Administrative Burden and Targeting: Evidence from SSI-Medicaid Automatic Enrollment for Children with Disabilities

• In Event: How State Policy Shapes Safety Net Access and Well-Being for People with Disabilities

Abstract

Administrative burden poses a significant barrier accessing Medicaid. While ordeals such as administrative burden can improve program targeting in some settings, they may also exacerbate selection in certain markets. In this paper we examine the implications of removing a targeted ordeal in the Medicaid application process: automatic enrollment for Supplemental Security Income (SSI) recipients. Thirty-four states and the District of Columbia have 1634 agreements with the Social SecurityAdministration (SSA) whereby SSA’s eligibility determination for SSI also serves as an eligibility determination for Medicaid, and SSI recipients are also automatically enrolled in Medicaid via this process. In states without 1634 agreements, SSI recipients must apply separately for Medicaid and/or meet additional eligibility requirements. Few states have changed their 1634 status since SSI’s inception in 1974; however, Ohio implemented a new 1634 agreement in August 2016. 
Using Medicaid T-MSIS demographic and outpatient claims data from 2015-2020 in Ohio, we examine enrollment and utilization patterns of children receiving SSI before and after the implementation of automatic enrollment. We develop a microsimulation model to simulate enrollment under a counterfactual without automatic enrollment and use this model to quantify the size of the compiler population. Then, to understand the characteristics and needs of the complier population, we use a regression discontinuity in time approach to examine differences in claim characteristics and utilization patterns between children enrolled in Medicaid via SSI before and after the introduction of automatic enrollment, and then implement a differences in differences approach comparing these utilization patterns to other children enrolled in Medicaid. We find that application administrative burdens significantly increase selection in this market: prior to the introduction of automatic enrollment, children on SSI who were enrolled in Medicaid had significantly higher rates of severe diagnoses and had significantly higher rates of utilization, in particular utilization of specialty services including behavioral health services compared to marginal beneficiaries enrolled via automatic enrollment. Children enrolled in Medicaid due to the policy use services less frequently than those previously enrolled, but still use significantly more services than children in Medicaid who do not have disabilities.

Author

• Stephanie Rennane, RAND Corporation
  Presenting Author