Does corruption influence young brain drain? evidence from Italy

In recent years, young Italian brain drain within provinces has increased at higher speed than ever. While is premature to assess whether this process is transitory or permanent, it is undoubted that it is relevant and needs to be analysed by researchers and monitored by policy makers constantly. Previous empirical studies have demonstrated that net skilled migration is influenced by economic factors, such as the search for higher income per capita and job opportunities, and, with a less extent, by the search of places endowed with more amenities. In the crossroad between these factors, this dissertation investigates the role of corruption, as proxy for meritocracy, as a key element of influence over young skilled mobility. To this end, a comprehensive framework of analysis, based on the comparison of results get by traditional and novel empirical methodologies, is used. Hence, the present dissertation develops its study in three chapters. Chapter 1 offers a detailed review of the literature that has analysed the general causes of skilled mobility and discusses the novel elements introduced by the current study. Then, Chapter 2 investigates the relationship between corruption and the Italian skilled mobility by exploiting traditional data and dynamic panel model. Evidence suggests that high corruption positively affects skilled flows from origin province, ceteris paribus. Although the adopted model is robust because it is widely exploited among researchers and controls for endogeneity, its main limit is represented by the fact that it does not fully exploit the potentialities offered by bilateral data of a tri-panel dataset, losing great part of information. Thus, Chapter 3 deals with the trade-off between robustness and completeness handled by the traditional method presented in Chapter 2 and adopts a gravity framework with a novel Pseudo Poisson. Results suggest the existence of push and pull mechanisms of corruption at play on young skilled mobility. Besides, evidence proves that sensitivity of the prospective tertiary students towards corruption varies according to their field of study of interest. Also, corruption widely affects long-distance skilled flow from the Centre-South to the North of Italy. Chapter 3 enriches its analysis with bilateral data on enrolments at university, additive research questions and results by adopting a remedy that does not give up completeness for gaining robustness in the empirical analysis. Results of the novel model of Chapter 3 present similarity with the results of the traditional model of Chapter 2, demonstrating that the coexistence of robustness and completeness features is possible if models are correctly implemented