Economics for the Common Good

How many unemployed people are there in France? Answering this question turns out to be devilishly difficult. The most inherently conservative estimate stems from the definition given by the International Labor Organization (ILO).2 It indicated 2.9 million unemployed during the third trimester of 2015, or an unemployment rate of about 10.6 percent. This was more than double the German level, for instance, and far higher than those of developed English-speaking countries or northern Europe. But this definition does not take into account almost 1.5 million unemployed persons who would like to work.³ Statistics from the French ministry of labor⁴ distinguish five types of jobseekers. The figure most commonly reported in the media is: “jobseekers expected to make active attempts to find work, but who are currently unemployed” (Category A). By this definition, there were 3,574,860 unemployed in metropolitan France in November 2015. Yet this Category A figure for unemployment also underestimates the total because it does not include the other categories of unemployed persons: those who are in training programs, doing internships, without a job but on sick leave or maternity leave, working on subsidized contracts or part time, etc.⁵ Including everyone in these categories, in November 2015 there were about 6,142,000 unemployed.

Two age groups suffer particularly from high unemployment in France. Those aged between fifteen and twenty-four have difficulty finding work. Their unemployment rate is 24 percent. Their employment rate (28.6 percent) is far lower than the OECD average (39.6 percent) and the rates in northern Europe (46.8 percent for Germany and 62.3 percent for the Netherlands).⁷ The French labor market is thus relatively closed to new entrants, particularly young people looking for their first job. To be sure, all countries have higher youth unemployment rates than among the rest of the population. Businesses are less eager to hire employees without prior experience, and especially to pay the cost of on-the-job training for young employees who, once they have completed their training or gained enough experience, may leave the company. However, young French people pay a harsh penalty compared with northern Europe or the English-speaking countries.

To reduce unemployment, successive governments in France (and in other southern European countries) have encouraged fixed-term contracts and subsidized jobs.

Subsidized jobs. These constitute, on the whole, a poor use of public funds, particularly for subsidized jobs in the noncommercial sector.¹² Rather than encouraging employers to use employees because it is cheap to do so, the money could be used instead to reduce social security contributions and thus incentivize companies to create the stable jobs they and their employees really need. Indeed, France has the highest social security contributions among OECD countries.¹³

In reality, employment on limited-term contracts is not good for either the employee or the employer. For the former, the contract offers little protection. Since in theory (although this is frequently ignored in practice) prolonging or renewing a limited-term contract legally transforms it into a permanent contract, regulation gives the employer a very strong incentive not to do so, even if the employee’s performance is satisfactory.¹⁵ This is precisely the opposite of what regulation is intended to achieve, in terms of employee protection. In fact, within Europe, France is the country where the transition from a temporary contract to a stable contract occurs the least frequently.¹⁶ Which means that elsewhere in Europe, a person hired on the basis of a temporary contract has a much greater chance of seeing this temporary contract transformed into a permanent job. The fact that companies resort extensively to fixed-term contracts, which neither they nor their employees like, is very revealing of the implicit cost imposed on French society by current laws regarding permanent contracts.

Even so, political debate is focused on dismissals of employees who have permanent contracts. Such dismissals represent only a small fraction of terminations (4.4 percent) and are supervised by the courts, with additional governmental pressure if the subject gets media coverage. On the other hand, the debate ignores almost completely the two principal causes of turnover in the labor market: resignations, which are few in number (9 percent of terminations) and decreasing, and above all terminations at the end of a fixed-term contract, which are far more numerous (77 percent) and are steadily increasing. (The rest consists of terminations by mutual agreement,¹⁸ those at the end of probationary periods, and retirements.)

