Markets For Labor - The Firm's Decision To Hire Labor

CHAPTER 9: MARKETS FOR LABOR Jobs are not what they used to be. Various studies conclude that, at least in the United States, the pattern of employment is changing, with fewer jobs in manufacturing and more in the service sector. In many areas, “good” jobs—jobs that offer decent wages, job security, health care, and retirement benefits—are on the decline.1 According to a 2016 study, “tectonic changes are reshaping U.S. workplaces” that “are affecting the very nature of jobs.”2 One prominent example is the growth of the “gig” economy, where employment isn’t defined by a steady, full-time job, but by shorter-term freelance or contract projects. 3 A 2021 analysis estimates that as many as one-third of U.S. workers are employed in the gig economy, with fewer people seeking traditional jobs during the economic recovery from COVID-19 lockdowns.4 A similar shift is occurring in most other countries. For example, China’s gig economy is also rapidly expanding, driven largely by food delivery and ride sharing work. 5 A job isn’t just a means to a paycheck for most people. Work is often central to our sense of identity.6 When asked “What do you do?” the most common answer is a job title. Research shows that our jobs can have a significant influence on our wellbeing. Those who are more satisfied with their jobs tend to have higher life satisfaction, are more likely to experience positive emotions, and less likely to experience negative emotions.7 Meanwhile, being unemployed has significant negative impacts on one’s mental and physical health.8 In this chapter, we take a detailed look at labor markets. We will explore how labor markets function, including some factors that make labor markets unique. We will consider the reasons why some jobs provide higher salaries and better benefits than other jobs. Finally, we will survey several recent trends in labor markets in the United States and around the world, along with government policies that explain or respond to these trends. 1. ECONOMIC THEORY OF LABOR MARKETS “The labor market” is a familiar phrase, but markets for labor are different from other markets in many ways. For a start, consider what is sold in a labor market. It is not human beings; slavery, one of the most despicable practices in human history, is illegal everywhere (although a 2019 analysis found that there are still over 40 million slaves throughout the world, including some in developed countries9). Rather, what is sold in labor markets is what is sometimes called “labor power”—that is, what a given person is able and willing to do in a given amount of time. An employer who hires a certain amount of labor power (X number of people working for Y hours) expects that it will produce a certain level of output. But it is not the actual output that is being purchased in this market—it is the contribution that employees make toward the production of output. This makes labor markets different from markets for the things that labor produces, such as sweaters, jet planes, or customer service over a telephone. The neoclassical model of labor starts with a familiar idea—that the market is based on the interaction of supply and demand. We will shortly focus on differences, but in some ways labor markets are similar to markets for other things. The demand side of labor markets comprises firms seeking to maximize their profits. The supply side comprises people seeking to maximize their utility, in this case by exchanging their labor power for payment. The stylized utility-maximizing consumers who were DRAFT 1 described in Chapter 8 are now simply dressed in overalls or suits and sent into the workplace, with the single goal of earning the money that will allow them to be consumers. Like the neoclassical consumer model in Chapter 8, the neoclassical labor model makes a number of simplifying assumptions. On the demand side, the model assumes that all firms are forced by competition to maximize profits and minimize costs, that labor productivity can be easily measured, and that historical and social contexts can be ignored. On the supply side, people are assumed to seek employment purely in order to earn money, disregarding any intrinsic motivations or benefits associated with work. Just like any other market model, the labor market model is used to study prices and quantities. Thus, we are concerned with how much labor will be supplied and purchased, and at what price (wage). To determine this, we need a theory both of the demand for and the supply of labor. THE FIRM’S DECISION TO HIRE LABOR 1.1 On the demand side of the labor market, consider a firm seeking to hire a specific type of labor. What should guide the firm’s decisions about how much labor to hire? From the viewpoint of a profit-maximizing firm, an additional person-hour of labor will be desirable if it increases profits, but not otherwise. Hiring an additional person-hour does two contradictory things to the firm’s profit position: • • Costs are raised by the amount of the additional wages paid. Revenue is increased by the value of the increase in output produced by the additional hour of work. Clearly, as long as the firm gets more additional revenue than it has to pay out in additional wages, it should keep hiring workers. But if it is getting less in additional revenue than it is paying out in additional wages, it should reduce the number of workers that it hires. The profit-maximizing decision rule for the firm can thus be expressed as hiring labor up to the point where: MRPL = MFCL where MRPL is the marginal revenue product of labor, or the amount that an additional unit of labor contributes to revenues, and MFCL is the marginal factor cost of labor, or the amount that the additional unit of labor adds to the firm’s costs. Note that hiring an additional unit of labor clearly adds to the wages the firm must pay, but it may also increase other costs, such as energy and supplies. For simplicity, we assume that the MFCL is just the additional wages, and that for the individual firm hiring another unit of labor doesn’t change the wages paid to any other workers. marginal revenue product of labor (MRPL): the amount that a unit of additional labor contributes to the revenues of the firm marginal factor cost of labor (MFCL): the amount that a unit of additional labor adds to the firm’s costs In other words, the firm should hire additional units of labor until the marginal benefits just equal the marginal costs. We will see very similar reasoning in Chapter 16, concerning a firm’s decision about how much to produce—for exactly the same reasons. A formal derivation of this rule is described in the appendix to this chapter. DRAFT 2 If the firm buys labor services in a competitive market, MFCL will simply be the competitively determined market wage, and the rule will simplify to: MRPL = Wage In most situations, the marginal revenue product of labor will decline as more workers are