Solutions to expanding heath care coverage in the U.S. are often incremental and focus on mitigating market failures. In new research, Katherine Baicker, Amitabh Chandra, and Mark Shepard suggest reframing the conversation around the establishment of a minimum floor of universally covered health care services. This article originally appeared on JAMA Forum.

The U.S. spends substantially more on health care per capita than other high-income countries yet leaves a greater share of the population uninsured. Traditional economic models—and common sense—point to the benefit of having at least some health insurance, if only for financial protection. In addition, ample evidence has shown that health insurance provides greater access to beneficial care and can improve health and save lives. Many people also place social value on others’ access to health care as part of a social safety net that also includes access to food, housing, and education.

Why, then, are so many in the U.S. uninsured? Understanding the underlying causes sheds light on different options for expanding insurance. The first explanation may be that insurance is expensive and many people simply cannot afford it. But this factor is not the only or main reason. About 40% to 50% of uninsured individuals likely qualify for no-cost insurance via Medicaid or an Affordable Care Act (ACA) exchange plan, and many others qualify for heavily subsidized insurance. Although some populations fall into gaps for subsidized coverage—notably undocumented immigrants and low-income people in states that have not adopted Medicaid expansion—lack of eligibility for affordable coverage is not the only barrier.

Other explanations point to market failures that make health insurance a bad deal for some people. Health insurance markets suffer from serious information failures—for example, insurers’ limited information about enrollees’ existing health needs and enrollees’ limited information about the potential plan’s quality and comprehensiveness of care—and from a lack of competition that drives up health care prices and insurance premiums in many areas. But these factors do not explain why many people do not take advantage of benefits available to them at no cost. Evidence shows that individuals’ behavioral biases and frictions, including the complexities of Medicaid and ACA exchange enrollment processes, may pose barriers.

The growing body of research on these barriers often encourages incremental policy approaches to expanding coverage: correcting each market failure and implementing nudges and administrative simplification to increase enrollment. Indeed, the ACA itself and more recent policies to amend it take just such an approach. The result is a patchwork of insurance policies that are incomplete and expensive in terms of the cost to administer them and the health consequences of inconsistent coverage and care. Tweaks to the existing system also perpetuate other shortcomings, such as job lock that comes from employment-based coverage, regressive financing mechanisms, and limited incentives for investing in population health.

Instead, it may be advantageous to begin with a policy that sets a social floor or basic policy that would be available to everyone. Starting with this premise would force explicit decisions about crucial tradeoffs that are already faced implicitly in the current system. The existing implicit social floor in the form of uncompensated care, emergency department visits, and free clinics is inefficient, unpredictable, and highly variable. Implementing a publicly financed basic policy with automatic enrollment could facilitate a move toward universal coverage in a financially sustainable way that ensures access to care with substantial health benefits.

We recently outlined how such an approach might work. First, this approach requires defining the floor to which everyone will be automatically entitled: How much insurance and health care access does society want to make universally available? Should publicly financed insurance cover all care, regardless of how low the health benefits or how high the costs, or should there be limits? We suggest that coverage decisions be grounded in how much health benefit a service generates, ensuring access to high-value care for all. High-value care is not the same as low-cost care: some very expensive treatments with dramatic health benefits are high-value care, and some cheap treatments with negligible health benefits are low-value care. Similar tradeoffs arise in deciding how much to pay health care professionals, which determines how many and which types of physicians and hospitals will accept basic coverage, as in Medicaid today.

The second step is determining who decides how much to pay for which services and for which patients? To mitigate concerns about the flexibility and innovation generated by one-size-fits-all public programs, public subsidies can be coupled with choice among plans, as in market-based social health insurance in the Netherlands and Switzerland as well as in Medicare Advantage and the ACA Marketplace plans in the U.S.

Third, decisions must be made about whether and how individuals can use private funds to buy additional coverage. For example, should people be able to opt out of the public system and replace it with separate private insurance as occurs in Germany? Or should they be allowed to “top up” the public insurance with supplemental private insurance that covers more treatments or reduces patients’ cost sharing, similar to supplemental policies in England and Canada that cover a wider set of clinicians and hospitals? These decisions have economic as well as ethical and distributional implications. Allowing additional coverage means that those with higher incomes are likely to have more health care and better outcomes than those with lower incomes. But this policy also enables people to find insurance that more closely matches their preferences and priorities. Furthermore, the presence of private market choices can drive innovation and quality. Lessons can be learned from the experiences of other countries, many of which have some version of a universal basic system, although with different answers to these fundamental questions. Almost all universal systems include options for supplemental coverage.

Beyond these fundamental questions, moving to such a system raises real concerns about disruption to clinical relationships, the risk of having the government as a monopsonist payer setting prices that are too low for access and medical innovation, and myriad logistical challenges. Despite these challenges, few would argue that the current U.S. health care system is serving the nation well; the system surely spends too much on health care that delivers too little benefit to too few people. Reconceptualizing universal coverage to ensure that public resources are devoted to care with high health benefit offers the opportunity to provide universal access to innovative care in an affordable system.

This article originally appeared on JAMA Forum as “A Different Framework to Achieve Universal Coverage in the US.” We have lightly edited the article for style.

Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.