The French exceptionalism in terms of the figures does not mean that other countries lack labor market rigidities whose effects can be counterproductive. For example, Spain has a high rate of unemployment among young people for similar reasons. Countries that have more flexible labor market institutions may still encounter difficulties in delivering good jobs, especially to the young. In the UK and US, employers have increasingly been creating insecure work in the form of zero-hours contracts, while self-employment and agency work has also become more prevalent, prompting much debate about whether policies need to change. While the US experiences a low unemployment rate, one in six working-age men without a college degree is not part of the workforce.¹⁹

For employees, unemployment and precarious employment are the tip of the iceberg. The hidden part is multiform …

Conflictual relationships. Relations between employers and employees are not harmonious in France. France ranks 129th out of 139 countries in terms of people’s perception of relationships at work.20 We can only guess at the causes of this unfortunate French peculiarity, which contributes to employee burnout. Perhaps the absence of mobility just mentioned plays a role. An employee whose relationship with coworkers or superiors has grown tense will normally change jobs in a fluid labor market, but in France he or she does not have that option and will remain, despite the conflict. It is not impossible, either, that some less than scrupulous employers may deliberately worsen an employee’s work environment in order to give him or her an incentive to accept a mutually agreed-upon termination and to leave the company; this allows the employer to avoid a hearing before a labor tribunal.

The French report suffering from considerable stress at work. Studies conducted on the basis of international data²¹ demonstrate a positive correlation between legislation protecting jobs and stress at work. This correlation is hardly surprising. As we have seen, rigidity in jobs and the scarcity of jobs damage relationships at work in several ways: employees unsatisfied with their jobs hold on to them nonetheless, and unscrupulous bosses can easily play on the fear of unemployment to intimidate employees.

A strong sense of job insecurity. Job insecurity is obviously felt by workers with fixed-term contracts, whose jobs are by definition precarious. More surprisingly, workers with permanent contracts also feel insecure, even though they benefit from what are in practice the world’s most protective labor laws.22 This observation is not as paradoxical as it seems, insofar as a worker with a “permanent” contract knows that if he is fired or his firm goes bankrupt and he becomes unemployed, his chances of finding an equivalent job are limited. This leads to a feeling of pessimism that pervades the whole of French society and paralyzes it, handicapping its ability to adapt and innovate. While the French example is extreme, it illustrates the degree to which employment laws may have unintended consequences.

One argument often advanced against the necessity of reforming labor institutions in France is that unemployment reflects a weak demand for companies’ products, and that a macroeconomic reflation would reduce unemployment. There is no doubt that France, like all European countries, is suffering from an uncertain outlook and is subject to the aftereffects of the financial and European crises.²³ Favorable prospects and a well-filled order book would obviously have very positive effects on employment. But the macroeconomic argument is not relevant, for several reasons.

Work contracts and rules about redundancies and layoffs have to reconcile two objectives. Employees are not responsible for and have no control over technological change or demand shocks faced by their employers; they must therefore be insured against the risk that their jobs might become obsolete or simply unprofitable. For its part, the company will insist on flexibility in its management of human resources as it copes with these supply and demand shocks. Without this, it will be reluctant to create jobs in the first place, because it will incur large losses in the event that these jobs are not very productive. Two diametrically opposed points of view? Not necessarily. But to reconcile these two aims, the employee, not the job, has to be protected.

The employer knows whether a job is profitable for the firm—profitability properly conceived, of course, because an employer can cope with temporary losses on a workstation or a production unit due to a temporary decrease in demand, and yet still profit in the long run by retaining the job. The employer has the information needed to manage human resources. But we also have to ask what impact his choice between keeping the employee or letting him go might have on the company’s stakeholders.

A first question regarding this principle is how to calculate the cost of dismissals for the unemployment insurance fund. Firing a thirty-year-old software engineer in Paris who will find a job the next day costs nothing; but firing a fifty year old with few skills in a depressed job market is quite another matter. How then should we calculate the cost of the dismissal?

A clever way of calculating the penalty for dismissal consists of looking at how much the fired employee cost the unemployment insurance fund after he or she has been dismissed. This approach goes back to FDR’s United States, which set up a system on this principle that is still in operation, called “experience rating.”27 This has a twofold advantage: that of applying a higher penalty for dismissing employees who will have a harder time finding a job, and that of giving the company an incentive to invest in on-the-job training, thus enriching the human capital of its employees, and so limiting the length of time that they will be out of work in the event that they are dismissed. Similarly, it provides an incentive for management and labor in the same sector to improve the quality of ongoing vocational training, because they will be sensitized to the typical length of time people are unemployed.