hired. A factory, for example, may be able to increase output by hiring more workers, but beyond a certain point additional workers will bring less benefit in terms of increased production. Further, the higher the market wage, the sooner a firm will reach the quantity of workers where MRPL equals the wage rate. Also, as the wage rate gets higher a firm is more likely to switch away from hiring labor towards other productive technologies, such as robot manufacturing or self-serve checkout counters. Overall, we expect these factors will create a downward-sloping demand curve for labor, as shown in Figure 9.1. Figure 9.1. An Individual Firm’s Labor Demand Curve 1.2 THE INDIVIDUAL’S DECISION TO SUPPLY LABOR The traditional model of consumer behavior presented in Chapter 8, in which consumers seek to maximize their utility, can be reframed to apply to decisions about labor supply. Specifically, we can consider how much time an individual is willing to work, given different wage levels. Again, we assume that people seek to maximize their utility. In this simple model, the potential worker is (rather unrealistically) assumed to have perfect information regarding available jobs and wages, and to be free to vary his or her hours of paid work. However, the labor market model differs from the model of consumer choice in that here the “budget line” is defined according to the number of hours that the individual has available to “spend” on activities, rather than according to the amount of money that he, she, or they has to spend on goods. Whereas different people have different budget lines, everyone has the same amount of available time (i.e., 24 hours per day, 7 days per week, etc.). In the neoclassical labor model, there are three different activities that an individual can spend his or her time on: • • • paid work unpaid work leisure DRAFT 3 Hours “spent” on paid labor result in wages, which in turn give opportunities for consumption. Hours spent on other activities yield utility either directly (as in the case of leisure) or indirectly through unpaid production such as cooking, cleaning, or even volunteer work. In this model, a paid job is itself generally assumed to yield no direct utility. Thus a potential worker must consider a trade-off between the benefits of earning an income against the lost utility from time spent working. The Opportunity Cost of Paid Employment As you might expect, the concept of opportunity cost is relevant for our labor market model. The opportunity costs of time spent working may include the following: • • • • Household production: A paid job may reduce the time that can be spent in productive but unpaid work at home—raising children, caring for elderly or sick relatives, cooking, cleaning, gardening, and the like. Education: As an alternative to seeking paid work immediately, individuals may decide to stay in school or return to school—either to prepare for better-paid future employment or simply to enjoy the life of a student. Self-employment: People can work for themselves rather than for someone else, starting a business such as opening a store or running a day care business in their home in order to make a profit. In this chapter, however, we focus only on people who work for wages or salaries. Leisure: Work cuts into the time available for leisure, whether playing music, fishing, camping, reading novels, watching sports, or whatever activities provide utility directly. To the extent that you value any of these four pursuits and reduce the hours you devote to them when you take a paid job, that job has a “cost.” The cost is the lost opportunity for other activities. In addition to the opportunity costs associated with your time, you may incur direct monetary costs when taking a paid job, such as the costs of work-related clothing and commuting. You may incur increased monetary expenditures for things that otherwise might have been home-produced (using your time resources), such as childcare and meal preparation. The Benefits of Paid Employment At the same time, of course, paid jobs have many benefits. Most obvious is the fact that they are paid. Even if paid work is unpleasant, boring, stressful, or even demeaning, wages and salaries are strong extrinsic motivators that encourage individuals to supply their labor. In addition, paid work itself has great intrinsic significance in most people’s lives. Evidence from actual or possible lottery winners illustrates this point. According to a 2016 survey, 80 percent of American office workers would continue working even if they won the lottery.10 And among those who actually won the lottery, another study found that 85 percent continued to work despite winning an average of around $4 million, and most of them stayed at the same organization.11 Reasons for working other than money include friendships on the job, a feeling of contribution to a broader cause, and the sense of identity that comes from one’s work. As we noted above, however, the neoclassical model of labor markets considers pay to be the only relevant benefit of work. We will therefore maintain this assumption for our initial model of labor supply. DRAFT 4 1.3 THE INDIVIDUAL SUPPLY CURVE FOR LABOR We look at the decision of an individual to supply various amounts of work hours over a given time period, say a week, assuming, for the moment, that the worker can find a paid job meeting some minimum criteria of working conditions, commuting time, etc. For now, we do not consider a worker’s choices among different kinds of paid jobs, focusing only on the decision about how much time to put into an existing job. As we did in Chapter 3, we can conceptualize a supply curve that shows the relationship between price and quantity. In this case, price is the wage rate and quantity represents the number of hours of paid labor supplied. In other words, we are looking only at the effect of different wage levels on the individual’s willingness to supply labor to the market. A reasonable first thought is that a labor supply curve will slope upward, just like all supply curves we have been looking at so far in the book. We show this in Figure 9.2, where the wage rate is on the vertical axis and the quantity of labor is on the horizontal axis. You might think of the wage rate as an hourly wage, as is common in many blue-collar and service jobs, or as a weekly or monthly salary that is common with professional and managerial jobs. Jobs may also pay in the form of tips, bonuses, or stock options, and they may provide fringe benefits such as health insurance. For our simple supply-and-demand analysis, we include all these in the concept of a “wage.” The quantity of labor is measured in time units, such as the number of hours one is willing to work per week. This upward-sloping labor supply curve is supported by some simple economic logic: The higher the wage rate, the more attractive work is relative to other activities. We can think of a substitution effect, where workers