Youth Employment in Europe: Do Institutions and Social Capital Explain Better Than Mainstream Economics?

Article excerpt

Abstract

Why did employment growth - high in the last decade- take place at the expense of young workers mainly, but not only. in the countries of Southern Europe? Youth unemployment is now exceeding 30%, after decades hovering around 20% and over, despite a variety of factors, common to most EU countries, that would be expected to reduce its evolution: population ageing and the demographic decline, low labor cost of young workers, flexibility of working arrangements, higher educational attainment, low unionization of young workers, early retirement practices of workers 50+. But neither seems to provide a convincing explanation for countries of Southern Europe. Historically based institutions and political tradition, cultural values, social capital - factors that go beyond the standard explanation of economic theory - provide a more satisfying interpretation.

JEL classification: J0, J6, J23, F01, F16

Keywords: EU labor market institutions & LM performance

Introduction

Why did employment growth - high in the last decade- take place at the expense of young workers in the countries of Southern Europe? This is the question addressed in this paper. Youth unemployment has approached or exceeded 20% despite a variety of factors, common to most EU countries. According to neo-classical economics all such factors would be expected to exert a positive impact on the work opportunities of young people: population ageing and the demographic decline, low labor cost of young workers, flexibility of working arrangements, higher educational attainment, low unionization of young workers, early retirement practices of workers 50+. But neither seems to provide a convincing explanation for countries of Southern Europe. Historically based institutions and political tradition, cultural values, social capital - factors that go beyond the standard explanation of economic theory - provide a more satisfying interpretation. My paper develops this interpretation. In this sense, it may be viewed as a new contribution to a problem that economists have mainly addressed with case studies of specific countries, neglecting the influence of G. Esping-Andersen's categorization of three "worlds" of welfare capitalism since the early Nineties,1 and the general intuitions that it provided.

Youth unemployment in Italy, Spain, Portugal, Greece and France has hovered around and above 20% (well before the 2008 crisis), with employmentpopulation ratios below 40% against 50% and over in the countries of Northern Europe (and Austria). In principle, long term GNP growth should be an important driving factor, but the empirical evidence is weak: growth has been high also in member states of Southern Europe (Spain and Greece) where youth employment has been lagging, while it has been modest in some Scandinavian countries where it fared much better.

Workforce ageing has been a common feature all over the EU: had natural replacement of the retiring cohorts taken place evenly in the EU, the impact on youth employment should have been roughly the same everywhere. But it did not: in first place retirement age has a wide cross-country variability. Secondly, youth employment and old age employment appear to be complementary rather than substitutes in several countries, here again especially of Northern Europe.

A closely related argument is the demographic decline after the baby boom of the Sixties. The decline hit Southern Europe where youth employment suffered the most, more than the rest of the continent. This is yet another surprising fact, as there is no reason to expect that, in addition to the direct impact on the size of labor supply, the demographic decline should decrease youth participation. To the contrary, if aggregate demand, skills and productivity were constant, a larger proportion of the fewer young people remained, would be called at work.

My exploration rests on simple associations between macro indicators of labor market performance, demography and institutional characteristics: I display the results of linear bivariate regressions as purely descriptive representations which carry no causal interpretation. …