How does receiving welfare benefits affect children later in life? On the one hand, we would expect more resources during childhood might translate into better outcomes—like higher earnings—because the payments allow parents to cut back on work and spend more time with their kids, or to buy things like food or learning materials. Of course, the size of the effect will depend on what parents spend the money on. On the other hand, disability programs might incentivize certain behavior to maintain eligibility, such as reducing household earnings or child school performance. In sum, how transfer programs affect children depends on how families respond to them, and on how those responses affect child development.
In this paper, the authors tackle these questions in the context of Supplemental Security Income (SSI) payments, which provide cash transfers and Medicaid eligibility to adults and children who have disabilities and live in households with limited income and assets. They utilize data covering parents’ responses to SSI—that is, whether mothers replace lost benefit income by working more, or leave their household income to fall–—alongside information on children’s own earnings in adulthood.
Effects on SSI children and siblings when removed during childhood
Using these data, the authors compare adult earnings for children who were more likely to be removed from SSI after a 2004 policy change to those who were less likely to be removed, based on when their families happened to enter the program. They observe these effects separately for SSI beneficiaries and their siblings, allowing them to disentangle whether the effects are owed to perverse incentives: eligibility rules that discourage improvement. Because SSI requires a child to keep meeting medical criteria, families may have reason to avoid investments that make the child appear less disabled. . (If parents hold back their qualifying child’s development in order to keep them eligible, being removed from SSI may lead to gains later in life. Siblings are not subject to eligibility considerations, so an earnings gain concentrated among SSI children (and not their siblings) would suggest perverse incentives.)
Comparing the adult earnings of children whose families entered SSI just before the review cutoff to those who entered just after—and comparing SSI recipients to their siblings within each group—the authors find the following:
- Children who were more likely to be removed from SSI because of an eligibility review, and their siblings, have modestly higher earnings in adulthood. A 4.9 percentage point increase in the probability of being removed from SSI during childhood is linked to a $520 increase in annual earnings after age 20.
- Parents respond to the child’s SSI removal by working more. Mothers choose to reduce their non-work time to prevent household income from falling, increasing their earnings by about the same amount as the SSI income drop and leaving annual household income roughly stable.
- There is no evidence that children benefit from removal from SSI because of perverse incentives. The effect of SSI removal on future earnings is no larger for the SSI children than for their siblings. Similarly, the effect is no larger for children with behavioral conditions, which may be more malleable and responsive to parents’ choices, than for children with physical and intellectual conditions.
Effects on younger siblings when SSI child removed at age 18
Building on these results, the authors analyze a second quasi-experiment, this time of a 1996 welfare reform that increased the likelihood of 18-year-olds being removed from SSI due to medical reviews. To measure the effect of the 18-year-old’s removal, they compare the adult earnings of younger siblings in families where the 18-year-old was more likely to be removed to those where removal was less likely, based on when the older sibling happened to turn 18. They find the following:
- In apparent contrast with the previous results, removing an 18-year-old from SSI reduces the adult earnings of younger siblings in the household. A 43 percentage point increase in the probability that the older sibling loses SSI is tied to a $515 drop in the average annual earnings of the younger siblings.
- Unlike in the first scenario, where parents offset the benefit loss by working more, in this case neither the parents nor the older sibling replaced the lost SSI income with additional earnings, and household income fell by over $17,000 in total during the younger sibling’s childhood. This reflects the fact that mothers who were already working at the time of an SSI loss tend to respond by working more, while mothers who were not working have no earnings response—and the age 18 sample has far fewer working mothers.
- The difference in results between the two experiments appears driven by diverging employment responses to losing benefits. When household income remains steady after SSI removal, children go on to earn more; when it falls, they earn less. This pattern holds across both experiments and regardless of the child’s age, sex, or condition, or the mother’s work history.
A model of parental work and child development
Together, these results point to household income as the mechanism that drives the impacts of receiving SSI on children’s outcomes in adulthood. Building on this result, the authors next estimate how much of the effect is owed to household resources and how much to perverse incentives. They develop a model of parental labor supply and children’s human capital formation, which reveals the following:
- Consumption plays a key role in child human capital formation for households receiving SSI, and it is difficult to replace. A decrease of $10 per week in consumption has the same effect on a child’s human capital development as a three-hour per week decrease in parent non-work time. This suggests that even a minimum wage job brings in enough to more than make up for the hours it takes away from time at home.
- Eliminating childhood SSI would lead parents to increase their earnings to some extent, but not enough to offset the lost SSI income. Household income would fall 11% if SSI were eliminated, and children’s human capital—measured as their adult wages—would fall by 4.7% for the SSI child and 4.9% for their sibling.
- Doubling the SSI amount in childhood would moderately reduce mothers’ employment while modestly raising the earnings potential of SSI children and their siblings. Even after mothers cut back, household income would rise 20% on net, and human capital would increase by 2.8% and 3.3% respectively.
- Doubling SSI in both childhood and adulthood, however, would reduce children’s human capital. This is due to two countervailing effects: although additional resources in childhood increase human capital investment in children, higher generosity of adult SSI increases perverse incentives to keep children on SSI in adulthood.
- Turning the SSI cash benefit into an in-kind benefit: goods or services provided directly—medical treatment, food, housing—rather than cash a household can spend as it chooses leads parents to work more, since a benefit delivered as goods or services leaves the family in greater need of cash. The effect of this change on the SSI child’s human capital depends on how much the in-kind transfer bolsters human capital compared to cash. Even if a dollar of in-kind support does only a quarter as much as a dollar of cash, the child still comes out ahead of the current cash benefit. Non-SSI siblings experience a decrease in human capital regardless of the efficacy of the transfer, because they no longer benefit from a general increase in household resources.
- Perverse incentives are relatively small compared to the benefits of SSI. In a scenario where parents no longer anticipate future medical reviews, and so have no reason to limit their child’s development, children’s human capital rises by roughly 1%. Most of that effect is due to the prospect of the age 18 review rather than reviews during childhood.
The long-term effects of SSI on children appear to depend on the family context: When parents continue working and household income increases, SSI appears to increase children’s earnings potential in adulthood. These findings suggest that resources are an important driver of human capital development, and understanding how resources are allocated within the family, including leisure and investments, is key to capturing the full effect of a change in a welfare program.