job openings and get hired. Given that racial segregation leads to largely unconnected Black, White, and Hispanic social networks, Black workers and Hispanic workers have a limited capacity to benefit from such practices because Black and Hispanic unemployment rates are higher and White people are more likely to own businesses or be in a strong position to affect hiring decisions.