We will see later that the reward-penalty system has other advantages, such as eliminating collusion between employers and employees at the expense of the social welfare system, and an improved allocation of activity among sectors of the economy.

The principle that the employer who dismisses an employee pays the associated social cost provides a general outline of how to achieve the desired accountability. It is, however, too simplistic; a few adjustments to the basic principle may be needed.²⁸

Progressive rights. Although fixed-term contracts are undesirable and should be abolished, enterprises do need short-term employment for temporary tasks or seasonal activities. A single labor contract system with progressive rights for the employee is compatible with this need for short-term jobs.

If it gets to litigation, dismissal procedures 30 result in major costs for the company far in excess of the payoffs to the employees who have been fired. The procedures are very lengthy (although they have been getting a little shorter lately). Suits relating to dismissals can be brought by the dismissed employee for up to two years.³¹ After the case is submitted to a court, a decision takes on average 13.6 months in the first instance and 35 months in the event of an appeal (which occurs in two-thirds of the cases). Not to mention that in a few cases there remains the risk of reinstatement—with payment of the interim salary.

The employer has to prove that there was a “real and serious cause” for the dismissal. In France, in contrast to many countries, the redundancy of a position, employee performance, or more generally the desire to cut costs are not considered legitimate reasons. A dismissal on economic grounds can only be justified if there are serious financial problems that put the company’s survival in danger, so a company that is otherwise healthy cannot stop an activity for which it no longer has a sustainable order book.³² Unsurprisingly, the cost to the company also includes the time its managers devote to dealing with the case (which takes them away from tasks involving the company’s future).

As for the employee, the process is costly, as well as unfair: those who are vulnerable and not very familiar with the complexity of the French system may be discouraged and may play the game less well than insiders³³ who know how the institutions operate. Furthermore, there is also a great uncertainty as to the final decision and important disparities between labor tribunals.³⁴

ANOTHER INSTANCE OF FRENCH EXCEPTIONALISM: COLLUSION BETWEEN MANAGEMENT AND LABOR AT THE EXPENSE OF SOCIETY

Relations between employers and employees are mediocre in France … except when it comes to colluding at the expense of the taxpayer. As always, economic agents react to the incentives they confront. The real guilty parties are the French labor market institutions encouraging management and employees to manipulate the system.

First of all, employers and employees have learned to systematically transform resignations into dismissals. Unlike a dismissal, a resignation does not give the employee the right to receive unemployment benefits. Companies and their employees thus have an interest in conspiring against the unemployment insurance system by presenting voluntary departures as dismissals. So long as the employee promises to give up his rights to sue the firm and leaves the company “on good terms,” this redesignation costs the company nothing, as it pays none of the unemployment benefits that the employee will receive after the “dismissal.”

In fact, employers and employees do not even have to game the system now, as this collusion was made legal in 2008 through the introduction of “termination by mutual consent,” a procedure that allows an employer and an employee with a permanent contract to agree on the conditions for breaking the work contract between them. This law made it easier to disguise a resignation as a dismissal. Its popularity (more than 358,000 terminations by mutual consent in 2015!) is thus not surprising.

This measure also has implications for the effective retirement age, possibly bringing it forward by as much as three years. When it was first implemented in 2009, French labor economists Pierre Cahuc and André Zylberberg noted:

The possibility of breaking a contract by mutual consent allows an employer and an employee to separate on good terms. The devil is in the details: insofar as the employee will end up keeping unemployment payments for three years, in reality the government has just made it possible to retire at fifty-seven! An older worker will in fact be able to leave a job almost without monetary loss by having the unemployment insurance system finance what amounts to an early retirement.35

Let us note that this collusion against the unemployment insurance system would not take place if the company were made accountable by a reward-penalty system. This would make it expensive for a company to have a former worker unemployed or in pseudoretirement, and so there would be less of this manipulation of the system.

