Intergenerational Transmission of Welfare Dependency

When parents are awarded DI, the likelihood that one of their adult children will participate in DI rises by 12 percentage points over the next decade.

The extent to which welfare dependency is perpetuated from one generation to the next is a question of great social importance for which there is only limited empirical evidence. In Family Welfare Cultures (NBER Working Paper No. 19237), Gordon Dahl, Andreas Kostol, and Magne Mogstad analyze this question using data from the Norwegian disability insurance (DI) system. They study the outcomes of appeals by claimants who were initially denied DI benefits. Judges are randomly assigned to DI appeals, and some are systematically more lenient than others.

The authors investigate whether the children of those whose appeal cases were assigned to lenient judges, and who were therefore more likely to be approved for DI receipt, are more likely to draw welfare benefits than the children of those whose DI appeals were heard by less lenient judges. They verify that the individual characteristics of the appellants are uncorrelated with the assignment of the judges. Their findings indicate that if parents become welfare dependents, the likelihood of their children eventually becoming welfare recipients also increases. Specifically, when parents are awarded DI, the likelihood that one of their adult children will participate in DI rises by 6 percentage points over the next five years, and 12 percentage points over the next decade. These findings suggest that a more stringent screening policy for DI benefits would not only reduce payouts to current applicants, but would also have long-run effects on participation rates and program costs. The results underscore how important accounting for intergenerational effects can be when making projections of how participation rates and program costs may be affected by program reforms.

The authors explore the cultural mechanisms behind the intergenerational relationship that they discover. They do not find any evidence that parental participation reduces the stigma of participation, or that their findings are driven by differential child investments by parents on DI. They conclude that the cross-generational effects may arise from children learning from a parent's experience with the DI program.

--Matt Nesvisky