The paradox of disappearing European unemployment

Tito Boeri 23 April 2008

a

A

An old European dream has come true, but it looks more and more like a nightmare.

The dream was written on the stone of the Treaty, signed in Rome on March 25, 1957: “The Community shall have as its task (…) to promote throughout the Community (…) a high degree of convergence of economic performance, a high level of employment and of social protection, the raising of the standard of living and quality of life, and economic and social cohesion and solidarity among Member States.”

A European dream came true…

In the last ten years European unemployment has fallen to a level not seen for over twenty-five years. There are currently almost 4 million fewer people unemployed in the EU15 than in 1996. Long-term unemployment almost halved: Europe is no longer a place where half of all job seekers have been on the dole for more than twelve months, as was the case in the mid-1990s.

The disappearance of mass unemployment in Europe is not the by-product of falling participation to the labour market; it is the average employment rate in the EU15 that has increased by more than six percent over the last ten years. This is the only area in which Europe is getting closer to the ambitious Lisbon targets. Nor there are more discouraged workers crossing the pourous borders between participation and non-participation to the labour market: there was no increase in the number of non-employed persons who decided to stop searching for another job as they realized that there were no vacancies for them.

It was mainly the countries with initially the largest unemployment rates that succeeded in reducing unemployment the most. The cross-sectional dispersion in unemployment rates across regions of the EU15 (Nuts II) also declined considerably as a result of both less cross country and less within country variation in unemployment rates. This looks like a major step towards achieving the social cohesion pursued by European Governments at least since the 1957 Rome Treaty. European regions are less and less different in terms of labour market conditions.

.... but it is turning into a nightmare

European Governments, however, are not capitalizing on these labour market success stories. Governments and coalitions ruling while millions of jobs were created have not been reconfirmed in office. Berlusconi’s 2001-6 Government succeeded in creating 1.3 million jobs in five years, far more than promised in the 2001 electoral campaign. This did not prevent the collapse of his popularity and the defeat at the next elections. Prodi’s 2006-8 Government had a very short life, in spite of creating more than 400,000 jobs in less than 2 years. Aznar lost in 2004 after halving Spanish unemployment and creating almost 5 million jobs during his mandate.

Public opinion polls also find rising dissatisfaction with working conditions, notably in those countries having experienced the strongest unemployment declines.

Why?

Why is a European dream turning into a European nightmare? The simplest explanation one could possibly offer is that the decline in unemployment was a demographic phenomenon, independent of changes in the incidence of unemployment among specific socio-economic groups. Europe is experiencing the ageing of its population and young people typically display higher unemployment rates than older workers. Thus, an ageing Europe is bound to have lower unemployment simply as a result of a changing age distribution of its workforce. However, this simple explanation does not work. It could at most explain one-tenth of the decline in unemployment. The remaining success is due to a fall of unemployment among all age groups. Nor does large-scale immigration, the second most important demographic phenomenon in Europe in the last decade, explain the disappearance of European mass unemployment. If anything, more migration should have involved increasing unemployment rates: the incidence of unemployment is typically higher among migrants than natives in the EU15.

In order to understand what has been going on in Europe and the paradoxical dissatisfaction of its citizens with the disappearance of unemployment, one has to go beyond labour market stocks and look at flows across labour market states. The first thing to notice is that unemployment has been declining in spite of an increase of unemployment inflows rates (inflows over the population at risk, that is working age population minus the unemployed). Put it another way, it was mainly an increase in outflows from unemployment that drove the fall in European unemployment. Secondly, one discovers that there has been an increase in mobility across labour market states, evident from computing mobility indexes over transition matrices mapping flows across the main labour market states. Significantly, the increase in mobility was stronger in the countries experiencing the largest falls in unemployment.

It was reforms!

The European labour market landscape looks much different from the sclerotic conditions of the early 1990s. Let us remind what a very influential report commissioned by the G7, the 1994 OECD Jobs Study stated at that time: “In inflexible Europe … the high incidence of long-term unemployment is associated with low inflow rates into unemployment.

