Lower and upper bounds of unfair inequality: Theory and evidence for Germany and the US

Lower and Upper Bounds of Unfair Inequality: Theory and Evidence for Germany and the US Previous estimates of unfair inequality of opportunity (IOp) are only lower bounds because of the unobservability of the full set of endowed circumstances beyond the sphere of individual responsibility. In this paper, we suggest a new estimator based on a fixed effects panel model which additionally allows identifying an upper bound. We illustrate our approach by comparing Germany and the US based on harmonized micro data. We find significant and robust differences between lower and upper bound estimates – both for gross and net earnings based either on periodical or permanent income – for both countries. We discuss the cross-country differences and similarities in IOp in the light of di¤erences in social mobility and persistence. JEL Classification: D31, D63, H24, J62 Keywords: equality of opportunity, fairness, redistribution, wage inequality Corresponding author: Andreas Peichl IZA P.O. Box 7240 53072 Bonn Germany E-mail: [email protected] * We are grateful for financial support by Deutsche Forschungsgemeinschaft DFG (GRK1461 and PE1675). We would like to thank Rolf Aaberge, Ingvild Almas, Paolo Brunori, Koen Decanq, Marc Fleurbaey, Dan Hamermesh, David Jaeger, Peter Kuhn, John Roemer and Alain Trannoy as well as seminar and conference participants in Bonn, Cologne, Marseille, Milan and Rome for helpful comments and suggestions. The usual disclaimer applies. 1 Introduction Preferences for redistribution are systematically correlated with beliefs about the relative importance of e¤ort and luck in the determination of outcomes (see Konow (2003), Alesina and Giuliano (2011) and Gaertner and Schokkaert (2011) for overviews). Individuals are more willing to accept income di¤erences which are due to individual e¤ort (or laziness) rather than exogenous circumstances (Fong (2001)). Theories of distributive justice distinguish ethically acceptable inequalities (e.g. due to di¤erences in e¤ort) from unfair inequalities (e.g. due to endowed characteristics).1 In empirical applications, the main problem is the identi…cation of the latter, i.e. the amount of inequality which is due to circumstances beyond the sphere of individual responsibility. It has been recognized that previous estimates of such inequality of opportunity (IOp henceforth) yield only lower bounds because of the unobservability of the full set of circumstances (e.g. Bourguignon et al. (2007) and Ferreira and Gignoux (2011)). In this paper, we suggest a new estimator of IOp based on a …xed e¤ects model which additionally allows identifying an upper bound for unfair inequalities in order to provide the full range of IOp estimates. We illustrate our approach by comparing estimates for Germany and the US –two countries with di¤erent welfare state regimes, attitudes towards inequality and redistribution (see Figure 5 in the Appendix) and social mobility.2 The concept of equality of opportunity (EOp) in contrast to equality of outcomes (EO) has received considerable attention since the seminal contributions of Roemer (1993, 1998), Van de gaer (1993) and Fleurbaey (1995).3 The traditional notion of EO refers to an equal distribution of economic outcomes (e.g. well-being, consumption or income) across the population.4 The EOp theory, in contrast, is interested in the sources of inequality and separates the in‡uences on the outcomes of an individual into circumstances and e¤ort. Circumstances are de…ned as all factors beyond the sphere of individual control, for which society deems individuals should 1See Sen (1980, 1985, 1992), Dworkin (1981a,b), Arneson (1989), Cohen (1989), Roemer (1993, 1998, 2002) and Fleurbaey (2008). 2According to Alesina and Glaeser (2004), Americans believe that social mobility is important and high in the US, whereas Europeans perceive lower chances to climb the social ladder. Hence, Germans are more in favor of redistribution than Americans (Alesina and Angeletos (2005)). 3See e.g. Roemer et al. (2003), Dardanoni et al. (2005), Betts and Roemer (2006), Lefranc et al. (2008, 2009), Devooght (2008), Checchi et al. (2010), Checchi and Peragine (2011), Dunnzla¤ et al. (2011), Aaberge et al. (2011), Almås et al. (2011) as well as Björklund et al. (2011). 4See, e.g., Katz and Autor (1999) for an overview as well as Autor et al. (2008) and Dustmann et al. (2009) for recent applications to the US and Germany.