Another way that management and labor evade the law is the transformation of a redundancy on economic grounds (sometimes of a whole group) into a dismissal for personal reasons.36 In this case, the termination is desired by the employer: Pierre Cahuc and Francis Kramarz observed in 2005 that

all the testimonies obtained from CEOs, labor union officials, and directors of human resources suggest that dismissals for personal reasons are frequently economic dismissals in disguise. For the employer, the alibi of the personal motive makes it possible to avoid the procedures required for dismissal on economic grounds, even when the dismissal of a group of employees is involved; the employer is thus encouraged to adduce a personal reason for firing a worker, even if it means making a transaction with the employee so that the latter gives up his right of recourse in exchange for an indemnity. The employee then accepts particularly advantageous severance conditions.³⁷

These behaviors underline the firm’s lack of accountability for the cost its actions impose on the unemployment insurance system. In 2013, there were thirty-eight thousand dismissals per month for personal reasons, as compared with sixteen thousand for economic reasons. About three-quarters of the dismissals for personal reasons were undisputed.

Above all, job creation is actually neglected here. If the weak protection given creditors has a negative impact on the financing and growth of companies, which is likely, the law’s net effect on employment is certainly negative.

Of course, even protected by grandfathered rights, these privileged workers might still feel concern. They might fear future promotions would give priority to employees under the new contract if they did not themselves convert to the latter; and, like their employers, they might worry about the state’s ability to make long-term commitments and fear, at some time in the future, the complete abolition of the old contracts, thus affecting those who had retained this status. That said, the increased likelihood of finding a new job under the new system, and better prospects for their children finding work, should overcome these concerns.

Another reason why one might want to grant grandfathered rights to employees with permanent jobs is that one could imagine an immediate wave of dismissals related to jobs that companies wanted to make redundant but have been unable to. More generally, the transition from the initial situation of the labor market, which necessarily takes time to improve following a reform, must be considered carefully. For instance, workers on the new contracts who are fired might still remain, on average, too long unemployed; it would probably be desirable to introduce the reward-penalty system gradually until unemployment has been reduced.

The poor functioning of the labor market in southern Europe is hard to ignore, and has indeed been evident for a long time (even though it has gotten worse due to the impact of the financial and European crises). It is the economist’s job to try to understand the resistance to change in light of this. When dysfunctional labor market institutions are still supported by a majority of citizens, it is not surprising that governments are in no hurry to explore controversial reforms and that they rather use the same old quick fixes to fight unemployment. It is easy for citizens to understand that their company could more easily terminate their permanent employment contracts if employers get more flexibility; it is much harder for these employees, as well as for the unemployed and employees in precarious jobs, to identify and analyze the economic mechanism that results in a more flexible system creating more numerous and better jobs.

The phenomenon of the identifiable victim, much studied by psychologists,43 refers to the observation that individuals feel much more empathy toward clearly identified victims (to the point of being prepared to help them) than toward victims defined more vaguely (“statistical victims”). For example, citizens are prepared to donate considerable sums to relieve the suffering of a person whose image and story they have seen on television, but are less disposed to help anonymous victims who have a greater need of their money. In the domain of the labor market, the identified victims are employees suffering from mass layoffs, and the anonymous victims are the unemployed for whom no job has been created.

Although it is emblematic, the character of the labor contract is not solely responsible for the current situation of high unemployment in southern Europe. For example, among other criticisms commonly directed at French institutions are the following:

• the gap between the skills provided by schools and the needs of companies (indeed, the high unemployment rate coincides with real shortages of labor for certain kinds of jobs);

• redistribution in favor of low-income workers primarily through the channel of the minimum wage (which is the highest in the European Union) rather than through the tax system—that is, through a tax credit on income from work (like the Earned Income Tax Credit in the US or the UK’s tax credit system);

• the management of the unemployed (usually called active labor market policy), which is very different from that of the Scandinavian countries, for example, where the unemployed are given greater incentives to resume work. Also, the replacement rate—the ratio of unemployment benefits to earnings from employment—is around the European average, but it is much more generous for workers with high salaries (three times higher than in Germany, for instance);

• the management of the public employment service (it is symptomatic that employers resort to Le Bon coin—a free Internet want-ads site—to recruit employees, despite a very extensive public employment service);

• the lack of flexibility in contractual arrangements, with rules being set mostly at the sector (rather than the firm) level;

• the closed nature of certain professions (for example, taxi driving), which prevents the creation of jobs many people would want to do.

