Parents’ Use of Government Assistance Drives Use in Next Generation

Gordon Dahl

Does the use of government assistance by parents make their children more likely to use welfare, too? Yes, suggests research coauthored by University of California, San Diego economist Gordon Dahl.

The question has been a difficult one and has fueled policy debates for decades. On the one hand, people have argued that “of course it’s cultural.” Kids learn from their elders in all sorts of ways. And it is not hard to imagine that they learn how to navigate the bureaucracy or that the family experience reduces stigma. On the other side, there has been the correlational argument: Poverty and ill health are the real drivers, and as poverty and ill health are inherited so too is the need for assistance.   

Dahl’s work shows a causal component to intergenerational welfare use. But he is careful to note that it doesn’t mean there isn’t also a significant contribution from the cycle of poverty itself or other factors that the study design cannot capture.

Dahl and colleagues from University College London and the University of Bergen analyzed 10 years of disability-insurance appeals claims in Norway and 15,000 parent-child combinations. The researchers focused on Norway, instead of the United States, chiefly because there is national data linking parents and children making it possible to map out family relationships over time for the entire population.

They “find strong evidence,” they write in their National Bureau of Economic Research working paper, “that welfare use in one generation causes welfare use in the next generation: when a parent is allowed [disability insurance], their adult child's participation over the next five years increases by 6 percentage points. This effect grows over time, rising to 12 percentage points after ten years.”

The 6- and 12-percentage points are a net difference, Dahl said. The reverse effect is also observed: Adult children of people whose initial claims are denied apply at much lower rates than would be expected in the general population.

Critical to their analysis – and being able to get at causation and not just correlation – is that these appeals claims are randomly assigned. One simply cannot shop around for a judge. It’s luck of the draw. And some judges are systematically more lenient, allowing up to 25 percent of appeals, while others are systematically more strict, approving as few as 5 percent.

“As far as we know, we are the first quasi-random experiment which can estimate causal intergenerational links in any type of welfare or social safety net,” said Dahl, professor of economics in the UC San Diego Division of Social Sciences, a research associate of the NBER and a fellow of the Stanford Center for the Study of Poverty and Inequality.

The study focused on appeals because that removes the problem of whether people who apply for disability insurance in the first place are somehow different from people who don’t. In the study’s scenario, all members of the parent generation had already all applied and so were identical in that respect. Also, about 70 percent of appeals are on “difficult to verify” conditions, such as musculoskeletal or mental disorders. This is a policy-relevant group and the one that would be most affected by changes in application criteria.

And why disability insurance, instead of some other form of government assistance? Because it is a large and growing share of the social safety net. “In the U.S., for example,” the researchers write, “the outlays for DI exceed those for food stamps, traditional cash welfare, or the [Earned Income Tax Credit].”

“For families without small children, DI is often the only cash benefit available after unemployment benefits run out,” Dahl and colleagues note. And, in most industrialized countries, disability-insurance rolls have been growing steadily over the last 50 years – rising, for example, from 2 to about 10 percent of the population in Norway, and from 1 percent to more than 5 in the U.S.

The study suggests that national budget decisions made without taking the intergenerational link into account will be flawed: Contraction or expansion of a social safety net program will change the rolls well beyond the people affected directly and in the immediate short term. 

“We don’t have a stand on whether people should or shouldn’t be on disability insurance,” Dahl said. “But we do say that you have to be careful when making a program stricter or less strict. There are important spillovers that will affect the next generation as well.”