Note: The authors would like to thank Erica Socker, Senior Research Associate, and Michelle Shaljian, Associate Director of Communications, for their review and editorial assistance.

According to data released by the Department of Health and Human Services, one in five Americans now receive their health insurance through a state Medicaid program. Despite this increase in enrollment, it is estimated that 6 million Americans will likely remain uninsured because 20 states have decided not to expand Medicaid as the Affordable Care Act (ACA) envisioned. There are at least four states that are considering expanding Medicaid but have yet to do so.

Medicaid expansion continues to be one of the most politically charged directives of the health care law, mainly because the Supreme Court decision left the choice to states. This decision has generated an ongoing debate about whether and how states should expand their Medicaid programs. For example, an intense debate has been underway in Virginia, over the decision to include Medicaid expansion in the state budget; putting Democratic Governor Terry McAuliffe at odds with the Republican State Legislature. Similar debates are occurring in states across the country, and are further complicated by states’ option to pursue alternative expansion approaches under a Medicaid waiver. For states that have not yet expanded the program, the success of these alternative expansion models may influence whether they can find a politically feasible path forward.

Apart from the political battles, some states are concerned about the financial impact of expanding Medicaid; an understandable concern given that Medicaid makes up an increasing share of state budgets. The ACA ensures that expansion states will receive a 100 percent match to cover the cost of new enrollees through 2016, but the share declines to 90 percent in 2020. Many states fear that the federal share will decrease further, potentially leaving them with a huge cost burden.

Other states believe that public assistance programs are not the best way to expand coverage to low-income individuals. For these states, expansion is a tradeoff between covering additional beneficiaries, many of whom are uninsured, and enlarging the traditional Medicaid program, which they see as costly and inefficient.

For states that are concerned about the fiscal implications of Medicaid expansion, three points are relevant. First, while Congress could choose to reduce the matching rates further in the future, it is unlikely. With millions of people enrolled and over half the states in the country adopting the ACA expansion, there is a powerful constituency to support the current payment arrangements.

Second, as researchers have pointed out, the ACA was paid for through many changes in provider payments and revenue increases, which affect providers and taxpayers in all states. States that choose not to accept federal support for Medicaid expansion are bearing the cost without reaping the benefits of the law. For example, hospitals in all states will be subject to Medicaid payment cuts that are based on the assumption that uncompensated care costs would decrease due to expanded coverage. Hospitals in non-expansion states will thus continue to pay high uncompensated care costs at the same time that their payments are reduced.

Third, the Medicaid expansion may have positive impacts on state economic growth and state budgets. Expansion would result in more federal funds for health care coming into the state, which would lead to greater economic activity (increased employment and earnings) within the state. Typically, if a state was expanding Medicaid, higher health care spending would be offset by the state funds needed to expand. As previously mentioned, the federal government will pay 100 percent of the expansion costs through the first few years.

The law also offers considerable flexibility in the form of Medicaid waivers, which gives the Health and Human Services Secretary authority to approve state-proposed modifications to the Medicaid program that do not cost additional federal money. Waivers are often used to expand eligibility to specific groups, cover non-traditional services, or fund innovations in care delivery. Due to political tensions, some states are seeking waiver approval from the Secretary to expand Medicaid, but in a way that they believe better serves their population.

A Private Option: The Arkansas Experiment

One of the first states to propose waiver authority to expand Medicaid was Arkansas, where a “private option” allows the state to use Medicaid funds to purchase private health coverage from Qualified Health Plans in the marketplaces. Politically the program facilitates a compromise, enabling states to accept billions of dollars from the federal Medicaid match without acquiescing to the demands of Obamacare.

Proponents of this approach believe it can help spread the risk among enrollees, increase competition in the marketplace, mainstream care for millions of poor Americans, and potentially save money in the long-term by reducing “churn” between Medicaid and private plans. With private payment rates to providers generally higher than Medicaid reimbursement, the question is whether these efficiencies can offset higher unit costs. That is a legitimate policy question, given that the fate of the private option will rest on keeping costs under control.

The state has already seen strong demand for private option coverage. As of April 2014, 170,000 Arkansans were enrolled, compared to 250,000 expected enrollees. These enrollees make up 77 percent of individuals in the state’s ACA health insurance marketplace. They also tend to be younger and healthier than the rest of the marketplace, thus diversifying risk and potentially lowering premium costs. While the preliminary evidence suggests Arkansas’s private option helped create a robust marketplace risk pool in the first year, their Medicaid expansion has come under scrutiny lately for exceeding estimated costs. Given that this was thought to be a more efficient approach than traditional Medicaid, and that other states have been considering an Arkansas-like model, a closer look at these arguments is merited. Our review of the evidence suggests that while it is too soon to draw a firm conclusion, the data do not suggest fundamental problems and indicate the concerns about costs may be misplaced.

Age of enrollees skewed premium expenditures. Arkansas estimated that the private option demonstration, a 3-year test program, would cost $118 million in 2014, the same amount it would cost to cover the new population under traditional Medicaid. During the first few months, the average enrollee premiums exceeded estimates, ranging from an estimated $472 to $478 per-member-per-month (PMPM), and then costing $485 PMPM by March. However, the primary reason for higher-than-expected premium costs was that actual enrollees were on average older than had been assumed for the cost estimates. Arkansas had originally estimated that the average age of private-option enrollees would be 37, but enrollment statistics revealed an average age of 39.

In fact, after adjusting for the actual age of the population, Arkansas’s costs were lower than the target. A difference between the estimated and actual enrollment is to be expected in any program, and this difference does not by itself indicate a fundamental cost problem.

“Beyond the basic” benefits increased premiums. One of the carriers offering plans on the Marketplace, Centene, added dental and vision benefits to its silver-level plan. While these are a valuable addition to existing benefits, they do add costs. Since, at the moment, private option enrollees can choose any silver-level plan, the plans with added benefits attracted more enrollees and increased total premium costs. The Arkansas Insurance Department has already recognized the issue and ruled that carriers in the Marketplace in 2015 must offer at least one plan that covers only the ten essential health benefits (EHB). The Department of Human Services expects to issue a related rule later this year that the private option will only purchase these EHB-only silver plans, which should help mitigate this problem and keep costs under control. The Department plans to take further action to limit costs in future years.

Despite some initial concerns, the Arkansas model remains a promising and reasonable path for states that are considering alternative approaches for expanding their Medicaid program. However, the recent primary defeat of Republican state Rep. John Burris, a private option supporter, has put the fate of the private option in jeopardy, at least in Arkansas. The program must be renewed under the state’s appropriations bill each year and was only narrowly renewed this year. In spite of the expanded coverage provided by the private option and the positive effect it has had on the state Marketplace, Arkansas may choose not to renew the program next year.

The potential of broader coverage from the ACA will require considerable flexibility on behalf of the federal government to help states overcome their concerns about expanding Medicaid. Each state proposal will have to be examined on a case-by-case basis, but options like the Arkansas model have the potential to help both state budgets and low-income beneficiaries.