Labour Markets
Rigidities and Unemployment

There seems to be a consensus among economists that labour-market rigidities are responsible for high unemployment in Europe. In particular, they are blamed for unemployment’s most alarming aspects: its long duration and its high incidence among youth. Recent work by – among others – two CEPR Research Fellows, Gilles Saint-Paul and Alan Manning, however, has shed new light on some aspects of both the sources and the consequences of labour-market rigidities. In particular, Saint-Paul’s research suggests that many of the rigidities – and much of the consequently high European unemployment levels – are a direct result of the powerful political influences exerted by people who already have jobs; and Manning’s analyses cast doubt on the widely held view that minimum wages have been a significant factor in the poor performance of European labour markets in recent years relative both to the ‘golden age’ of the 1960s and to the United States.

Three sources of rigidity are singled out in the conventional wisdom: the welfare state, minimum wage rates and employment protection legislation. Thus, unemployment benefits lower the incentive for job search and increase wage pressure from those in work; minimum wages price the least skilled out of the market; and firing costs deter hiring, reducing labour demand and hampering the economy’s ability to deal with uncertainty and structural change. Hence the frequent recommendations – exemplified by the 1995 OECD Job Study – for more flexible labour markets.

In practice, however, few of the remedies economists advocate pass the test of political viability. Examples abound: in 1994, an attempt by the French government to lower the minimum wage for young workers was followed by violent demonstrations, eventually leading to withdrawal of the reform proposal; the Swedish government lost the 1994 election because it had lowered the unemployment benefit replacement ratio from 90% to 80%; and after German reunification, despite large productivity differentials and the need to restructure the eastern economy, the government gave in to pressure from western unions and allowed eastern wages to converge rapidly to western levels, thus leading to substantially higher unemployment rates in the east than in the west.

Such phenomena raise a number of questions, according to Saint-Paul and Manning. What is the coalition of economic agents that benefits from existing labour-market institutions and policies? What strategies to reform the labour market are politically viable? What economic environments are most favourable to economic reform? And how valid is the belief that minimum wage rates cause unemployment?

Both researchers have addressed some of these questions in a series of recent papers. Using both theoretical and empirical perspectives, Saint-Paul has investigated the mechanisms that account for the persistence of labour rigidities and how they affect the political viability of reform. One important mechanism is described by the political insider-outsider model. This refers to the fact that, when voting or lobbying in favour of a given institution, incumbent employees will take into account the institution’s impact on their ability to bid up wages at the firm level. Thus, labour-market rigidities can be thought of as devices that allow incumbent employees indirectly to achieve monopoly power in wage-setting, even though at the firm level the scope for wage increases beyond the level set by competitors is very limited.

Another mechanism reflects policy complementarities: many channels can be identified through which labour-market rigidities mutually reinforce each other because the existence of one institution increases support for another. For example, by reducing the exposure of employed workers to unemployment, job-protection legislation increases the political support for institutions that reduce job creation, such as generous welfare benefits funded by payroll taxes. A third mechanism is social cohesion: some institutions, such as minimum wages and wage ladders, typically reduce the dispersion of wages compared to the market equilibrium. Because they create unemployment, they are not an efficient tool to redistribute income to the poorest and may even increase overall inequality.

Saint-Paul argues that such institutions benefit the middle class by reducing the degree of redistributive conflict between the middle class and lower class. The key mechanism is that those members of the lower class who have jobs are richer and thus less likely to support redistributive policies. At the same time, the unemployed are poorer but have little impact on political outcomes because they remain a minority. In examining the logic of unemployment benefits, Saint-Paul notes that the most straightforward way to think about them is as an insurance device. But the political insider-outsider model suggests that they may also benefit incumbent employees by increasing their bargaining power in wage-setting negotiations.

Saint-Paul has assembled a range of empirical evidence on these mechanisms. By looking at the determinants of unemployment benefits in a panel of countries, he has sought to determine whether insurance or wage-setting motives have a greater impact on the setting of these benefits. He finds little evidence of insurance motives: there is no correlation between the level of unemployment benefits and employed people’s exposure to unemployment. At the same time, economies where employment is less reactive to wages – implying a greater scope for wage increases – seem to have more generous unemployment benefits. This suggests that the wage-setting motive is playing a role.

Saint-Paul’s work also examines the reform of employment protection legislation. In the 1980s, many European governments attempted to boost job creation by reducing firing costs. Saint-Paul argues that both the design of these reforms, and the economic environment in which they were implemented, were important determinants of their success. First, many were designed so as to protect the interests of incumbent employees. This was done by using a two-tier system, in which the new regulations would apply only to new contracts, and by limiting the use and renewal of the new contracts. Such limits reduced the likelihood of ‘flexible employees’ eventually outnumbering ‘rigid employees’, and hence threatening their political power. Second, the two-tier reforms were implemented at times of relatively high job destruction and rising unemployment, so that the exposure of employed people to unemployment increased their support for measures that create jobs.