Why did this sea change from sclerotic to a more mobile Europe occur? The driving factor of the increase in labour market flows would seem to have been reforms of employment protection legislation. During the 1990s, major reforms reduced costs of dismissals: there were only four such reforms over the entire EU15 in the 1986-90 period and 16 in the 1996-2000 period. Most of these reforms were marginal in that they were confined to reducing employment protection for new hires, expanding the scope of fixed-term contracts and introducing new and more flexible contractual types (from temporary work agency to job-on-call). This dramatically changed the conditions at entry. In the countries with the strictest provisions concerning the dismissals of permanent workers, the majority of new hires are currently in these new and highly flexible contracts. For example, in Spain nine out of ten transitions from unemployment to employment occur in fixed-term contracts. The increase in unemployment outflows in Europe was largely associated with this new entry channel.
The trouble is that, rather than being just a port of entry, these contracts often become a dead-end: the probability of moving from a fixed-term to permanent contract within a year is indeed very low, of the order of one out of 20 or one out of 10. In other words, dual track reforms created long-lasting asymmetries in career paths, concentrating risk on the workers with flexible contracts. On the basis of transition matrices, it is predicted that, in the long-run, up to one third of employment will be in such flexible contracts.

Can Europeans be happier about their new labour market?

The dissatisfaction of Europeans with respect to their labour market is ultimately related to a new and apparently less favourable risk-return combination. Labour markets are becoming more risky and this entails a welfare loss for risk-averse workers unless this greater risk is compensated by higher returns. Pressures are mounting everywhere in Europe for greater state involvement in wage setting. These pressures can also be interpreted as a request of compensation for greater labour market risk. Nobody can be fully insulated from it. Even the insiders are somewhat exposed as they form expectations about job loss.

The pressures to go back are strong. But after having reduced state involvement in employment adjustment, it would be a mistake to have governments more involved in wage setting. Setting statutory and industry-specific wage minima, as recently done in Germany, exposes governments to even stronger pressures from national lobbies and leapfrogging games across industries. And there is no reason to reintroduce even the mild forms of income policy that were adopted in several EU countries in preparation for EMU membership. The problem is that centralised wage setting is not an appropriate instrument under the EMU, as macroeconomic shocks are more regional or industry-specific in nature. Thus, national-union-based systems of industrial relations are ill-suited to address new demands for microeconomic adaptability.

The best response that can be offered to the paradoxical concerns of Europeans with respect to lower unemployment is to decentralise even more wage setting and make it more responsive to productivity. Risk is magnified by the fact that any labour market transition involves a large wage loss. Centralized wage agreements tend to reward automatic adjustments of wages to tenure. Moving across jobs or experiencing even a short unemployment spell prevents moving up the wage ladder. A better risk-return combination can be offered by linking wages more closely to idiosyncratic productivity. Insofar as changing jobs involves pairing better matches, wages would increase after changing jobs rather than the other way round.

At the same time, something must be done to tackle the increasing dualism between temporary and permanent jobs in most European labour markets. This dualism is costly for the society at large as it reduces incentives to accumulate human capital: workers with fixed-term contracts are less subject to on-the-job training than the other workers. A sensible policy would be to offer a clear “tenure track” prospect to young workers by completing reforms of employment protection. Currently, there is no long-term prospect after the expiration of a temporary contract. Governments could promote permanent entrance in the permanent labour market in stages, introducing employment protection with gradualism and avoiding the formation of a long-term dual market. Job security provisions, in the form of mandated severance payments, should increase smoothly as workers acquire tenure without large discontinuities.

Waking up from the nightmare

There is ultimately a trade-off between employment and productivity growth behind the dissatisfaction of Europeans with low unemployment. Employment growth is occurring at the cost of negative or low growth of labour productivity. This prevents workers from being compensated for their higher risk exposure. The fact of the matter is that Europe is still in the middle of the river of labour market reforms. Pressures to go back are strong. Governments should resist these pressures, as they would have huge employment costs. Increasing both employment and productivity in Europe requires doing just the opposite. Governments should take us on the other side of the river: a tenure track to more stable jobs should be introduced and wage setting should be decentralized to link it more closely to productivity gains.

a

A

Topics:  Labour markets

Tags:  elections, European unemployment, employment risk, electoral campaigns, dissatisfaction

President, Italian Social Security administration (Inps)

Events

CEPR Policy Research