The Impact of Medicaid Expansion on Employer Provision of Health Insurance Jean Abraham University of Minnesota Anne Royalty Indiana University – Purdue University Indianapolis Coleman Drake University of Minnesota ARM 2017 Presentation June 26, 2017
Acknowledgements & Disclaimer • We thank the SHARE program of the Robert Wood Johnson Foundation for financial support. • Any opinions and conclusions expressed herein are those of the authors and do not necessarily represent the views of the U.S. Census Bureau. All results have been reviewed to ensure that no confidential information is disclosed.
Background • Employer-sponsored insurance (ESI) – 46% establishments offered ESI in 2015 – Declining over time • ACA changed insurance options – Low/Middle-Income: Marketplaces – Low-Income: Medicaid – Employer Mandate (2015)
Research Question How has Medicaid expansion impacted ESI offers, workers’ ESI premiums, and workers’ ESI eligibility?
Prior Literature • Insurance Expansions and ESI – Shore-Sheppard et al. (2000) – Buchmueller et al. (2005) • Effect of ACA on Employer Behavior – Abraham, Feldman, Graven (2016) – Abraham, Royalty, Drake (2016)
Conceptual Framework • Pauly and Goldstein (1976): Employers max. employees’ utility by offering their pref. comb. of wages, benefits s.t. employers min. labor costs • ACA and Medicaid Expansion – Reduces value of benefits vs. wages – Reduce benefits on various dimensions – Likely focused on low-wage workers
Data • MEPS-IC 2010-2015 – Repeated cross-sections – Approximately 141,900 establishments • Supplemental Data – County-level unemployment (BLS) – State-level minimum wage law (DoL) – Medicaid expansion details (KFF)
Key Variables • Outcomes 1. Establishment offers insurance 2. Lowest single out-of-pocket premium 3. Percentage workers not eligible for offer • Policies – Medicaid Expansion – Employer Mandate
Establishment Characteristics Outcome 2010-2013 2014-2015 49.13 44.19*** Offer $1,033 $1,010 S. OOP Prem. 19.43 20.47*** Pct. Not Elig. 33.55 30.56*** Pct. Low Wage 99,000 42,900 Sample Size
Econometric Model (DID) 𝑃𝑣𝑢 𝑗𝑢 = 𝛾𝑌 𝑗𝑘𝑢 + 𝛿𝑆 𝑗𝑢 + 𝜀𝑁 𝑡𝑢 + 𝜃 𝑡 + 𝜄 𝑢 + 𝜈 𝑗𝑘𝑙𝑢 – Estab. 𝑗 , county 𝑘 , state 𝑡 , year 𝑢 – Outcomes 𝑃𝑣𝑢 – Controls 𝑌 – Employer mandate 𝑆 – Medicaid expansion 𝑁 – Fixed effects 𝜃 (state), 𝜄 (year)
Estimation • Offer: Linear Probability Model • Single OOP Premium: Two-Part Model – First Stage: Logit – Second Stage: GLM log link, gamma dist. • Pct. Not Eligible: Negative Binomial • MEPS-IC survey design
Average Marginal Effects Outcome Offer (pp) SOOP ($) N. Elig (pp) -0.0046 -36.5 4.01* Medicaid Exp. (0.0068) (34.4) (1.82) 0.061*** 149** -3.88* Emp. Mandate (0.0064) (49.6) (1.83) -0.0017*** 2.01*** 0.30*** Pct. Low Wage (0.00005) (0.3) (0.016) Dec. Inc. Not Sig. Year FE
Sensitivity Checks • DID Identification – Common Trends Assumption – Placebo Test for Pct. Not Elig • Other – Medicaid, Low Wage Interaction – Measures of Medicaid
Conclusions • Medicaid Expansion – No impact on offers, single OOP premiums – Decreased worker eligibility for offers (4pp) – No differential effect in low-wage estabs. • Employer Mandate – Increased offers – Increased SOOP – Increased eligibility
BCRA Implications • Medicaid expansion phase out – Unlikely to change offers, worker premiums or accompanying trends – Will increase worker eligibility, but perhaps at the expense of wages • Employer mandate repeal – May decrease offers, wrkr prems., eligibility
Coleman Drake Division of Health Policy & Management School of Public Health drake222@umn.edu I am on the job market!
Recommend
More recommend
