Intergenerational Wealth Formation over the Life Cycle

Wojciech Kopczuk (Columbia)

Abstract:

This paper provides novel insights on intergenerational wealth mobility and its relationship to lifetime economic resources using thirty years of wealth records for the Danish population. Non-parametric evidence reveals an almost linear relationship between wealth ranks of children and parents, except at the very top of the distribution where the association is stronger. The slope of the graph—the rank correlation—is 0.27 when both parents and children are in mid-life (age 45-50). We find a U-shaped pattern when looking at the rank correlation as a function of child age with a correlation of 0.35 when children move into adulthood (age 20), going down to 0.17 in the mid-twenties and then moving gradually up again to 0.27 in the forties. After death of parents, the correlation lies in the range 0.35-0.40. We provide a simple theoretical frame- work to understand intergenerational wealth mobility over the life-cycle. The theory explains the life-cycle pattern in measured wealth mobility through life-cycle patterns of transfers and earnings: wealthy parents make inter vivo transfers early in childrens’ life, their children have low income in the twenties when investing in human capital, but a high permanent income and, finally, they receive large bequests. The U-shaped pattern requires that inter vivos transfers are quantitatively important and the increase in correlation at the receipt of bequests reveals the quantitative importance of bequests. Our main interest is in the correlation across gener- ations in lifetime resources, which according to the theory may be captured by appropriately estimating intergenerational correlation in wealth. Our preferred estimate of the intergenera- tional correlation in lifetime resources is 0.25, which is significantly higher than the correlation of permanent incomes equal to 0.20.