Physician Payment and Demand for Health Insurance: Evidence from Medicaid Primary Care Payment Parity

Although the means-tested health insurance program in the U.S. (namely, Medicaid) provides insurance almost free of charge to the disadvantaged population, the take-up rates of the program have long been far below 100 percent. This study seeks to examine a potential determinant of the non-take-up: access to primary care proxied by the reimbursement rate to primary care providers. Previous literature on the reasons for the incomplete take-up of Medicaid mainly focused on welfare stigma, information barrier, and transaction costs, little research has been done to understand the take-up barriers related to the program benefits. However, potential Medicaid beneficiaries face great access barrier to primary care. Primary care physicians are often reluctant to accept new Medicaid patients and, even for existing Medicaid patients, offer low-quality services. The most-stated reason for the nonparticipation is low reimbursement rates from physician surveys. Indeed, primary care is reimbursed at a particularly low rate. Medicaid reimbursed 66% of Medicare rate for all services and 59% of that for primary care services in 2012. Built upon this evidence, Hahn (2013) examined the relationship between the Medicaid benefit of access to primary care and Medicaid coverage among children. Compared to children, the uninsured rate among adults is twice as high, while the Medicaid participation rate is substantially lower. Further, adults tend to face a larger access barrier to care than children enrolled in Medicaid.

The Affordable Care Act (ACA) raised the primary care physician reimbursement rate of Medicaid to that of Medicare during 2013—14. Utilizing this change, this paper applies a generalized difference-in-differences method to examine the impact of Medicaid primary care physician payment on the Medicaid coverage rate among low-income non-elderly adults, using a novel dataset of Medicaid-to-Medicare primary care reimbursement rate ratios and the American Community Survey of 2010—14. The results show that the reimbursement rate ratios of office-based patient visits are positively associated with adults’ Medicaid coverage rate. For instance, a 10-percentage-point increase in the reimbursement rate ratio of 30-minute new patient office visit is associated with a 0.40-percentage-point increase in the Medicaid coverage rate among adults whose family income is below 250% of the federal poverty line. This association is most significant among the near-elderly (aged 50—64), non-parents, African-Americans, and those living in urban areas. Also, the reimbursement rate ratios are negatively associated with the uninsured rate, but not with the privately insured rate among adults.

The contribution of this study is three-fold. First, this study, to my knowledge, is the first to examine the effect of Medicaid primary care reimbursement rate on the Medicaid coverage among adults. Second, the findings of this study suggest that the Medicaid physician payment policy is effective in promoting public insurance take-up among potential beneficiaries. Third, this implication signifies the importance of the high Medicaid reimbursement rate in the context of the ACA’s Medicaid expansion. With the increased competition for physicians among newly enrolled and existing Medicaid patients, terminating the payment parity might substantially increase the patients’ access barrier to care and thus uninsured rate.