Supplemental Security Income youth face unique challenges in the transition to adulthood and have poor labor market outcomes (Davies et al. 2009; Deshpande 2016). Each year, 50,000 children turn 18 and undergo an age 18 redetermination for qualification, based on adult medical criteria. Evidence indicates that families underestimate the likelihood of removal at age 18 and, as a result, may underinvest in their child’s human capital because they expect the child’s SSI benefits to continue in adulthood.
Despite this qualitative evidence, there is no empirical evidence on the effect of beliefs about the likelihood of SSI removal at age 18 on human capital investment. In Year 1 of this project, currently funded through the DRC, we are conducting a randomized controlled trial (RCT) of the effect of providing “knowledge” (information about the likelihood of removal at age 18) and “nudges” (opportunities for human capital investment) on human capital investment and disability receipt. While Year 1 outcomes will provide immediate evidence on the effect of age 18 information on human capital investment, measuring the effects on the most meaningful outcomes and understanding the mechanisms behind these effects will require a combination of administrative and survey data over a longer time period.
We propose to evaluate the effect of the “Knowledge and Nudges” RCT, implemented in Year 1, on intermediate outcomes of SSI youth and the mechanisms through which these effects operate. To evaluate intermediate outcomes, we will use SSA administrative data on age 18 redetermination outcomes, DI and SSI receipt, and parent and youth earnings. Conditional on data linkages, we will also study effects on criminal behavior, grade completion, and program participation by linking the RCT data to state criminal records and school records. Finally, we will conduct an endline survey of beliefs and self-reported investment behavior to understand mechanisms, specifically the change in beliefs about the age 18 redetermination and human capital investment responses to those changes in beliefs. Specifically, we will do the following:
• Use SSA administrative data to evaluate effects on disability receipt and employment
• Use state data to measure effects on educational achievement, criminal involvement, and participation in government benefit programs, conditional on data linkages
• Implement an endline survey to measure changes in beliefs and investment behavior