On the belief that minimum wages are significantly responsible for Europe’s unemployment problem, Manning notes that it would be difficult to point to much in the way of persuasive research on the issue. It was with this in mind that a group of European economists began to look in more detail at the impact of minimum wages.

An obvious first point is that the impact of a minimum wage rate will depend on the level at which it is set: not even the most ardent supporter of minimum wages would suggest that the minimum rate can be raised continually without eventually running into serious problems. Some perspective is therefore needed on this issue. It is often claimed that minimum wages are much higher as a fraction of average earnings in Europe than in the United States. This is generally, but not universally, confirmed by the data. Thus, minimum wages in the United States in the mid-1990s were approximately 37% of average earnings, but 50–60% in many European countries.

Manning argues, however, that these economy-wide statistics are potentially misleading. In the youth labour market – where the minimum wage has a larger impact – US minimum wages are a higher fraction of average earnings (85% in 1992) than in most European countries. The reason is that in the United States there is effectively no variation in the minimum wage by age, whereas virtually all European countries have a lower minimum wage for younger workers. In the parts of the labour market where it matters, therefore, it is not clear that the European minimum wage is a more serious constraint.

The data also reveal that minimum wages, as a proportion of average earnings, generally are no higher now than in the 1960s. If minimum wages are causing problems for European labour markets, therefore, it is not because they have been aggressively increased. Some would argue that this misses the point, in that the failure of European countries to reduce minimum wages has made their unskilled workers more and more uncompetitive in an increasingly cut-throat global marketplace. But most minimum-wage workers are not found in traded-goods sectors: they are heavily concentrated in retail and personal services, hotels and catering. The only manufacturing sectors with a high concentration of minimum-wage workers are clothing, leather and furniture. The few remaining European firms in these sectors have had to cope with wages that, for reasons quite unconnected with the minimum wage, commonly are ten times higher than those in competitor developing countries. Any problems posed by the minimum wage pale into insignificance when compared with this gap in wage costs.

Furthermore, if it is true that minimum wages are increasingly preventing low-skilled workers from pricing themselves into jobs, we would expect to see increasing numbers of workers paid at, or near to, the minimum rates. In France, the proportion of workers affected by the annual increase in the SMIC (salaire minimum interprofessionel de croissance) did increase from 5% in the 1970s to around 8% in the late 1980s. But the corresponding proportion was also around 8% in 1960 – and even higher in the 1950s – while the unemployment rate has risen from under 2% to more than 12%.

Against this general background, Manning notes that more detailed case-studies of the impact of changes in minimum wages revealed a complex picture. There were episodes where higher minimum wages seemed to be associated with lower employment: the rise in the youth minimum wage in Spain in 1990 seemingly had large effects on youth employment; and there was similar, though weaker, evidence from the cut in youth minimum wages in the Netherlands in the early 1980s. But other evidence suggests this is by no means always the case. In Spain, while youth employment may have fallen as the minimum was raised, total employment seemed, if anything, to rise. In the UK, as well, there is no evidence that total employment falls when the minimum wage is raised.

Perhaps the most interesting case is France, often said to have a minimum wage so high that it causes serious economic harm. Yet Manning and his colleagues find little evidence in support of this claim. Because the SMIC is a national minimum wage, its impact is much greater in low-wage than in high-wage regions. Yet, when the SMIC was raised by a large amount between the late-1960s and the mid-1980s, both wage and employment growth in the low-wage regions were stronger, relative to the high-wage regions, than in the period since the mid-1980s during which the SMIC has been slightly reduced.

All of this suggests that, at current levels, the impact of the minimum wage on European labour markets is often exaggerated. The point is further underlined by the fact that rising unemployment has occurred in all labour-market segments, including those in which highly educated workers are paid well above the minimum. This does not render discussion of the minimum wage irrelevant: while most people have no need of its protection and are unaffected by it, it is of vital importance for a minority who would otherwise be seriously exploited.

Both Manning and Saint-Paul believe that effective political discussions of labour market issues would be assisted by wider recognition of the sources and effects of the rigidities that are so widely, yet misleadingly, blamed for the problems of unemployment.

The research on which this summary is based is reported in greater detail in Economic Policy no 23, published in October 1996. Both Gilles Saint-Paul and Alan Manning presented this research to an audience of policy-makers and CEPR corporate members on 21 October 1996 in London.