substitute away from leisure and other activities, and toward more work, as the wage rate increases. We can also note that the opportunity cost of not working increases as the wage rate goes up. So, when offered a very low wage, an individual may be reluctant to join the labor market, or to supply many hours of work, because they may get more benefits from leisure, selfemployment, or other activities. The higher the market wage rate, the more attractive it is to engage in additional paid labor instead of other activities.12 Figure 9.2. Upward-Sloping Labor Supply Curve But we next need to think a little more deeply about the labor supply decision. Of course, one of the important reasons that an individual works is to earn an income, DRAFT 5 which in turn is used to buy goods and services that they can then enjoy. Yet as one’s wage continues to rise, will the person always want to work more and more? Probably not. Economists explain this in terms of the fact that leisure pursuits (and perhaps other unpaid activities) are usually “normal goods,” in the sense explained in Chapter 4. As people earn higher incomes, they may also want more time to enjoy the fruits of their labor. So, while the substitution effect described above suggests that a higher wage will result in more labor supplied, we also need to consider an income effect. Specifically, the higher the market wage, the more leisure (and other unpaid activities) people will want to “buy.” People may also have a target level of income in mind, beyond which they have less need for additional money. As we saw in Chapter 8, workers early in the industrial era often had such income targets. Increases in wages above the traditional level led individuals to take longer weekends and offer fewer hours of work the next week. Imagine you are already making a good wage rate, and you get a raise. You may well decide to work less, rather than more, assuming you have the choice. In this case, a higher wage rate leads to a lower quantity of labor supplied, and a backwardssloping labor supply curve! Figure 9.3. Backward-Bending Individual Labor Supply Curve Economists usually believe that, at relatively low wage rates, the substitution effect will dominate. But eventually, when wages rise enough, the income effect will dominate. The overall result is a backward-bending individual paid labor supply curve, as shown in Figure 9.3. So, the existence of the income effect means that the individual labor supply curve looks rather different from our normal supply curve in previous chapters. backward-bending individual paid labor supply curve: a labor supply curve that arises because, beyond some level of wages, income effects outweigh substitution effects in determining individuals’ decisions about how much to work DRAFT 6 1.4 THE MARKET SUPPLY CURVE FOR LABOR We now broaden our thinking, moving from labor demand by an individual firm and labor supply by an individual worker, to the market level. We first extend our analysis of labor supply, and then consider market-level labor demand. The supply of labor to a particular market, such as the national market for aerospace engineers or the market for restaurant wait staff in Chicago, can be thought of as the horizontal sum of the supply curves of those individuals who could participate in the market. Although the supply curves of some individuals might bend backward, the supply curve for a particular market can generally be assumed to have the usual upward slope, similar to Figure 9.1. This is because employers can obtain a larger quantity of labor in two ways. The first is by persuading workers already in the market to supply more hours. The second way is to attract more workers to enter the particular market, either by drawing them away from other jobs or by drawing them into the paid labor force from other activities. For most of these workers, we can assume that the substitution effect dominates, and so the supply curve will slope upward. Market labor supply is relatively wage elastic if a variation in the wage brings a large change in the quantity of labor supplied. This could occur if the (upward-sloping sections of) individual worker’s supply curves are elastic. It also occurs when a rise in the wage readily draws more workers into the particular market. Markets for types of labor that use general or more easily acquired skills generally tend to have relatively elastic supply curves. If the local wage for restaurant wait staff rises, for example, people may leave jobs as salesclerks and delivery truck drivers in order to work at restaurants. If the wages paid by restaurants fall, wait staff may look for jobs as salesclerks and drivers. Market labor supply is relatively wage inelastic, however, if a variation in the wage brings little change in the quantity of labor supplied. At the extreme, the supply of labor might be “fixed” for some occupations, at least in the short run. For example, there are only so many aerospace engineers in the United States at any point in time. (What slope would the supply curve have?) Raising the wage might draw a few engineers out of retirement or self-employment, but it cannot instantly produce a large quantity of new engineers, because obtaining the skills necessary for this job requires many years of education. A drop in the wage, similarly, might not much decrease the quantity of labor supplied in the short run, because the engineers’ specialized skills are not valued nearly as much in other markets. Changes in the quantity supplied will occur only over the long run, as high wages attract more students to train for the job or low wages cause more engineers to become dissatisfied and retrain for something else. The United States has also used immigration policies to increase the quantity of labor supplied in certain high-skilled areas where there are labor shortages. For a realworld example of a labor shortage, see Box 9.1. We can also think of the supply of labor not just in terms of the market for a specific type of job, such as aerospace engineers, but with regard to the labor market for an entire region or country. In this case, we are considering the total number of people willing to supply their labor as a function of the average wage rate. Economists define the percentage of adults willing to work at current wages as the labor force participation rate. In the United States, the labor force participation rate in 2021 was about 62 percent. That is lower than it was in the 2000s and 2010s, but higher than it was in the 1950s and 1960s.13 We will discuss trends in labor force participation later in the chapter. DRAFT 7 labor force participation rate: the percentage of the adult, noninstitutionalized population that is either working at a paid job or seeking paid work BOX 9.1 A SHORTAGE OF DOCTORS A shortage of doctors in the United States is only expected to get worse, due to changes on both the demand and supply side. According to a 2021 study, the shortfall