Effects of early Medicaid expansions on disability income receipt and labor market outcomes

The 2010 Patient Protection and Affordable Care Act was intended to broaden Medicaid eligibility to include most adults with incomes below 138% of the Federal Poverty Level by January 2014.  In preparation for this change, five states -- Connecticut, New Jersey, Minnesota, Washington, California -- and the District of Columbia expanded their Medicaid programs early, in 2010 and 2011.[1]  In addition, Massachusetts expanded its Medicaid program in 2006, as part of a state health care reform initiative.

In this paper, we test whether these early Medicaid expansions affected receipt of Supplemental Security Income (SSI) and Social Security Benefits, as well as labor market outcomes, among low-income adults. The early Medicaid expansions may have affected these outcomes via three competing channels:

• Qualifying for SSI makes individuals eligible for Medicaid in most states. After the Medicaid expansions, people who receive/ will receive SSI income are encouraged to continue to work and leave /not enter the SSI rolls since they now can obtain Medicaid without enrolling in SSI.

• The Medicaid expansions eliminate insurance-motivated employment lock, and encourage some currently employed (full-time) workers to leave their job (or become part-time workers). 

• Medicaid coverage improves health outcomes and encourages people to stay at jobs rather than seek disability assistance

Data for this study come from the 2000-2013 American Community Survey.  We use a difference-in-difference approach, comparing average outcomes among low-income individuals before and after the Medicaid expansion in states that did and did not expand Medicaid. 

The disability income outcome variables include: (1) in past year, individual only received Social Security (SS); (2) in past year, individual only received SSI; (3) in past year, individual concurrently received SS and SSI; (4) in past year, individual received either SS or SSI income. We also examine a range of labor market outcomes including labor force participation, employment, hours worked, and earned income.  The control variables include state fixed effects, year fixed effects, state-specific linear time trends, as well as socio-demographic characteristics and measures of disability.  We estimate robust standard errors adjusted for clustering on state. The sample is limited to individuals living in states in the treatment group (the early expander states) and the comparison group (states that expanded Medicaid later). 

Our preliminary findings indicate that the early Medicaid expansions are associated with reductions in the likelihood of receiving SSI income only as well as reductions in the likelihood of receiving concurrent SS and SSI income.  We also find that the early Medicaid expansions are associated with increased likelihood of current labor force participation, but other effects on labor market outcomes are not statistically significant.

Currently, we are working to refine the comparison group for this study, and we are considering alternative specifications which are better able to capture state-specific trends.  In future work, we also will consider issues related to the relatively small number of clusters in this study.

---

[1] New Jersey and Washington only transfer individuals on pre-existing state-operated public health insurance to Medicaid rolls and do not accept new applicants.