Abstract:
This paper provides a new approach to the empirical estimation of inequality of opportunity (IOP). Previous studies usually conclude that, in most countries, unfair inequality, arising from circumstances outside the realm of individual control, represent a small share of total income inequality. This result is largely driven by observational constraints that hamper the full observability of relevant circumstances and lead to lower-bound estimates. This also stands at odds with estimates of intergenerational and siblings correlation, which indicate a strong influence of differences in family background on individual success. This paper bridges the gap between these different approaches and relies on family and municipality fixed-effect models to account for shared unobservable circumstances alongside a rich set of individual characteristics. We apply this methodology to the estimation of IOP in Sweden. Our results point to a larger share of IOP than previously estimated and reaching up to 46% of observed income inequality.