Cancian, M., Yang, Mi-Youn, & Slack, K. S. (2013). The effect of additional child support income on the risk of child maltreatment. Social Service Review, 87(3), 417–437.
In the first longitudinal study of its kind, Cancian, Yang, and Slack (2013) found that small amounts of additional household income lowered the risk of child maltreatment. Researchers have long noted the strong correlation between poverty and child maltreatment, particularly child neglect, but identifying why this relationship exists has been more difficult to understand. This groundbreaking finding supports the hypothesis that income has a causal relationship to risk of child maltreatment.
The US-based study took advantage of a statewide pre-existing quasi-experimental research project that randomly assigned additional income to low-income mothers already receiving public financial support. In the late 1990s, Wisconsin proposed a cash welfare policy reform to allow mothers to receive the full amount (full pass-through) of child support payments from their child(ren)’s father(s) without a corresponding reduction in cash welfare benefits. Federal guidelines required a randomized control designed study evaluating the policy change in order to assess the benefits of the proposed reform. Thus, the social assistance recipients were randomly assigned to two groups: 1) those receiving full pass-through of child support (the experimental group); and 2) those receiving partial pass-through of child support (the control group). Participants that received full pass-through received only a small amount more than those that received partial pass-through (on average $100 more per year). The two groups of mothers did not significantly differ on demographic or case-level details; half were aged between 17 and 25, 68% were Black, 55% had not graduated high school, and 60% had children under three years of age. The majority of mothers had low earning potential, low labor market participation, and had not received child support in the year prior to the random assignment.
Administrative data on the final sample of 13,065 mothers were collected for the two-year period following the mother’s assignment to full pass-through or partial pass-through child support groups. In that two-year period, the group of mothers who received the full pass-through of child support was found to be approximately 10% less likely to have a child involved in a screened-in child maltreatment report, significantly less than the group of mothers who received the partial pass-through.
Because the study had a strong random assignment design, it used relatively simple multiple logistic regression in its analyses. Additional covariates, such as education level, age, and child support payment amount were included in the regression models to control for chance variation between experimental and control groups. Sensitivity tests explored the strength of the association between experimental group status and child maltreatment risk, revealing a strong negative association between child welfare involvement and child support income. Limitations of the study are many, including that the sole indicator of child maltreatment risk was a screened-in child maltreatment report, which may not be the best proxy for child maltreatment. Other limitations include possible endogenous post-assignment outcomes such as fertility, child support paid, and family formation. Importantly, the families studied were generally very low-income and low-asset, and thus the findings can only be generalized with caution. While the $100 per year difference in additional income was found to have a significant impact on child maltreatment risk for these families, this small amount of additional income may not have the same impact on higher income/asset families.