The top priority in health policy today should be providing health insurance to people in the so-called coverage gap. This group — which includes at least 2.2 million people —faces a Catch-22, an unanticipated result of the 2012 U.S. Supreme Court decision that made the Medicaid expansion provision of the Affordable Care Act (ACA) optional for states.

As a result, low-income people — those with incomes below 100 percent of the federal poverty level (i.e., $21,960 annual income for an individual) in the 14 states that have not yet expanded Medicaid are not eligible for traditional Medicaid coverage but also not eligible for marketplace coverage.

All states that have not yet expanded Medicaid (except Wisconsin) have restrictive coverage rules. In Texas, a parent in a family of three with annual family income at 17 percent of poverty — just $3,750 — would not be eligible for Medicaid. In nearly all the states, a nondisabled adult without dependent children would not be eligible for Medicaid at any income level.

There is consensus — at least among Democrats — that it is important to insure this population. States have new incentives to expand Medicaid but there is disagreement about what to do if they don’t act. There are three options under discussion: expanding marketplace coverage eligibility to include the coverage-gap groups, with subsidies to create zero-cost premiums and no cost sharing; implementing a new public option plan; or creating a federal Medicaid-like plan, which would contract with existing Medicaid managed care organizations.

Expanding marketplace coverage has many advantages. The federal government already operates successful marketplaces, with a choice of plans in all the nonexpansion states. These marketplaces have experience adjusting premiums and cost sharing based on beneficiaries’ income. While marketplaces operate open enrollment only once a year, they already have the capacity to enroll beneficiaries off-cycle. Finally, the marketplaces and their plans operate at a distance from adversarial state governments, which can ease implementation and monitoring burdens.

Some public option proponents reject the marketplace approach. They would prefer to create a new federal program that might initially enroll the coverage-gap population and then serve as a launching pad for a broader program. This may be unrealistic. In addition, politicians and analysts use the term “public option,” but it has no agreed-upon meaning. This lack of consensus works in abstract discussion but makes an immediate translation into policy very challenging.

Some analysts prefer a federal Medicaid approach, which would allow this population access to Medicaid’s special protections, including nonemergency medical transportation, certain behavioral health benefits, and Early and Periodic Screening, Diagnostic, and Treatment (EPSDT) services. These benefits are not included in the marketplaces’ essential health benefits (EHBs). Medicaid also has robust grievance and appeals protections and an automatic option to enroll at any time.

The benefits of a federal Medicaid plan approach are more modest than advocates hope for and the disadvantages — in terms of time to establish the program, political accountability for operations, and complexity facing beneficiaries — are greater than they acknowledge. Benefits included in Medicaid but not in the marketplace EHB are of use mostly to the disabled, elderly, or under-21 Medicaid populations. There are few people in the coverage-gap population over 65 or under 21. Some have disabilities but have not met a formal disability determination. In most cases, EHB benefits are usually sufficiently broad to address their needs. The limited evidence available suggests that claims denials are about as common in Medicaid as in marketplace plans, and that under existing regulations, few denials are appealed. While some benefits, such as behavioral health, are more generous in Medicaid than in the marketplaces — at least on paper — Medicaid’s low payment rates and limited provider participation often belie that promise.

A Medicaid-based expansion also would come with some very real costs. Currently, there is no federal Medicaid program, meaning a new bureaucracy would have to be built. Getting existing Medicaid managed care plans to participate could be a struggle; some states, like South Dakota and Wyoming, have small populations and no existing Medicaid managed care plans, so plans would need to be established. Partisan state governments could manipulate rates and benefits to load extra costs onto the federal plan. Some beneficiaries might need to navigate three entirely separate enrollment bureaucracies — traditional Medicaid, the new federal Medicaid program, and the marketplaces — if they experienced minor changes in work hours or wages over time.

All strategies to address the gap population would provide nonexpansion states with more federal funds than currently flow to states that have already expanded, because state governments pay 10 percent of the cost of the Medicaid expansion population and none of the cost of the marketplace. Under a marketplace-based plan or a public option, there also would be cross-state inequity at the provider level. Provider payment rates in marketplace plans are close to Medicare rates, and typically higher than Medicaid rates. Medicaid providers in expansion states might prefer that their states end the Medicaid expansion so they could collect the higher marketplace rates.

The federal government could address this by fully financing enhanced payment rates for selected groups of providers in existing expansion states, as it did for two years under the ACA. Adding this feature might even encourage nonexpansion states to expand, so that their existing Medicaid providers could gain access to this fee bump. Congress also could require pharmaceutical companies to offer Medicaid-equivalent rebates for prescriptions purchased by gap-population members in marketplace plans.

These complexities strengthen the case for a marketplace-based approach. This approach could be implemented relatively quickly and easily, which would be particularly useful for a temporary program. This is an important consideration: Congress will need to level the state and provider playing fields, and a temporary expansion will give them time to work out the details.

For the past seven years, in most states, Americans of all income levels have been eligible for affordable insurance coverage. A temporary expansion of the existing ACA marketplaces would effectively and rapidly extend the benefit to people in the coverage gap in nonexpansion states.