Chapter 8 Jobs in the Macroeconomy

Unemployment in the United States and Europe

Figure 8.2 "Unemployment Rates in France, the United States, and the Euro Area, 1985-2011" shows quarterly unemployment rates for the United States, France, and Europe as a whole. In the late 1980s, unemployment fell in both the United States and France, although the US unemployment rate was about two percentage points lower than the French rate. The 1990s were a different story. Unemployment rates increased in both countries at the beginning of the decade. Thereafter, the unemployment rate decreased in the United States, but it continued to increase in France for about half of the decade and decreased only near the end of century. From the early 1990s up to about 2008, the unemployment rate in Europe was substantially higher than that in the United States. The pattern for Europe as a whole closely matches the pattern for France, although unemployment in France is typically a little higher than the European average.

The crisis of 2008, however, led to a dramatic rise in the unemployment rate in the United States. At the end of 2007, the US unemployment rate was just under 5 percent. Two years later, at the start of 2010, the rate was over 10 percent. Unemployment also rose in Europe, but to nothing like the same degree. In early 2011, US and European unemployment rates were almost identical.

One other feature of the data is noticeable: there is a regular seasonal pattern in the data. For example, in the United States, unemployment is almost always higher in the first quarter of the year than it is in the preceding or following quarter. This is because some sectors of the economy are heavily affected by seasonal patterns. For example, stores may hire extra people during the Christmas holiday period, while construction firms may employ fewer people during the winter months. Sometimes, data such as these are “seasonally adjusted” to remove these effects.

The French labor law reforms with which we began the chapter were aimed at young workers, so let us also look specifically at the unemployment experience of this group. Between 2000 and 2010, the unemployment rate in France for the age group 20–24 ranged between 17 and 21 percent, with an average of 18.6 percent. In the United States, in contrast, for the same period and the same group of workers, the unemployment rate averaged 10 percent.The figures on youth unemployment come from “Statistics Portal: Labour,” OECD, http://www.oecd.org/topicstatsportal/0,2647,en_2825_495670_1_1_1_1_1,00.html#499797. In both countries, the unemployment rate is higher for younger workers than the overall unemployment rate.

Although there are some similarities between France and the United States, there is also a clear puzzle: unemployment, for both the overall population and young workers, was, until very recently, much higher in France. We need to understand the source of this difference before we can evaluate different policy remedies.

The Labor Market

Unemployment suggests a mismatch between supply and demand. People who are unemployed want to have a job but are unable to find one. In economic language, they are willing to supply labor but cannot find a firm that demands their labor. The most natural starting point for an economic analysis of unemployment is therefore the labor marketThe market that brings together households who supply labor services and firms who demand labor as an input into the production process..

The labor market is depicted in Figure 8.3 "Labor Market". “Price” on the vertical axis is the real wageThe nominal wage (the wage in dollars) divided by the price level., which is the nominal wage divided by the price level. It tells us how much you can obtain in terms of real goods and services if you sell an hour of your time. Recalling that the price level can be thought of as the price of a unit of the real gross domestic product (real GDP), you can equivalently think of the real wage as the value of your time measured in units of real GDP.

At a higher real wage, households supply more labor. There are two reasons for this. First, a higher real wage means that, for the sacrifice of an hour of time, households can obtain more goods and services than before. Households are therefore induced to substitute away from leisure to work and ultimately consume more. Second, as the wage increases, more individuals join the labor force and find a job. Embedded in the upward-sloping labor supply curve is both an increase in hours worked by each employed worker and an increase in the number of employed workers.

At a higher real wage, firms demand fewer labor hours. A higher real wage means that labor time is more expensive than before, so each individual firm demands less labor and produces less output. The point where the labor supply and demand curves meet is the equilibrium in the labor market. At the equilibrium real wage, the number of hours that workers choose to work exactly matches the number of hours that firms choose to hire.