of doctors is predicted to be between 38,000 and 124,000 by 2034.14 The most important change on the demand side is the aging of the American population, given that health care needs generally increase with age. By 2034 the number of Americans over age 65 is predicted to increase by over 40 percent, with even larger growth (74 percent) for those age 75 and over. On the supply side, one problem is that the supply of doctors is relatively inelastic in the short term. It commonly takes about 10 to 15 years from the time an undergraduate decides to become a doctor to the time when they can actually start practicing. Thus, the labor market for doctors is an example of a market that can persist in a condition of disequilibrium for many years.15 The COVID-19 pandemic has had multiple impacts on the market for doctors. While illnesses from COVID-19 have increased the demand for medical care, the number of patients undergoing routine procedures such as colonoscopies has declined.16 On the supply side, the pandemic has caused many doctors to leave the profession, primarily through early retirement.17 About one-third of doctors are over age 60, and the pandemic increased stress, the risk of infection, and costs such as protective equipment.18 The COVID-19 pandemic also magnified already-existing inequalities in the U.S. health care system, and addressing those inequalities would create an even greater shortage of doctors: The COVID-19 pandemic has only exacerbated our nation’s doctor shortages. Furthermore, it shined a spotlight on already profound inequities in health care access. [We estimate] that if all insurance-, place-, and racebased inequities in health care access were eliminated, then the country would need an additional 180,400 physicians.19 The elasticity of labor supply at this broadest level considers how the labor force participation rate changes as average real wages change. Based on the results of dozens of studies from the United States and Europe, labor force participation tends to be relatively wage-inelastic.20 Most estimated elasticities are between 0.05 and 0.60, indicating that higher real wages do increase labor force participation rates, but only modestly. Labor participation tends to be somewhat more wage elastic for females. In addition to considering the elasticity of labor supply, it is important to note that labor supply curves can also shift in response to nonprice factors, just like the shifts in other supply curves that we studied in Chapter 3. For the economy as a whole, for example, labor supply curves tend to shift outward over time because of population growth. Changes in gender norms and household technology have resulted in outward shifts in labor supply curves for professions such as law and medicine as increasing numbers of women have entered those fields. The COVID-19 pandemic, in contrast, reduced the supply of people willing to work in jobs such as health care and teaching. DRAFT 8 Changes in one labor market may also have repercussions in other markets. For example, a rise in the wages of salesclerks (a movement along the supply curve for salesclerks) might decrease the supply of wait staff (that is, shift the supply curve for wait staff back), as people exit the wait staff market in order to take advantage of the higher wages now being offered for salesclerks. 1.5 MARKET DEMAND CURVES The demand curve for paid labor—whether for a specific job or for the entire labor market—can generally be thought of as downward-sloping, like the demand curves we examined in previous chapters. When wages are high, employers have incentives to economize on the use of labor. They may cut back on production or try to substitute other inputs (e.g., another type of labor, machinery, or computerization) for the type of labor whose wage is high. But when wages are low, employers may be able to expand production or substitute relatively cheap labor for other inputs. In terms of the marginal revenue product logic discussed above, we can say that firms are willing to accept a lower marginal revenue product of labor when wages are low but will insist on a high marginal revenue product when wages are high. Labor demand will tend to be relatively wage elastic if there are good substitute inputs available and if the wage bill is a large proportion of total production costs (so that the employers are motivated to seek out substitutes). Labor demand will tend to be relatively inelastic if no good substitute inputs are available and the wage bill is a small proportion of total costs. A 2015 meta-analysis based on over 100 studies from all regions of the world finds that labor demand tends to be inelastic, averaging around 0.20 in the short run and 0.40 in the long run.21 We would expect labor demand to be more elastic in the long run as employers have more time to find substitutes for relatively expensive labor. The article also found that demand is more elastic for lowskilled workers, as it is easier to find substitutes for workers without specialized skills. The labor demand curve may shift if there is a change in the demand for the good or service that it is used to produce, if technological developments alter the production process, if the number of employers changes, or if the price or availability of other inputs changes. For example, when an organization experiences a fall in demand for its products, its labor demand curve will shift back as well. 1.6 MARKET ADJUSTMENT Still using the same simplifying assumptions as in Chapter 3 about how markets work, we can examine how market forces might influence wage rates and the quantity of labor employed. For example, Figure 9.4 depicts a stylized labor market for real estate agents in the United States. In the early 2000s, home sales were booming, and demand for the services of real estate agents was high, as depicted by demand curve D1. Stories in the newspapers at the time touted the fat salaries real estate agents could make as many people were buying homes in order to “flip” them in a short time period at a profit. When the housing bubble burst during 2006 and 2007, home sales declined by about half, and the demand for real estate agents similarly declined. The labor market for real estate agents went from boom to bust. We can think of this as the demand curve shifting to D2. Comparing equilibrium E1 to equilibrium E2, we can see that the model predicts that the number of real estate agents will fall and that the wage will fall as well. In fact, DRAFT 9 employment for real estate agents declined by more than 25 percent during the Great Recession and real wages declined as well.22 Figure 9.4. The Labor Market for Real Estate Agents Labor market adjustment takes time—the movement from E1 to E2 is not instantaneous. It takes time for workers to change their career plans and for employers to adjust wages and salaries, which may be set by labor contracts. Given that labor market conditions are constantly changing, it may be