Rivista Italiana di Economia Demografia e Statistica Volume LXXI n.3 Luglio-Settembre 2017 A RIF REGRESSION APPROACH TO EVALUATE WAGE CHANGES: A FOCUS ON ITALY Mariateresa Ciommi, Gennaro Punzo, Gaetano Musella, Francesco Maria Chelli, Rosalia Castellano 1. Introduction The recent structural changes in the European labour markets and in their income distribution are being encouraged by the ongoing economic crisis (see Acemoglu 1999; Autor 2003; Goos et al. 2009 among others). In Italy, the effects of the crisis have been made more serious because of the political instability and geographical disparities (Ballarino et al. 2014). Moreover, its impact on politics and society has been as relevant as its impact on the economy (Di Quirico, 2010). In this context, our paper aims at investigating the dynamics and the strength of changes in wage and wage inequality in Italy in the years of the Great Recession by analysing the role of individuals’ skills and of countries’ labour markets in rewarding employees. In line with the aim of identifying the driving forces of income changes over time and their intensity, we perform the Recentered Influence Function (hereafter, RIF) regression (Firpo et al., 2007; 2009; 2011) of Gini, variance, median and the two extreme deciles (q10 and q90) on log-wage. The RIF methodology is an extension of the Oaxaca-Blinder decomposition (Blinder, 1973; Oaxaca, 1973). However, unlike the latter can be applied only to the mean, the RIF decomposition is suitable to different distributional statistics. This allows us to explore the primary factors of wage levels and wage inequality and to decompose their changes over time into the composition and wage structure effects and, finally, to evaluate the contribution each factor gives to the overall changes. While the first component refers to the effect attributable to workers’ characteristics, the second captures the effect due to the capability of the country’s labour market to valorise individual skills and endowments. We use the Italian section of the EU-SILC data (European Union Statistics on Income and Living Conditions) with regard to two different years (2005 and 2013), which enables capturing the potential impact of the economic and financial crises on wage distribution and inequality. The paper is structured as follows. Section 2 offers a methodological overview and discusses some descriptive statistics of the crucial variables. Section 3 argues the results of the RIF regressions and decompositions. Section 4 concludes.