Supply and demand in the labor market determine the real wage and the level of employment. Variations in either labor supply or labor demand show up as shifts in the curves. If we want to talk about unemployment, however, the labor market diagram presents us with a problem. The idea of a market is that the price adjusts to reach equilibriumAn equilibrium price and an equilibrium quantity such that the quantity supplied equals the quantity demanded at the equilibrium price.—the point where supply equals demand. In the labor market, this means the real wage should adjust to its equilibrium value so that there is no mismatch of supply and demand. Everyone who wants to supply labor at the equilibrium wage finds that their labor is demanded—in other words, everyone who is looking for a job is able to find one.

Remember the definition of unemployment: it is people who are not working but who are looking for a job. The supply-and-demand framework has the implication that there should be no unemployment at all. Everyone who wants to work is employed; the only people without jobs are those who do not want to work.

Theories of Unemployment

So where do we go from here? One natural approach is to start from Figure 8.3 "Labor Market" but look for circumstances in which we would see unemployment. Figure 8.4 "Unemployment in the Labor Market" shows us that there will be unemployment if the real wage in the market is too high—that is, above the equilibrium real wage. In this case, the amount of labor that workers want to sell is greater than the amount that firms want to buy. Some workers will want a job at this wage but be unable to find one. They will be unemployed.

Figure 8.4 "Unemployment in the Labor Market" shows us what the labor market must look like for there to be unemployment, but it is hardly an explanation of unemployment. Economists typically expect markets to look like Figure 8.3 "Labor Market", not Figure 8.4 "Unemployment in the Labor Market". That is, they think that the price in a market—in this case, the real wage—adjusts quickly to ensure that supply equals demand. If we want to explain unemployment with a picture like Figure 8.4 "Unemployment in the Labor Market", we also need some story of why real wages might be sticky, so they remain above the equilibrium wage.

Facts

Our starting point is the classification of individuals in the civilian working age population. Recall that economic statistics place them as one of the following: employed, unemployed, or not in the labor force. Imagine taking a snapshot of the US economy each month. For a given month, you would be able to count the number of people employed, unemployed, and out of the labor force. We could call these the stocks of each kind of individual.

Figure 8.5 "Worker Stocks in the United States" shows the number of people between 16 and 64 years old in the United States in three different “states”—employment, unemployment, and out of the labor force—over the period 1996–2003.These data come from a study using a monthly survey conducted by the Bureau of Labor Statistics (BLS) called the Current Population Survey and were compiled by Stephen J. Davis, R. Jason Faberman, and John Haltiwanger. The numbers here come from S. Davis, R. J. Faberman, and J. Haltiwanger, “The Flow Approach to Labor Market: New Data Sources and Micro-Macro Links” NBER Working Paper #12167, April 2006, accessed June 30, 2011, http://www.nber.org/papers/w12167. On average, there were 122 million people employed, 6.2 million unemployed, and 59.3 million considered out of the labor force. Adding these numbers together, there were 187.5 million working-age individuals, of whom 128.2 million were in the labor force. The average unemployment rate was 4.8 percent over this period, and the employment rate was 95.2 percent. Notice, though, that many individuals are out of the labor force: only 65 percent of the population is employed.

Figure 8.5 "Worker Stocks in the United States" shows an average over many months, but you could also look at how these numbers change from month to month. Even more informatively, you could count the number of people who were employed in two consecutive months. This would tell you the likelihood of being employed two months in a row. These calculations for the US economy are summarized in Figure 8.6 "Worker Flows in the United States".

Look, for example, at the arrows associated with the box labeled unemployed. There are two arrows coming in: one from the employed box and one from the out-of-the-labor-force box. There are two arrows going out: one to the employed box and one to the out-of-the-labor-force box. Each of these four arrows has a percentage attached, indicating the fraction of people going from one box to another. Thus, on average, 28.3 percent of the unemployed people in one month are employed in the next and 23.3 percent leave the labor force. The remaining 48.4 percent stay in the group of unemployed.

The numbers in the figure are averages over a long period. Such flows change over the course of the year due to seasonal effects. Around Christmas, for example, it may be easier for an unemployed worker to find a job selling merchandise in a retail shop. These flows also change depending on the ups and downs of the aggregate economy.