unclear whether a particular labor market is in equilibrium. Much of the recent labor economics research has focused on the persistence of “friction” in labor markets, which slows the transition of workers from one job to another. In particular, unemployed workers may spend considerable time searching for a job that meets their specific requirements. The existence of labor market friction means that a significant number of job openings typically exist, even during periods of unemployment. For example, in February 2022 the unemployment rate was 3.8 percent in the United States and 6.3 million people were unemployed. Yet there were also 11.3 million job openings. Discussion Questions 1 Suppose that your college or university substantially raises the wages that it offers to pay students who tend computer laboratories, monitoring the equipment and answering questions. What do you think would happen to the quantity of labor supplied? Why? Where would the extra labor hours come from? Do you think the supply of this kind of labor is elastic or inelastic? Why? 2 Opticians fit people who have poor eyesight with glasses or contact lenses, prescribed by an optometrist. Beginning in the 1990s, technological developments in laser eye surgery made surgery an increasingly popular way of correcting bad eyesight. What effect do you think this development had on the market for opticians? Draw a graph, carefully showing whether the shift is in demand or supply and showing the resulting predicted changes in the quantity of labor demanded and in the wage. DRAFT 10 2. EXPLAINING VARIATIONS IN WAGES Among the things that economists are especially eager to understand about labor markets is differences in wages. Why do professional basketball players (average annual salary over $8 million in 202223) make so much more than aerospace engineers (average salary of $121,000), who in turn earn so much more than preschool teachers (average salary of $37,000)?24 In addition, within the same job definition it is possible to find workers who receive very different compensation, even though they seem to have equivalent qualifications and are hired from the same job market. Women in the United States are paid, on average, only 82 percent of what men make on average.25 Are such patterns of wage differentials determined by the logic of markets? If not, what other forces affect them? 2.1 WAGE VARIATIONS IN THE NEOCLASSICAL LABOR MODEL Some economists stress productivity differences as nearly the sole source of wage variation over the long run. This emphasis requires a number of restrictive assumptions: that people behave in a rational, self-interested way, that market forces are strong, and that markets are perfectly competitive. A high wage, in this view, is merely a sign that an individual is making a highly valued contribution. In the neoclassical labor model, the demand for labor—the employers’ willingness to pay for different types of labor services—is solely a function of how productive workers are. Employers are not willing to pay their workers more than the value that each one contributes to what the employer finally sells—what we have called the marginal revenue product of labor. Employers who can get away with paying workers less than their MRPL are motivated to do so in order to minimize costs and therefore to maximize profits. Theoretically, this should not be possible, if the labor market is truly competitive. In that case, workers who do not receive a “fair” wage from one employer (a wage equal to their MRPL) can find another employer who will offer the wage that actually represents the worker’s contribution. In the neoclassical model, the main reason for variations in labor productivity, and hence in wages, is human capital, which consists of people’s knowledge and skills. Human capital is mostly affected by: • • • • formal education and job-related training informal education and job-related experience innate talents the physical and mental health of the worker26 Obviously, different kinds of jobs require different kinds of human capital. Different levels of human capital often result from different levels of investment, in terms of education and training. The wages for skilled occupations, such as aerospace engineers (e.g., compared to farm manual laborers), reflect in part the fact that aerospace engineers have specific formal training, whereas farm laborers use more common skills that many people possess. We can see the impact of education levels in Figure 9.5, which shows median earnings by education level in the United States. Those with a master’s degree earn about 18 percent more than those with just a bachelor’s degree, and those with a bachelor’s degree earn 67 percent more than those with just a high school diploma. The income benefits associated with education increased in recent decades. According to analysis by the U.S. Census Bureau, the increase in the economic benefits of education DRAFT 11 may be explained by both the supply of labor and the demand for skilled workers. In the 1970s, the premiums paid to college graduates dropped because of an increase in their numbers, which kept the relative earnings range among the educational attainment levels rather narrow. Recently, however, technological changes favoring more skilled (and educated) workers have tended to increase earnings among working adults with higher educational attainment, while, simultaneously, the decline of labor unions and a decline in the minimum wage in constant dollars have contributed to a relative drop in the wages of less educated workers.27 Of course, educational attainment is one factor employers look at when evaluating potential employees. A firm that is seeking to employ an aerospace engineer would look for someone with the appropriate degree. A college degree is a common screening method that employers use to limit their job search to specific candidates. Other screening methods used by employers include requirements that applicants have a minimum number of years of work experience, or certification that they have been trained to do specific tasks. Figure 9.5. Median Weekly Earnings of U.S. Workers, by Educational Attainment, 2021 Source: U.S. Bureau of Labor Statistics, 2021a. Note: Professional degrees include medical, dentistry, veterinarian, and law degrees beyond the bachelor’s level. screening methods: approaches used by employers to limit their job search to specific candidates In a subtler, but also common, example, firms may require credentials such as a college degree, or even a graduate degree, not because they are convinced that the undergraduate or graduate education has directly provided essential knowledge or skills but because the possession of the degree signals that the person is a certain kind of worker. The signaling theory of the value of education suggests that the value DRAFT 12 of a college education may be not so much in the way it creates human capital as in how it solves information