In policy debates, dynamic societies are often seen as desirable since they offer a fair chance of moving up the income ladder (Chetty et al. 2014, Alesina et al. 2017). To assess the extent to which societies provide individual members such opportunities to escape their origins, a common approach has been to measure the degree of income mobility in society – with income mobility itself often calculated as the lack of correlation between past (father’s) income and present (son’s) income.1 This approach has an important drawback – the opportunity for catch-up is measured using an indicator, income mobility, that depends on drivers that may be good or bad for society. The following simple example illustrates the point.
Consider a society in which a father’s lifetime income uniquely determines the income of his son at the time of the son’s labour market entry (the son’s ‘initial income’). Suppose further that after labour market entry, the incomes of individual workers grow at the same, constant rate. In this society, inter-generational income mobility is zero – the income of sons is perfectly correlated with their fathers’ incomes. Consider now the same society with one change – the labour income of workers is subject to random shocks that are unpredictable and long lasting. In the second society, income mobility is positive since sons’ incomes are less than fully correlated with that of their fathers, but members of the society are subject to risk and are therefore worse-off in welfare terms. Thus, in the presence of income risk, moving to a society with higher measured income mobility can reduce the well-being of all members of society.
In a recent paper, we propose a parsimonious framework that overcomes the shortcomings of the standard approach to income mobility described above (Krebs et al. 2017). Specifically, we present an empirical framework of labour income and a tractable theoretical model of consumption-saving with incomplete insurance markets that allow for a more nuanced analysis of income mobility and economic welfare. Our approach proceeds in two steps.
First, following a large labour literature,2 we assume that measured labour income of individual workers is determined by a number of observable worker characteristics (education, age, etc.) and a random component. The random component is comprised of persistent shocks representing risk, transitory shocks representing measurement error (or short-term fluctuations in income), and an opportunity term that measures the extent to which low-income individuals catch up with high-income individuals (i.e. move up the income ladder). Further, we show how the parameters of this income process can be quantitatively estimated using repeated cross-sectional data with a short panel dimension. Thus, our approach is applicable to most developed countries for which panel data on income are available, and many developing countries for which income data with only a short panel dimension are available.
Second, we provide a simple theoretical mapping from the estimated income process to income mobility, equilibrium consumption, and welfare (expected lifetime utility). We show how the two parameters representing opportunity and risk affect income mobility and expected lifetime utility (welfare) of risk-averse workers. The opportunity parameter increases income mobility and welfare – it measures ‘good’ mobility. The risk parameter increases income mobility, but reduces welfare – it measures ‘bad’ mobility. Thus, our analysis distinguishes between good and bad drivers of income mobility and offers a methodological framework to evaluate these drivers quantitatively.
Application to Mexico
In our paper, we present a quantitative implementation of our framework that underscores the importance of separating the main drivers of measured income mobility that we discussed above. Our analysis of income mobility in Mexico using longitudinal data on individual incomes yields striking results. We find that opportunity and risk are equally important drivers of income mobility.3 Further, both have large (but opposing) effects on welfare. For example, eliminating or insuring income risk would generate welfare gains that are equivalent to an increase in lifetime consumption by about 10%, even if workers are only moderately risk-averse (log-utility).4 Eliminating the catch-up of low-income individuals with high-income individuals yields a loss in social welfare of similar magnitude. Decomposing mobility into its fundamental components is thus seen to be crucial from the standpoint of welfare evaluation.
The application in Krebs et al. (2017) is concerned with the study of intra-generational mobility in one country (Mexico), but our approach can be applied to any country as long as income data with some panel dimension are available. Further, our finding that opportunity and risk are equally important drivers of intra-generational income mobility has important implications for the welfare evaluation of inter-generational income mobility. The application of our approach to intra-generational and inter-generational mobility in various countries is the subject of ongoing research.
Our research has two important implications. First, policy conclusions drawn from previous empirical results on income mobility measures stand on shaky ground since opportunity and risk are equally important drivers of measured income mobility, but have very different effects on welfare. In other words, differences in estimated income mobility across countries or observed changes in income mobility over time might just be due to differences or changes in welfare-reducing income risk. Second, future research on income mobility can (and should) provide policymakers with reliable results by separating the opportunity-component of income mobility from its risk-component.
Alesina, A, S Stantcheva and E Teso (2017), “Intergenerational mobility and preferences for redistribution”, VoxEU, 7 June.
Chetty, R, N Hendren, P Kline and E Saez (2014), “Where is the land of opportunity? Intergenerational mobility in the US”, VoxEU, 4 February.
Corak, M (2013), “Income inequality, equality of opportunity, and intergenerational mobility”, Journal of Economic Perspectives 27: 79-102.
Fields, G and E Ok (1999), “The measurement of income mobility: An introduction to the literature", in J Silber (ed), Handbook of Inequality Measurement, Dordrecht, Germany: Kluwer Academic Publishers.
Krebs, T, P Krishna and W F Maloney (2017), "Income Mobility, Income Risk and Welfare", NBER Working Paper No. 23578.
Meghir, C and L Pistaferri (2011), “Earnings, consumption, and life-cycle choices”, Chapter 9 in Handbook of Labor Economics, North Holland.
 See Corak (2013) and Fields and Ok (1999) for surveys of the literature.
 See Meghir and Pistaferri (2011) for a survey the literature.
 We also find that the largest part of measured income mobility is driven by measurement error or transitory income shocks and therefore (almost) welfare-neutral, and that the impact of measurement error on measured income mobility declines with the panel length.
 In comparison, for the same preference parameters, Lucas (2003) computes the welfare cost of aggregate consumption fluctuations for the US that are two orders of magnitude smaller. Thus, even though our estimates of persistent income risk seem small when measured mobility is the yardstick, their welfare effects are large indeed.