Do European countries exhibit similar patterns? Portugal makes for a good comparison with the United States because the unemployment rates in the two countries were broadly similar over most of the last two decades. Yet Portugal has very strong employment protection laws, to the point where they are enshrined in the Portuguese Constitution:“Article 53,” Portugal-Constitution, adopted April 2, 1976, accessed June 30, 2011, http://www.servat.unibe.ch/icl/po00000_.html#A053_.

A study that compared the labor markets in Portugal and the United States uncovered the following facts:See Olivier Blanchard and Pedro Portugal, “What Hides Behind an Unemployment Rate: Comparing Portuguese and U.S. Labor Markets,” American Economic Review 91, no. 1, (2001), 187–207.

• The flows into unemployment from employment and the flows from employment to unemployment are much lower in Portugal compared to the United States.
• Average unemployment duration in Portugal is about three times that of the United States.
• Job protection is very high in Portugal relative to the United States.

Even though Portugal and the United States have similar overall unemployment rates, the underlying flows are quite different in the two countries. Flows between employment and unemployment—and vice versa—are much smaller in Portugal. This means that if you lose your job, it is likely to take a long time to find a new one. If you have a job, you are likely to keep it for a long time. As we would expect from this, people typically spend much longer periods of time in unemployment in Portugal than they do in the United States.

If we compare the United States with Europe more generally, we see similar patterns. In 2010, the average unemployment durationThe amount of time a typical worker spends searching for a new job. for workers ages 15–24 was about 10.6 months in Europe but only 5.9 months for the United States. For workers in the 25–54 age group, the duration was higher in both Europe (13.7 months) and the United States (8.2 months) than for younger workers.See “Unemployment Duration,” Online OECD Employment database, accessed June 30, 2011, http://www.oecd.org/document/34/0,3746,en_2649_33927_40917154_1_1_1_1,00.html#uduration. Recall that in 2010, Europe and the United States had similar rates of unemployment. Employment duration, however, is still much higher in Europe than the United States. In both places, older workers tend to be unemployed for longer periods than younger workers. But European workers are typically unemployed for much longer periods of time than US workers.These figures come from “Average Duration of Unemployment,” OECD, accessed June 30, 2011, http://stats.oecd.org/Index.aspx?DataSetCode=AVD_DUR.

The Organisation for Economic Co-operation and Development (OECD) conducted a large study on the employment protection legislation in a variety of developed countries. The main study (OECD Employment Outlook for 2004, http://www.oecd.org/document/62/0,3746,en_2649_33927_31935102 _1_1_1_1,00.html) created a measure of employment protection and then attempted to relate it to labor market outcomes in different countries. The reasoning we have just presented suggests that in countries with relatively high levels of employment protection, labor markets would be much more sluggish.

Formulating a comprehensive measure of employment protection is not easy. In principle, the idea is to measure the costs of firing workers and various regulations of employment. Examples would include requirements on advance notice of layoffs and the size of severance payments that firms are obliged to pay. In some countries, a firm must go to court to lay off workers. For temporary workers, there are specific restrictions placed on this form of contract, as in the discussion of France that opened this chapter. In reality, these costs are difficult to detect and convert to a single measure. The OECD findings should be interpreted with these challenges in mind.

Another OECD publication (http://www.oecd.org/dataoecd/40/56/36014946.pdf) examines employment protection legislation across OECD countries in 1998 and 2003.This discussion is based on Figure A.6 of OECD, “Annex A: Structural Policy Indicators,” Economic Policy Reforms: Going for Growth, accessed June 30, 2011, http://www.oecd.org/dataoecd/40/56/36014946.pdf. Portugal was the country with the highest level of employment protection legislation, while the United States was the lowest. France was above average, while the United Kingdom and Canada were below average. The OECD analysis highlighted two effects of such legislation on labor market flows:

1. It limits flows from employment into unemployment because it is costly to fire workers.
2. It limits flows from unemployment to employment because firms, when deciding to hire a worker, will realize that they may wish to fire that worker sometime in the future.

