How much inequality can be explained by family background? A new intergenerational decomposition approach

It is well-known that family background predicts child income. This is often taken to imply that family background is an important driver of cross-sectional income inequality. However, there is surprisingly little research on the strength and the nature of this link.

What share of income inequality is explained by family background? How sensitive is this share to the choice of inequality index? In other words, does the importance of family background depend on value judgments about what is considered "inequality"? How important are unobserved family variables relative to observed ones? Has inequality driven by family background increased or decreased over time?

We will answer these questions by developing new econometric methods and applying them to Swedish register data. Our approach combines the best of two existing approaches: (1) sibling correlations from the intergenerational mobility literature and (2) an inequality decomposition from the inequality of opportunity literature. Our novel econometric approach will use recent developments in the literature on short panels to measure the inequality effect of latent sibling fixed-effects. This allows us to account for unobserved family variables.