problems for employers, signaling that a potential employee has the organization and self-discipline needed to complete a college degree— qualities that will also be essential in the workplace. signaling theory: a theory of the value of an education that suggests that an educational credential signals to an employer that a potential worker has desired character traits and work habits In the neoclassical framework, a worker’s productivity, and consequently his or her wages, may depend on other factors besides human capital. One example is the level of work effort, such as the pace at which a worker works, and the attention paid to a task. This not only depends on the worker’s motivation, but also on employer management practices, which could include both positive reinforcement or encouragement and negative reinforcement or discipline. In some jobs, workers are paid on a “piecework” system, meaning a fixed amount for each unit produced, so that greater productivity means higher income. Finally, a worker’s productivity depends on the quantity and characteristics of the resources available to them. In the simplest terms, those who work with more, newer, and better equipment and technology are more productive and, according to the neoclassical model, likely to receive a higher wage. 2.2 SOCIAL NORMS, BARGAINING POWER, AND LABOR UNIONS As suggested by some of the examples that we have already mentioned the neoclassical labor model does not explain all variations in wages across occupations and across workers. We now turn to various other factors that influence wages. As we saw in Chapter 5, wages may be prevented from reaching equilibrium due to minimum wage laws. But even well above the relevant minimum wage, social norms may prevent wages from adjusting downward due to market forces. According to the neoclassical labor model, wages should adjust downward when demand decreases or supply increases. In reality, employers are usually slow to offer significant wage reductions, and people will resist taking the offers if made, because the social norm is for wages to adjust upward, not downward, over time. In addition to norms, an essential aspect of most labor markets is the bargaining power on each side. One obvious situation that allows a firm to keep wages low is if it is the only employer to whom a certain group of workers can look for work. This is called a condition of monopsony. In the 1900s, for example, some manufacturing companies (including Hershey’s for chocolate and Pullman for railway cars) set up “company towns” in which they were the sole major employer. Remote mining towns and logging camps are other examples. In such cases, workers may have to accept the company’s low wages as the price of keeping their jobs—unless they have the ability and the determination to leave the area. Immigrant workers today are often in the position of being bound to a single employer, perhaps because the employer has impounded their passports or, if they lack proper documentation, has threatened to report them to immigration authorities. monopsony: a situation in which there is only one buyer but many sellers. This situation occurs in a labor market in which there are many potential workers but only one employer DRAFT 13 More common in labor markets are cases of oligopsony, in which there are just a few buyers. For example, if someone is looking for work as a supermarket stocker or salesclerk, there are just a few major supermarket chains, giving these large employers considerable power in setting wages. In theory, even a few large employers could compete against each other for employees, thereby bidding wages up, but in situations where the employees do not have specialized skills this is unlikely. oligopsony: a situation in which there are only a few major buyers but many sellers. This situation occurs in a labor market when there are many potential workers but just a few large employers Labor unions are legally recognized organizations that collectively bargain for their members regarding wages, benefits, and working conditions. Unions first appeared in the mid-nineteenth century, but they were not legally recognized in the United States until 1935. As seen in Figure 9.6, membership in labor unions in the United States peaked in the mid-1950s, when over one-third of all wage and salary workers were unionized. Since then, membership in unions has gradually but steadily declined. In 2020, less than 11 percent of workers in the United States belonged to a union. Labor union membership is much higher in the public sector than the private sector. About 35 percent of government employees, but only 6 percent of private sector workers, belong to a union. labor unions: legally recognized organizations that collectively bargain for their members (workers) regarding wages, benefits, and working conditions Figure 9.6. Union Membership in the United States, 1950–2020 Source: U.S. Bureau of Labor Statistics, 2021b. One of the reasons for the decline in union membership in recent decades has been an anti-union regulatory environment. Perhaps most famously, in 1981 President Ronald Reagan responded to an illegal strike by air traffic controllers, who were federal DRAFT 14 employees, by giving them 48 hours to return to work or face termination, resulting in many losing their jobs permanently. More recently, since 2011 states such as Wisconsin and Indiana have passed new laws limiting the power of labor unions. Another reason for the decline of labor unions has been a shift in employment from traditional unionized occupations such as manufacturing to service occupations in which it is more difficult to unionize, such as retail and restaurant workers. Union membership rates are higher in most other industrialized countries. For example, union membership is 14 percent in Australia, 24 percent in the United Kingdom, 33 percent in Italy, 50 percent in Norway, and 65 percent in Sweden.28 However, even in most of these countries union membership rates have been declining in recent years. Labor unions have been effective at providing well-paying jobs for their members. According to the U.S. Bureau of Labor Statistics, the average weekly earnings of unionized private sector workers in 2020 were $1,144 per week, compared to earnings of $958 for non-union workers.29 Union workers are also more likely to have employer-provided benefits such as health insurance and paid vacations. Some economists argue that unions push wages to above-market levels.30 According to this view, while unions were probably necessary to counter the excessive power of corporations in much of the twentieth century, they have more recently become a source of market inefficiency. Other economists see labor unions as a necessary way for workers to bargain on an equal footing with management. The decline of unions is widely considered a contributing factor in the rise of economic inequality in the United States, which we discuss in more detail in Chapter 