The first effect is the more obvious one; indeed, it provides the rationale for employment protection. If it is hard to fire workers, then firms are less likely to do so. The second effect is less obvious and more pernicious. If it is hard to fire workers, then firms become more reluctant to hire workers. Put yourself in the place of a manager wondering whether to make a hire. One concern is that the person you are considering will turn out to be unsuitable, or a bad worker. Another is that conditions in your industry will worsen, so you may not need as many employees. In those circumstances, you want to be able to let the worker go. If you will not be able to do so, you may decide it is safer simply to make do with the workers you already have.

The OECD analysis particularly stressed the effects on the labor market experience of relatively young workers. The report emphasized that stronger legislation is linked to lower employment of young workers. If it is costly to sever a relationship, then a firm will not give a young worker a chance in a new job. The OECD also noted an important benefit of employment protection legislation: it enhances the willingness of young workers to invest in skills that are productive at their firms. Without a strong attachment to the firm, workers have little incentive to build up skills that are not transferable to other jobs.

Job Creation and Job Destruction

In place of the supply-and-demand diagram, we can think about the decisions that workers and firms make when they are trying to form or break an employment relationship. Individual workers search for available jobs, which are called vacancies. On the other side, vacancies are searching for workers. When a vacancy and a worker are successfully matched, a job is created. When we say that a vacancy is searching for a worker, we, of course, really mean that a firm with a vacancy is seeking to hire a worker. You can think of a firm as being a collection of jobs and vacancies.

Whereas the standard supply-and-demand picture downplays differences among workers and jobs, this “search-and-matching” approach places these differences at the center of the analysis. Workers differ in terms of their abilities and preferences. Jobs differ in terms of their characteristics and requirements. For an economy to function well, we need to somehow do a good job of matching vacancies with workers. When a successful match occurs, we call this “job creation.”

Search theoryA framework for understanding the flows of workers across periods of employment and unemployment along with the creation of job vacancies by firms. is a framework for understanding this matching process. Let us think about how this process looks, first from the perspective of the worker and then from the perspective of the firm. Workers care about the various characteristics of their jobs. These characteristics might include how much the job pays, whether it is in a good location, whether it offers good opportunities for advancement, whether it is interesting, whether it is dangerous, and other attributes.

Vacancies are likewise “looking” for certain characteristics of workers, such as how much they cost, what skills they possess, whether they have relevant experience, whether they are hardworking and motivated, whether they are trustworthy, and so on. The firm cares about these characteristics because it cares about profitability: its goal is to make as much profit as possible.

Over time, the quality of the match between a worker and a vacancy may change. A job may become less profitable to the firm and/or less attractive to the worker. To put it another way, the amount of value created by the job may change. The worker may come to dislike particular aspects of the job or may wish to change location for family reasons. The worker may feel that he or she would be better matched with some other firm, perhaps because of changes in his or her skills and experience. From the firm’s side, demand for the firm’s product may decrease, or the firm might shift to a new production technique that requires different skills. If the value created by a job decreases too much, then the firm or the worker may choose to end the relationship, either by the worker’s choice (quitting the job) or the firm’s (firing the worker). This is “job destruction.”

Jobs are created and destroyed all the time in the economy. The flows of workers among jobs and employment states are a key characteristic of the labor market. As these flows occur, workers often spend time unemployed. After a job is destroyed, the worker may spend some time unemployed until he or she finds a job with a different firm.

Labor Flows and Productivity

In a rapidly changing economy, the value of different jobs (worker-firm matches) changes over time. To function efficiently, the labor market needs to be able to accommodate such changes. For this discussion, we will think about efficiency as simply being measured by the productivity of the match between workers and firms. In an efficient match, the worker is productive at the chosen job. For the overall economy, if all matches are efficient, then it is not possible to change the assignment of workers to jobs and produce more output.

