Wage distribution, educational wage premia and labour market institutions in Italy

The leading wire of the thesis is the Italian labour market, which is investigated along several dimensions: the wage distribution, the educational wage premia and the labour market institutions. Chapter 1 shows that educational wage premia in the private sector decreased over time and across the whole wage distribution, while they remained stable in the public sector. The evidence for the private sector is not in line with the international literature, especially the one of Anglo-Saxon countries, where educational premia increased markedly and favoured skill-biased changes. As for Chapter 2, it points out that inequality of net wages decreased in the private sector and increased in the public sector (1993-2006). Using a quantile decomposition methodology I show that the decrease in wage inequality in the private sector is driven by the negative coefficients component, which is consistent with the falling educational premia observed in chapter 1. I relate this evidence to changes in the job quality distribution that have penalized top occupations and to patterns of over education. As for the increase in inequality in the public sector I claim that it can be explained by the reforms of the wage setting system introduced in 1993 and 1998. To some extent, falling educational premia derived in the first two chapters could be also due to an inefficient matching process between characteristics of jobs and workers. To address this explanation, chapter 3 presents an evaluation of the Public Employment Services, showing that these services entails lower probability of finding a job in the short run, because of a lock-in effect, and higher probabilities in the long run. I also point out that PES effects are less pronounced when considering a proxy for the match quality as outcome variable. In chapter 4 I address the relation between wage distribution and productivity advantages stemming from spatial externalities, showing that much of the impact of spatial externalities is instead due to sorting of workers and firms. I also test the existence of a positive spatial correlation between individual skills and firm size, i.e. assortative matching. A further finding concerns the negative relation between the degree of assortative matching and local market size: the larger the size of the market the lesser the general human capital attributes should count.Doctorat en sciences économiques (ECON 3)--UCL, 200