The poor stay poor, the rich get rich: essays on intergenerational transmission of economic status

Persistence of inequality across generations is an important field of research with many implications in terms of policy. Economic inequality related to the intergenerational transmission of economic opportunities may strongly influence policies designed to reduce earnings or wealth concentration. Empirical research has usually focused on the intergenerational persistence of earnings or income, considered as good measures of differences in economic well-being and consumption capacity of individuals. On the contrary, only a limited number of recent studies attempt to estimate the degree of intergenerational mobility by using net wealth as a measure of economic status of individuals. This Ph.D thesis includes three autonomous chapters regarding the intergenerational persistence of wealth and earnings inequality and their mechanisms. The first one reviews research on wealth inequality and persistence across two or more generations. Broadly speaking, wealth is more unequally distributed than income and unlike other flow variables, may be transmitted across generations directly, by means of bequests or donations. This means that it may be a good proxy of permanent economic disparities. Unfortunately, measuring wealth is not an easy task since data on real and financial assets are incomplete and provided with many differences across countries. Regarding the extent of correlation in wealth across generations, only a limited number of studies are able to use suitable data on wealth which cover two or more generations. In any case, according to few recent empirical works, intergenerational rank correlations in wealth seem to be usually higher than intergenerational rank correlations in income. These findings derive from the fact that wealth is more representative of cumulate resources and less affected by transitory shocks than earnings or income. Moreover, wealthy parents seem to transmit many resources to their children at the beginning of the adulthood, by making donations. This may explain why, unlike intergenerational correlations in income, intergenerational correlations in wealth seem to be very high also considering children in their 20’s. The second chapter exploits retrospective socio-economic information about both parents to impute parental wealth in order to assess the degree of wealth mobility across generations in Italy and highlight some of the mechanisms linking parental wealth to offspring’s economic outcomes. Using the Bank of Italy’s survey on household income and wealth (SHIW) and two samples of offspring and pseudo-parents in their 40s, I find an intergenerational age-adjusted wealth elasticity (IWE) of 0.451 and a rank-rank slope of 0.349 which appear to be robust to the use of different predictors of parental economic status. These results suggest that Italy is a low mobility country also when wealth is taken as an alternative measure of economic status. As in the only previous study by Boserup et al. (2016) which analyses the pattern of wealth mobility over the lifecycle, the second chapter shows a U-shaped pattern of the intergenerational wealth correlation as a function of the second generation’s age with higher estimated intergenerational correlations when children are taken at the beginning of their adulthood or in their 40’s. Geographical differences in the extent of intergenerational wealth mobility are analysed by estimating elasticities and rank-rank slopes in two different macro-areas of the country. Results suggest that the southern part of Italy is extremely less mobile than the northern part of the country. Regarding the analysis of the mechanisms behind the intergenerational wealth correlation across two generations, the second chapter suggests that income seems to be the main intergenerational mediating factor. On the contrary, the correlation across generations of saving preferences and attitude to risk seems to explain only a small fraction of the IWE. Finally, the third chapter (which is part of a research work with Michele Raitano and Teresa Barbieri) provides new and detailed estimates of intergenerational earnings mobility in Italy and sheds light on mechanisms behind the association of gross earnings between fathers and sons. Being not available panel data following subsequent generations in Italy, we make use of a recently built dataset that merges information provided by IT-SILC 2005 (i.e., the Italian component of EU-SILC 2005) with detailed information about the whole working life of those interviewed in IT-SILC recorded in the administrative archives managed by the Italian national Social Security Institute (INPS). This dataset allows us to rely on the two-sample two-stage least squares method (TSTSLS) to predict father earnings and, then, compute point in time intergenerational elasticities (IGE) and imputed rank-rank slopes. Furthermore, the characteristics of the dataset allow us to extend point in time estimates considering, for both sons and “pseudo-fathers”, average earnings in a 5-year period and observing sons at various ages, thus assessing the robustness of our estimates to attenuation and life cycle biases. Confirming previous evidence (Mocetti 2007; Piraino 2007), we find that Italy is characterized by a relatively high earnings elasticity in cross country comparison – the size of the estimated β is usually over 0.40 – and the size of the intergenerational association increases when older sons and multi-annual averages are considered. We then investigate mechanisms behind this association both: i) including a set of possible mediating factors of the parental influence (e.g., sons’ education, occupation, labour market experience) among the control variables when regressing sons’ earnings on fathers’ earnings and ii) following the sequential decomposition approach suggested by Blanden, Gregg and Macmillan (2007). Results show that a limited share of the intergenerational association is attributable to sons’ educational and occupational attainment, while the largest part of the association is mediated by sons’ employability, i.e., by their effective experience since the entry in the labour market. Results show that the mediating role of education in Italy is limited, especially if compared with evidence obtained for other countries such as the US and UK.