The Natural Rate of Unemployment

We expect there to be some frictional unemployment, even in a well-functioning economy. We also know that there is cyclical employment associated with the ups and downs of the business cycle. When cyclical unemployment is zero, we say that the economy is operating at full employment. The natural rate of unemploymentThe amount of unemployment we expect in an economy that is operating at full employment. is defined as the amount of unemployment we expect in an economy that is operating at full employment—that is, it is the level of unemployment that we expect once we have removed cyclical considerations.

The natural rate of unemployment can seem like an odd concept because it says that it is normal to have unemployment even when the economy is booming. But it makes sense because all economies experience some frictional unemployment as a result of the ongoing process of matching workers with jobs. Government policies that affect the flows in and out of employment lead to changes in the natural rate of unemployment.

Hours Worked in Europe and the United States

Figure 8.7 "Hours in Europe Relative to the United States" and Figure 8.8 "Annual Hours in Various Countries" show some basic facts about hours worked in the United States and Europe.Richard Rogerson, “Understanding Differences in Hours Worked,” Review of Economic Dynamics 9 (2006): 365–409. Figure 8.7 "Hours in Europe Relative to the United States" shows how hours worked in a number of different European countries compare to hours worked in the United States. More precisely, it shows the total hours worked by individuals between 15 and 64 years old divided by the number of people in that age group. The table does not distinguish by employment status: all working age people are counted, not just employed people.

Three of the largest European countries—France, Germany, and Italy—average less than 75 percent of the hours worked in the United States. Part of this difference is due to longer holidays in Europe, and part is due to the fact that the workweek in Europe is typically shorter. Because the table counts all working age people, the higher unemployment rate in Europe also contributes to the difference.

Figure 8.8 "Annual Hours in Various Countries" looks at the hours worked in various countries over the 40 years from 1970 to 2009. The measure of hours is calculated in the same manner as Figure 8.7 "Hours in Europe Relative to the United States". Average hours worked have declined significantly in most of these countries. Meanwhile average hours worked in the United States have been more or less flat over these four decades. As a result, hours worked are now significantly higher in the United States than in any of these countries.The data come from OECD (2010), "Hours Worked: Average annual hours actually worked", OECD Employment and Labour Market Statistics (database). doi: 10.1787/data-00303-en (Accessed on 18 October 2011) http://scholar.harvard.edu/alesina/files/work_and_leisure_in_the_u.s._and_europe.pdf. Figure 1 shows a similar pattern of divergence in hours worked for employed people, though the hours worked per employed person has declined in all countries over this period.

Research by the Nobel Prize–winning economist Edward Prescott paints a similar picture. He reports that from 1993 to 1996, the hours worked per person in France were about 68 percent of the level in the United States. In addition, US output per person was much higher than in Europe. Prescott explains this difference based on the number of hours worked, not by differences in output per hour worked. In other words, the United States is richer, not because it is more productive but simply because people work more.

Where Do Differences in Hours Worked Come From?

The immediate question is, why do people work more in the United States? A natural place to look for explanations is the labor supply decisions of households. One possibility is simply that the tastes of US and European households are different. Perhaps Europeans prefer having fewer goods and more leisure. Although this is possible, economists prefer to start from the presumption that people have broadly similar tastes and look first to see if there are other plausible explanations.

The differences in hours worked are not explained by Europeans having poorer technology. Both the United States and European countries are highly developed, so technologies used in one country are used in the others as well. Supporting this is the fact that, as we already noted, productivity does not appear to be lower in Europe.

Another candidate explanation is that there are differences in the tax system. Figure 8.9 "Labor Supply" shows an individual labor supply curveA curve that indicates how many hours of labor an individual supplies at different values of the real wage.—in either Europe or the United States. Notice in Figure 8.9 "Labor Supply" the wage on the vertical axis is the real wage after taxes. This is defined as follows:

real wage after taxes = real wage × (1 − tax rate).

In this equation, the tax rate is a marginal tax rate. This means that it is the tax paid on the extra amount you earn if you work a little bit more. Suppose the tax rate is 0.40 and your real wage per hour is $10. Then, if you work an extra hour, you pay $4 to the government, and you retain $6.

