In-work income stochastic frontiers: methodological advances for inequalities investigations

Graziella Bonanno, Filippo Domma, Camilla Mastromarco

Abstract

The existence of inequalities in income distribution is closely related to the standard conceptualization of efficiency. In this paper, the idea is to measure the differences between a potential in-work income, that should be obtained for an individual with particular socio-economic characteristics given his investment in human capital, and the actual received income. In detail, we estimate a similar Mincer equation for individual income, which incorporates human capital variables such as experience, education and occupation.In studies related to stochastic earnings frontier, some scholars use the Stochastic Frontier (SF) approach to get efficiency of logarithms of wages and refer to the traditional approach employing Normally distributed error components. The implicit hypothesis of this specification is that wages follow a log-Normal distribution. However, in the literature it has been shown that the latter distribution is not suitable due to the poor ability to describe both the upper and lower tails of the observed incomes distribution. To overcome this problem, the starting point of our specification is to use the Dagum function, for which it has been shown that it fits very well to the income distribution.

To test the new SF specification, the paper uses individual data from 9 longitudinal components of IT-SILC (Statistics on Income and Living Conditions for Italy). It collects timely and comparable cross-sectional and longitudinal data on income, poverty, social exclusion and living conditions and covers the period 2004-2015. This time span allows us to evaluate the impact of the economic crisis on the difficulty of workers in achieving their potential income in different countries and regions. Moreover, the effect of welfare politics will be investigated