Economists widely reject 46 the fallacious claim that there is a fixed quantity of employment, a concept according to which the total number of jobs in an economy is fixed and thus must be distributed equitably (the “lump of labor fallacy”). The introduction of the thirty-five-hour work week in France in 2000, which had received no significant support from professional economists and which sought to create jobs by spreading the work around, took people by surprise. And yet the idea that employment is scarce is age old and regularly resurfaces, particularly in periods of recession. Paradoxically (given that it tends to be favored on the left of the political spectrum), the idea that there is a fixed amount of work—and so working time should be reduced in order to allow the employment to be shared—is the same one that underlies the discourse of far-right parties when they claim that immigrants “take employment away” from resident nationals. Others use it to advocate lowering the retirement age (don’t older workers take work away from the young?). Still others use it in support of protectionist goals (don’t foreign companies take away our jobs?). Finally, others worried about the impact on available work in 1996, when it was decided to end compulsory military service in France.

Where does the idea that work is scarce, and that the government must intervene to distribute it, come from? Indirectly, from Malthus. At the beginning of the nineteenth century, the scarce resource was still largely land. Land was more or less limited in quantity, and thus jobs working the earth were limited as well. Of course, work has always required not only labor, but also complementary factors of production, for example machine tools, premises, computers, or factories. But unlike the production factor that Malthus focused on (land), these production factors are not at all fixed in quantity, at least in the medium and long term. Even in the short term, production factors can be adjusted in certain circumstances. In a famous article,47 David Card (now at the University of California, Berkeley) studied the impact of 125,000 Cuban migrants who came to Miami in the space of a few months in 1980. This migration, of significant magnitude in relation to the size of Miami (7 percent increase in the labor force), had practically no impact on either unemployment or on the salaries of the groups competing with the Cuban immigrants on the labor market (essentially African Americans). Work was not fixed in quantity; investments in textile production quickly created the necessary jobs.

In the case of sharing work time, the reasoning is slightly different. Suppose there is a legal obligation to reduce working time that favors employees and disadvantages companies—for instance, no change in salary (in practice, the reduction in work time is often accompanied by changes in salary agreements or state subsidies—we will return to this possibility). In the short term, it is possible that employment will increase in order to compensate for the reduction in working hours so as to meet the orders on hand and to make use of existing factors of production. However, this is a flash in the pan; in the medium term, the order book, other inputs, and employment adjust downward. Yet a sustainable employment policy has to create jobs over the medium and long term. Furthermore, the cost of public policies to compensate employers, if adopted, must be taken into account too, because taxes will have to be increased or public expenditures decreased elsewhere.

This fallacious idea that work has a fixed quantity also crops up in attitudes toward technological progress. Employment changes continually. For at least two centuries we have been afraid that automation—from weaving looms (the revolt of the Luddites in England in the early nineteenth century) to assembly lines in the 1950s, and more recently robotics—might cause employment to disappear. All these technological changes lead in fact to the disappearance of certain jobs, but fortunately not to the disappearance of employment (otherwise we would all be unemployed). Similarly, studies show that migrants bring economic benefits to a country, including for its workers in terms of more jobs and higher economic growth, with very little or no effect on earnings.⁵¹ Needless to say, this is just a broad-brush picture, as precise effects depend on the context (such as migrants’ skills, labor market and welfare institutions, and the complementarity or substitutability between immigrants and natives in the job market).⁵²

This brings me to the issue of protectionism. Around the world, populist politicians are riding the wave of popular indignation against job losses or falling wages for unskilled workers. This indignation reflects real difficulties: governments have not paid enough attention to the damage caused to some of their citizens by globalization. Until 1990, international trade created relatively few losers. Since then, developing countries—notably China—have turned their backs on their earlier protectionist policies of import substitution and have opted instead for a market-driven and open economy. Simultaneously, container transport costs have fallen drastically. These two phenomena have realigned trade to take place along the global north-south axis. Whilst developed countries are winners overall, many of their workers, often unskilled, are experiencing difficulties finding jobs close to home similar to the ones that have gone. And retraining policies to increase their skills and equip them for other jobs have not kept up.