Figure 8.9 "Labor Supply" shows that an increase in the after-tax real wage will cause an individual to supply more time to the market and thus consume less time as leisure. The increase in the wage creates an incentive for the individual to substitute away from leisure because it has become more costly.

Suppose that we compare two identical individuals in Europe and the United States. If the marginal tax rate in Europe is higher than it is in the United States, then the after-tax wage in Europe will be smaller. Since labor supply is upward sloping, individuals in Europe will work less than individuals in the United States. For this to be a convincing explanation, two things must be true:

1. Marginal tax rates must be higher in Europe.
2. Labor supply must slope upward enough to match the differences in hours.

Marginal tax rates are indeed lower in the United States than in Europe. Recent research finds that the marginal tax rate on labor income is about 34.5 percent in the United States compared to 57.7 percent in Europe (Germany, France, Italy, and the United Kingdom).Alberto F. Alesina, Edward L. Glaeser, and Bruce Sacerdote, “Work and Leisure in the U.S. and Europe: Why So Different?” (Harvard Institute for Economic Research, Working Paper #2068, April 2005), accessed June 30, 2011, http://www.colorado.edu/Economics/morey/4999Ethics/AlesinaGlaeserSacerdote2005.pdf. So, if you work an extra hour and earn a pretax wage of $10, then you would keep $6.55 in the United States and $4.23 in Europe.

The evidence is also consistent with the view that labor supply increases as the after-tax real wage increases. Figure 8.10 "Differences in Hours Supplied" shows the implication of this. On the vertical axis are two different levels of the after-tax real wage: a low one for Europe and a higher one for the United States. These differences in the after-tax real wage translate into differences in hours, using the labor supply curve of an individual. Thus, as in Figure 8.10 "Differences in Hours Supplied", individuals in the United States work more hours than in Europe. As this is true for everyone in the labor force, this argument immediately translates into a statement about hours worked for the aggregate economy.

Can the difference in the after-tax real wage explain the observed difference in hours worked? This depends on how responsive labor supply is to changes in the real wage. Figure 8.11 "Responsive and Unresponsive Labor Supply" shows two labor supply curves. In one case (the solid curve), labor supply is very responsive to changes in the wage. Relatively small differences in taxes then have substantial effects on hours worked. In the other case (the dashed curve), labor supply is not very responsive to the wage. Differences in tax rates are then unlikely to be able to explain the differences in hours worked.

Prescott argues that the difference in taxes between the United States and Europe is enough to account for the differences in hours worked. To make this argument, Prescott holds fixed the labor supply curve (Figure 8.10 "Differences in Hours Supplied") across countries and asks how much of the observed difference in hours can be explained by tax policy. This is a movement along the labor supply curve because the vertical axis measures the after-tax real wage. To support this argument, however, Prescott assumes that labor supply is indeed quite responsive to changes in after-tax wages.

Unemployment Insurance

In Figure 8.6 "Worker Flows in the United States", we described the flow of workers between situations of employment, unemployment, and out of the labor force. We also argued that having a flexible labor market in which people can change jobs easily may more than compensate for the fact that people may sometimes spend time in unemployment.

But this is abstract economist-speak. People who lose their jobs, even if only temporarily, see their livelihood vanish. The reallocations of jobs that are beneficial to the economy as a whole may be costly, even devastating, to the affected individuals. For this reason, most developed economies have some kind of unemployment insurance to protect their workers. Unemployment insurance means that, if you are unemployed, you will receive some income from the government. Exactly how long you receive this income for and exactly how much you get depends on where you live. Some countries have much more generous unemployment insurance than others. Even if you live in the United States, the amount of insurance varies from state to state.

When it comes to buying car insurance, home insurance, or life insurance, households typically decide for themselves how much insurance to purchase. It is not a decision made by the government. Unemployment insurance is different: it is provided by the government rather than by private companies. This insurance is funded by taxes levied on firms and workers together.

