Innovation and inequality in Europe and Italy

In a phase of large increases in innovations and structural changes with lower economic growth and higher inequality and polarization, it is important to know and study the theories on the effects of innovations on inequality. In this paper we will examine the different theories on the link between these two variables, namely the skill-biased and the routine-biased technological change theories, the evolutionary theory and the geographical theories, with particular focuses on the roles of indirect effects and new technologies and on the empirical evidence. Innovations and income inequality have been increasing over time. Is there a link between the increasing trend of these variables? In this paper we will focus on these trends investigating whether patterns of innovations affect income inequality. To this aim we use two new databases (TechEvo and ARDECO) for data on, respectively, patents and other socioeconomic variables, from several European countries for the period 2003-2015 and carry out estimates on the associations of innovativeness with income inequality (quintile ratios). We find that innovations and population density, differently from other studies, are not correlated with income inequality in the European regions. Italy is less innovative and more unequal than the rest of Europe. At the same time, it has probably the worst performance in terms of growth and employment. In this paper we will study the link between innovations and inequality in Italy. We use four different panel models, two with the Arellano-Bover/Blundell-Bond GMM system estimator and two Dynamic Durbin Spatial Models and several different variables to measure different kinds of inequality (total income inequality from 2003 to 2015 at the regional level, inequality of taxpayers from 2000 to 2011 at the provincial level), finding that the relationship between innovations and inequality depends on the measure of inequality employed and on the geographical level analysed and that other factors, such as population density, are more important.