10. Also, the benefits of unions may extend beyond those who actually belong to them: [Unions] affect non-union pay and practices [by instituting] norms and practices that have become more widespread throughout the economy, thereby improving pay and working conditions for the entire workforce. . . . Many fringe benefits, such as pensions and health insurance, were first provided in the union sector and then became more commonplace. Union grievance procedures, which provide due process in the workplace, have been adapted to many non-union workplaces. . . . [Unions] remain a source of innovation in work practices (e.g., training and worker participation) and in benefits (e.g., childcare, work-time flexibility, and sick leave).31 2.3 EFFICIENCY WAGES AND DUAL LABOR MARKETS Economists have theorized that employers may sometimes pay wages somewhat above the market-determined level as a way of motivating and retaining workers. Efficiency wage theory proposes that workers will work harder and “smarter” when they know that their employer is paying them more than they could receive elsewhere. Because these wages are above the market-clearing level, there is likely to be a queue of potential workers who would like to get the relatively high wages. This fact adds to employee motivation, because they understand that if they were to shirk and be fired there would be plenty of applicants for their position. Employees may also be motivated by a sense of gratitude, or identification with the firm, because we tend to like people who treat us well. Thus, it is theorized that efficiency wages can be profit maximizing: The cost to the firm of the extra wages may be more than made up for by the superior work effort and loyalty that they elicit. (See Box 9.2 for more on the potential benefits of efficiency wages.) DRAFT 15 efficiency wage theory: the theory that an employer can motivate workers to put forth more effort by paying them somewhat more than they could get elsewhere BOX 9.2 GOOD JOBS ARE GOOD FOR BUSINESS According to a recent study,32 providing employees with “good” jobs and paying efficiency wages can frequently be good for business, too. This idea runs counter to prevailing notions of cost minimization. The conventional wisdom is that many companies have no choice but to offer “bad” jobs—especially retailers whose business models entail competition by offering low prices. If retailers invest more in employees, customers will have to pay more, so the assumption goes. Several businesses in the study provide their employees with “good” jobs, including Trader Joe’s and Costco. Trader Joe’s starting salary of around $40,000 per year is about twice what many of its competitors offer. Costco’s wages are about 40 percent higher than those of their main competitor, Walmart’s Sam’s Clubs. Both Trader Joe’s and Costco also offer good opportunities for advancement. Turnover at these companies is low, and employee morale is relatively high. They are both also known for high-quality customer service. The study found that, rather than hurting these firms’ profits, they actually financially outperform their competitors. For example, annual revenues per square foot are $986 at Costco, but only $588 at Sam’s Club. Sales per square foot at Trader Joe’s are about three times that of a typical U.S. supermarket. Companies offering well-paying jobs also typically institute policies that promote worker efficiency, including training workers for a variety of tasks and allowing them to make relatively small decisions on their own. The study concludes: Today many retail managers believe that there is a tradeoff between investing in employees and offering the lowest prices. That is false. Retailers that persist in believing in it forgo the opportunity to improve their own performance and contribute the kind of jobs the U.S. economy urgently needs. When backed up with a specific set of operating practices, investing in employees can boost customer experience and decrease costs. Companies can compete successfully on the basis of low prices and simultaneously keep their customers and employees happy.33 Of course, not all workers benefit from efficiency wages. The theory of dual labor markets is based on the idea that there can be different sectors within a labor market. Workers in the “primary” sector of a labor market receive high (possibly efficiency) wages, opportunities for advancement, job security, and perhaps other favorable working conditions. Employment in the “secondary” sector, by contrast, is more closely driven by market conditions. These workers generally receive lower wages, enjoy few opportunities for advancement (even if they increase their human capital), and have little job security. dual labor markets: a situation in which primary sector workers enjoy high wages, opportunities for advancement, and job security, while secondary sector workers are generally hired with low wages, no opportunities for advancement, and little job security DRAFT 16 Such labor market segmentation may take place across firms. A primary sector of large, established firms may exist side-by-side with a secondary sector of smaller organizations that are more subject to competitive pressures. Dual labor markets may also exist within a single organization. For example, a firm may employ regular workers with health and retirement benefits and, alongside them, hire temporary workers on short contracts with no benefits. In many colleges and universities, tenured faculty constitute the “primary” workforce. Lecturers, adjuncts, and research associates, who constitute a secondary workforce, are hired as the need arises—and let go when the need falls. Such a structure allows an employer to keep a loyal core of employees and to avoid making new long-term commitments in times of temporary high demand. But individual workers may become stuck in the secondary sector as they have fewer opportunities to build up human capital and may quickly develop an “unstable-looking” work history. An extreme type of dual labor market is what some economists call a “winnertake-all” market, such as the one for star athletes, famous actors, and top corporate executives. In such markets, the rewards for being in first place are vastly greater than the rewards for being a step down, even if the actual difference in talents and skills between the top tier and the next is small. Welfare analysis applied to such markets would find significant inefficiency. Very few people can actually get into the top tier in, for example, Olympic sports or an acting career. Yet the rewards for winning are so appealing that many individuals devote huge amounts of time and effort trying to “reach for the gold.” Except in cases in which the effort to be the best is rewarding in itself, those who unsuccessfully devote their lives to the effort would probably have happier, more productive lives working toward different goals. 