The reason unemployment insurance is provided by the government is because it might be difficult for private firms to provide this coverage. Private insurance companies rely on the fact that not everyone makes claims on insurance at the same time. For example, a provider of home insurance knows that 20 percent of the houses that they insure will not burn down in the same month. But in a recession, the high rate of unemployment means that a lot of people claim benefits at the same time. If private insurers were providing the benefits, insurance companies might go bankrupt, leaving workers without insurance. The government, by contrast, can use its ability to borrow, so it can finance unemployment insurance in one year from tax receipts it will receive in the future.

In the United States, the amount of insurance you receive typically depends on how much you have earned over the past year. A rule of thumb is that workers get about 25 percent of their wage income paid back through unemployment insurance. Benefits are available for only 26 weeks, although this is usually extended when the economy is in a recession. Other countries have much more generous programs.“The Ins and Outs of Long-Term Unemployment,” OECD Employment Outlook 2002, accessed June 30, 2011, http://www.oecd.org/dataoecd/36/48/17652683.pdf. Table 4.1 provides an extensive cross-country comparison. In Denmark, for example, unemployment benefits are about 90 percent of labor income and can last for up to 4 years.

Unemployment insurance has two main effects. First, and most obviously, this insurance makes it easier for unemployed people to sustain their level of consumption until they regain employment. Thus this form of insurance helps support consumption smoothingThe idea that households like to keep their flow of consumption relatively steady over time, smoothing over income changes.. Second, unemployment insurance affects the incentives of the unemployed. If individuals know they will receive some income even when they are unemployed, they are more likely to be willing to search extensively for good jobs. Instead of feeling the need to take the first job that comes along, people can wait longer and search longer for a job that is a really good match.

Unemployment insurance therefore contributes to labor market flexibility. It is, however, tricky to decide just how much unemployment insurance should be provided. After all, if unemployment insurance is too generous, then unemployed workers will be tempted to defer getting a new job for a long time—perhaps indefinitely. For this reason, governments usually restrict the period of time for which a worker can collect insurance to provide an incentive for them to search for a job.

Firing Costs

Imagine that you are the human resources (HR) manager of a firm in the United States. Suppose that the demand for your firm’s product has declined, so you need to lay off some workers. You will be obliged to provide two weeks’ notice to them. In many cases, that will be the end of your firm’s obligations, although workers may sometimes be entitled to additional severance payments as part of their employment contracts. In the United States, employment contracts are largely a private matter between a firm and its workers. A firm cannot fire a worker for a discriminatory reason, but otherwise the government stays out of the contractual agreements among workers and firms. According to the Department of Labor, “In general, if the reason for termination is not because of discrimination on these bases, or because of the employee’s protected status as a whistleblower, or because they were involved in a complaint filed under one of the laws enforced by the Department of Labor (see Whistleblower and Non-Retaliation Protections), then the termination is subject only to any private contract between the employer and employee or a labor contract between the employer and those covered by the labor contract.”“Termination,” US Department of Labor, accessed June 30, 2011, http://www.dol.gov/dol/topic/termination/index.htm.

In other countries, matters are not so simple. Imagine now that you are the HR manager of a firm in Portugal. Your product demand has fallen off, and you want to reduce output. In contrast to the United States, you may not be able to simply lay off workers. In Portugal, and in many other countries, there are numerous laws that make it costly to dismiss workers.

If you want to design a public policy to reduce the unemployment rate, it is tempting to make it harder to fire workers. If it is difficult to fire people, then fewer individuals will move from employment into unemployment. As we discussed earlier, though, spells of unemployment are sometimes necessary if workers are to move from less productive jobs to more productive ones. An increase in firing costs makes the labor market less flexible, so the economy will adjust less effectively to changes in workers’ productivities.

There is also a more subtle unintended consequence of firing costs. If it is harder to fire workers, then firms become more reluctant to hire workers. Neither firms nor workers know the true value of a match in advance. When you take a part-time job, your productivity at that job and job satisfaction cannot be known ahead of time. Suppose there was a law that stated that once you accept a job you must stay with that employer for five years. You would certainly become very careful about deciding to accept a job offer. Exactly the same applies to firms. If the cost of laying off a worker is very high, then the firm will simply not hire the worker. A policy designed to promote employment can actively discourage it.