THE COMPLEXITY OF THE FRENCH LABOR CODE

In another example of France’s exceptionalism when it comes to the labor market, French labor law is both complex and directive. Complex: everyone agrees. The French labor code is 3,200 pages long and continues to grow as the years go by. Even highly specialized professors of labor law do not fully master it. Hence, the adage “ignorance of the law is no excuse,” which normally makes perfect sense, becomes almost laughable in this context. Do we really expect the CEO of a small or middle-sized company, who has no specialized legal department and who is busy with other tasks, to be able to master this legal corpus? Even a large company with a solid legal department is capable of violating labor law without knowing it.

The directive aspect of French labor law is more controversial. France remains one of the countries where the state and industry bodies intervene the most in contractual relationships between employers and employees. Elsewhere, there is more room for negotiations between employers and employees (Great Britain) or between sector bodies and employees (the Scandinavian countries). Some countries (Denmark) have virtually no labor law, giving contractual arrangements free rein and thus more opportunities to adapt their employment relations to the specific context of a business or sector.

To be sure, the French legislature has provided some room for negotiation at the level of sectors and enterprises. In particular, since 2004, a company-level agreement can deviate from the sector agreement if the latter does not explicitly rule it out. However, in practice there are very few such deviations from the norm, the hierarchy of rules being: 1) the labor code, 2) the sector agreement, 3) the company agreement. Sectoral agreements are systematically extended by the Ministry of Labor in such a way that the possibility of a distinct company agreement becomes moot. Sector agreements may solve the problem of free-riding in the provision of ongoing vocational training, but they also allow member companies to come to agreements among themselves and to make the final consumer pay the increase in costs (especially in sectors not subject to international competition); in the latter respect, they are bad for demand and thus for employment.55 Because of the quasi-systematic extension of sector agreements, more than 90 percent of French employees are covered by collective conventions, compared to only 1 percent in Germany.⁵⁶

This obviously does not mean that sectors should play no role in labor contracts. Management and labor unions in a company do not always have the expertise necessary to draw up contracts, or a clear view of their consequences, especially in small and medium-sized enterprises. The sectors thus have an important role to play as providers of services to help management and labor organize their relations better, or by conceiving sectoral agreements that are an option rather than a constraint on negotiations within the enterprise.

Finally, let us note that even when sector or company agreements are extensive, the labor code can still play an important role as in the case of a minimum wage.57

2. Immigration. The 2015 European migrant crisis has renewed many people’s concerns about a shortage of jobs; no matter how considerable it seems to be, this crisis is only a prelude to what will happen if we do not succeed in controlling global warming.⁵⁸ However, migrants represent an economic as well as cultural opportunity for any country. They must therefore be welcomed as contributors to society, which requires that they not be excluded from jobs by labor market institutions—yet another reason to reform these institutions.

3. Technology. The digital revolution⁵⁹ will have two effects that exacerbate the social cost of rigidities. First of all, it increases the speed at which employment is being transformed and jobs destroyed and created, making the overly rigid permanent contracts even less attractive for employers than they are today, and increasing the need for better ongoing vocational retraining. Second, work is itself in the process of changing, with a growing number of self-employed people and individuals working for several employers. Many observers are calling for the creation of a worker’s law rather than a more restrictive employee’s law. French labor law, as massive as it is, focuses almost exclusively on the employee, and is based on a conception of work that dates from employment in factories. Consequently, France has a long way to go to prepare itself for the changes ahead. And while France stands out for the rigidity of its labor market institutions, the same point applies to every other developed economy. Even in the flexible US, UK, or Danish labor markets, the legal and regulatory frameworks center around employees rather than workers.