2.4 DISCRIMINATION Labor market discrimination exists when, among similarly qualified people, some are treated disadvantageously in employment on the basis of race, sex, age, sexual preference, physical appearance, or disability. Workers who belong to disfavored groups may be paid less for the same work, may be denied promotions, or may simply be excluded from higher-paying and higher-status occupations. labor market discrimination: a condition that exists when, among similarly qualified people, some are treated disadvantageously in employment on the basis of race, sex, age, sexual preference, physical appearance, or disability Historically, much labor market discrimination, particularly against African Americans and other minorities, was based on racist beliefs that certain groups were innately inferior. Some discrimination against women was similarly based on sexist notions of inferiority. Gender discrimination is also historically rooted in social norms that reserved better-paying jobs for men (who were assumed to be supporting families), while making women (who were assumed to have husbands to rely on) solely responsible for providing unpaid household labor and family care. Discriminatory attitudes are not just held by employers. They may also be held by customers or co-workers, which can pose a dilemma for employers, even if they themselves are not prejudiced. For example, suppose that a law firm hires a skilled minority lawyer, but racially biased clients prefer to be represented by EuropeanAmerican lawyers. The firm may find that the new lawyer attracts little business to the firm, and thus may receive lower pay and not be considered for promotion. Such DRAFT 17 discrimination can be eliminated only by socially coordinated action or major changes in social norms. In Figure 9.7 we compare median weekly earnings in the United States of fulltime, year-round workers in various groups, using government data from 2021. Median earnings are at a level where half the people in the group make more and half less. We see that median earnings vary significantly by both race and sex. The median earnings of black male workers were only 75 percent of the earnings of their white male counterparts, and the median earnings of Hispanic male workers were only 70 percent of white male earnings. Disparities among female workers of different races also exist, although the differences are somewhat less pronounced. White female workers only earn 82 percent of the earnings of their white male counterparts. Gender disparities are also evident among male and female workers of other races. The data in Figure 9.7 are not necessarily evidence of wage discrimination. Some variations in wages may be due to factors outside the labor market itself, such as differences in experience, education, and occupational choice (although some of these differences may also be a result of discrimination and outdated social norms). For example, we saw earlier in the chapter that educational attainment can have a significant impact on earnings. Education levels vary by race. About 40 percent of white individuals over age 25 have a bachelor’s degree or higher, but only 26 percent of black and 19 percent of Hispanic individuals have at least a bachelor’s degree.34 So differences in education levels may explain the variation in wages by race, rather than discrimination. But again, differences in educational attainment may be a result of past, and current, discrimination not in the labor market but in terms of access to education. Figure 9.7. Median Weekly Earnings, Select Groups of U.S. Workers, Age 25 and Over, 2021 Source: U.S. Bureau of Labor Statistics, 2021d. Educational attainment does not vary significantly by gender in the United States (actually, a slightly higher percentage of women have a college degree or higher).35 So the differences in earnings between male and female workers in Figure DRAFT 18 9.7 cannot be attributed to differences in education. Part of the explanation for women’s lower earnings is that women have traditionally had less work experience than men, on average. Men as a group have tended to work more continuously at their jobs, whereas, given social norms and, sometimes, individual preferences or requirements concerning family responsibilities, many mothers participate in the labor market less than full-time or not at all when their children are young, leading to reduced advancement opportunities. Another important factor in explaining earnings differences by gender is occupational segregation—the tendency of men and women to be found in different kinds of jobs. For example, in the United States, jobs like bookkeeper, dental hygienist, child-care worker, registered nurse, and preschool teacher are held overwhelmingly by women. Meanwhile, men dominate in occupations such as construction trades, metal working, truck driving, and engineering. Occupational segregation could be a result of differences in preferences, or it could also reflect discrimination. For example, existing stereotypes may lead more women to become nurses while doctors are more likely to be men. occupational segregation: the tendency of men and women to be employed in different occupations gender wage gap: the difference in average wages between men and women; women are paid, on average, less than men The gender wage gap—the difference in average wages between men and women—has declined in the United States in recent decades, as shown in Figure 9.8 (which repeats the figure from Chapter 0).36 In 1970 women’s average wages were only 58 percent of men’s wages. By 2020 the ratio had increased substantially, although even then women earned only 82 percent of men’s wages. Figure 9.8 shows that the gender wage ratio tended to rise prior to 2000, remain somewhat constant in the 2000s, and rise again during the 2010s. According to a comprehensive 2016 analysis, about half of the difference between men’s and women’s pay in the United States is associated with differences in industry and occupation choice.37 The study also concludes that workforce interruptions, such as taking time off to raise children or care for family members, also help explain why women earn less than men, on average. However, even after accounting for gender differences in education, experience, occupational choice, and other variables, about 40 percent of the gender pay gap remains unexplained. At least part of this unexplained difference can most likely be attributed to discrimination. According to the International Labour Organization, women are paid about 16 percent less than men globally.38 While data are not available for all countries, the gender wage gap is the highest, above 30 percent, in India, Pakistan, and South Korea. Developed countries with a comparatively low gender wage gap include Australia (13 percent), Italy (8 percent), and Belgium (7 percent). Women are paid more than men, on average, in more than a dozen countries including Argentina, Costa Rica, Thailand, Bangladesh, and the Philippines. DRAFT 19