Breaking the cycle? The effect of education on welfare receipt among children of welfare recipients

Abstract

We examine the impact of high school graduation on the probability individuals from welfare backgrounds use welfare themselves. Our data consists of administrative educational records for grade 12 students in a Canadian province linked with their own and their parents' welfare records. We address potential endogeneity problems by: 1) controlling for ability using past test scores; 2) using an instrument for graduation based on school principal fixed effects; and 3) using a HeckmanSinger type unobserved heterogeneity estimator. Graduation would reduce welfare receipt of dropouts by ½ to 3/4. Effects are larger for individuals from troubled family backgrounds and low income neighbourhoods. Welfare systems in a number of developed countries have undergone reforms in recent years, with some of the reforms being extensive. A key focus in many reforms has been the desire to break "cycles of dependency" in which welfare receipt at one point in time is believed to generate higher probabilities of receipt in the future. Of greatest concern are inter-generational welfare traps where welfare receipt by parents may lead to increased welfare dependency among children once they grow up. In several countries, empirical work has shown a strong positive correlation between welfare rates of successive generations within families. Whether this correlation reflects a causal impact of parental welfare use on children's behavioural patterns or simply correlations in unobserved characteristics common to the parents and children has been the focus of some debate.Several, though not all, investigations of this issue for the US point to the conclusion that a substantial part of the correlation is causal. This result has led some to advocate a policy of reducing access to welfare for current adults in part in order to reduce dependency of future generations. We are interested in investigating an alternative policy response: breaking the cycle by improving the educational outcomes of children of welfare recipients. Thus, we examine the effects of increasing the education levels of children from welfare households on their future welfare use . 1 We study the impact of high school graduation on the probability individuals from welfare backgrounds use welfare themselves at each age between 19 and 24. Our data is a unique linking of educational records from all individuals entering grade 12 in the Canadian province of British Columbia (BC) between 1991 and 1996 with both their own and their parents' welfare records. The education data includes information on whether the individual graduated from high school (i.e., completed grade 12), whether he or she had failed any earlier grades, test scores from exams in grade 11, and characteristics of the school. We identify children from welfare backgrounds as 2 students from households which received welfare income at any time before the child turned 17. For the children themselves, we know whether they received welfare at each age from when they first become eligible for welfare (age 19) to the end o...