The French government, as we saw at the beginning of the chapter, made an attempt to introduce labor market reforms based on exactly this reasoning and tried to make the argument that we have just outlined to the protesters in the streets. If there were more flexibility in the firm’s employment decision, they argued, firms would become more willing to hire young workers. This would help to reduce youth unemployment. The following New York Times article tells what happened next.Elaine Sciolino, “Chirac Will Rescind Labor Law That Caused Wide French Riots,” New York Times, April 11, 2006, accessed June 30, 2011, http://www.nytimes.com/2006/04/11/world/europe/11france.html?_r=1.

President Jacques Chirac crumbled under pressure from students, unions, business executives and even some of his own party leaders on Monday, announcing that he would rescind a disputed youth labor law intended to make hiring more flexible. The retreat was a humiliating political defeat for both Mr. Chirac and his political protégé, Prime Minister Dominique de Villepin […]

It also laid bare the deep popular resistance to liberalizing France’s rigid labor market, and makes any new economic reform politically impossible before a new government is in place, and perhaps not even then.

“Dead and buried,” is how Jean-Claude Mailly, leader of the leftist union Force Ouvrière, described the fate of the labor law. “The goal has been achieved.”

[…]

The new law was intended to give employers a simpler way of hiring workers under 26 on a trial basis without immediately exposing companies to the cumbersome and costly benefits that make hiring and firing such a daunting enterprise. Opposition to the law reflects the deep-rooted fear among the French of losing their labor and social protection in a globalized world.

[…]

In its initial form, the law allowed employers to fire new employees within two years without cause. In the face of mounting pressure, Mr. Chirac watered it down so that employers could subject new employees to only a yearlong trial period, and then would have to offer a reason for any dismissal.

Students and unions, bolstered by support from the opposition Socialists and even some business leaders, had vowed to continue their street protests until the law was rescinded.

The Socialists were quick to proclaim victory on Monday. “This is an unquestionable retreat,” Francois Hollande, the leader of the Socialist Party, told reporters. “It is a grand success for the young and an impressive victory for the unity of the unions.”

[…]

Restrictions on Hours

Another tempting policy to increase employment is to limit the number of hours an employee can work. Suppose that a firm needs 1,200 hours of labor time a week. If a typical worker works 40 hours per week, then the firm will need to hire 30 workers. But if the government were to legislate a 30-hour workweek, then the firm would need to hire 40 workers instead.

This idea of “spreading work” through restrictions on hours was part of the response in the United States to the Great Depression. During the early 1930s, the US government instituted such restrictions under the heading of the “National Economic Recovery Act.” The idea persists to the present day. In France, the government passed a law limiting hours worked to 35 hours per week (for workers at large firms) starting in the year 2000. In Germany, the government operates a policy called Kurzarbeit, whereby it subsidizes firms who retain workers for shorter hours in times of recession.

One problem with such policies is that restrictions on hours reduce the value of a match between a worker and a firm. Consequently, fewer matches will be formed, and more workers will be unemployed. Another problem is that it reduces flexibility in the labor market, which leads to less efficient functioning of the economy.

As a concrete example, consider auto manufacturers in the years following the Great Depression in the United States. This industry had substantial variations in hours worked over the model year. During times of high demand for cars (the spring), factories and their workers were working overtime to meet the increased demand. Restrictions on hours meant that overtime working had to be replaced by increased hiring. Firms that wanted to produce more output had to hire and train new workers. This was costly, so firms sometimes found it was better simply to accept that they would not meet the high demand.

In the case of France’s 35-hour workweek, matters were a bit more complicated. The mandated short workweek imposed some rigidity on firms. However, during the negotiations for this change in the laws, French labor unions agreed to some other changes that improved the flexibility of the labor market. France later moved away from the 35-hour workweek by permitting firms and workers to agree to longer work hours if they wish.