By Mark Shepard, Katherine Baicker, and Jonathan S. Skinner
National Bureau of Economic Research, November 2019
Abstract
There is increasing interest in expanding Medicare health insurance coverage in the U.S., but it is not clear whether the current program is the right foundation on which to build. Traditional Medicare covers a uniform set of benefits for all income groups and provides more generous access to providers and new treatments than public programs in other developed countries. We develop an economic framework to assess the efficiency and equity tradeoffs involved with reforming this generous, uniform structure. We argue that three major shifts make a uniform design less efficient today than when Medicare began in 1965. First, rising income inequality makes it more difficult to design a single plan that serves the needs of both higher- and lower-income people. Second, the dramatic expansion of expensive medical technology means that a generous program increasingly crowds out other public programs valued by the poor and middle class. Finally, as medical spending rises, the tax-financing of the system creates mounting economic costs and increasingly untenable policy constraints. These forces motivate reforms that shift towards a more basic public benefit that individuals can “top-up” with private spending. If combined with an increase in other progressive transfers, such a reform could improve efficiency and reduce public spending while benefiting low income populations.
From the Introduction
In this paper, we focus on the design of the current Medicare program for the elderly to assess its tradeoffs and provide insights about the implications of using it as a foundation for expanding coverage. We first ask about the efficiency and equity tradeoffs involved with its current generous, uniform design. Second, we address the question of how rising income inequality, ongoing medical technology innovation, and the budget pressures imposed by an aging population affect the efficiency of the current benefit structure. Finally, we examine the effects of an alternative, non-uniform benefit structure on economic efficiency and equity.
To study these questions, we build on a rich literature in health economics and social insurance design to develop a simple economic model of Medicare that incorporates income inequality, medical technology growth, and distortionary taxes. The model allows us to assess how the welfare consequences of Medicare’s uniform benefit structure have evolved, as well as the welfare effects of potential alternative public insurance designs.
The model suggests that while Medicare’s uniform benefit has the advantages of simplicity and lower administrative costs, it also comes with a cost of uniformity. While high-income households would likely prefer a very generous plan, low-income households would likely prefer lower health care spending and higher take-home pay or more generous non-medical benefits such as food stamps or housing assistance. A uniform program pools everyone into the same plan, creating an inefficiency due to mismatch between the public benefit and privately optimal generosity.
Our central argument is that three macro trends have increased this cost of uniformity appreciably since Medicare’s creation in 1965. First, income inequality has risen substantially. Rising inequality leads to growing divergence between rich and poor in willingness (and ability) to pay for generous medical care. Second, there have been dramatic innovations in medical technology: there was much less health care available to buy in the 1960s, and even advanced technologies of the day were relatively inexpensive. Third, average marginal tax rates have increased from less than 25% in 1965 to 30% in 2012, commensurately increasing the deadweight loss (or economic cost) associated with publicly financed benefits – a trend that will likely continue with the budget pressures from population aging.
These changes imply that demand among the rich for generous medical care increasingly diverges from what a uniform public system can afford to fund. While a universal, generous Medicare program may have been efficient in 1965 when options for treatment were both limited and relatively inexpensive, tax rates were lower, and income more evenly distributed, the efficiency cost of maintaining uniform coverage has grown over time. The current benefit design thus may not be a sustainable foundation upon which to expand public health insurance.
We describe an alternative insurance benefit design in which the government provides basic insurance but allows higher-income households to “top up” by purchasing additional coverage for additional services. (Medicare does have in place supplemental “Medigap” plans, but these are primarily designed to cover copayments and deductibles, rather than cover additional services.) The basic plan is intended to be similar to public insurance provided in many other countries – with low patient copayments and deductibles but with more modest provider payment rates and with coverage of treatments restricted to those with proven effectiveness relative to lower-cost alternatives.
Supplemental “top-up” plans are also common in other countries. Governments often underwrite a basic insurance plan (or mandate the purchase of regulated and subsidized private plans), but then allow households to add on private supplemental insurance.
Our calibrated model suggests that switching from a uniform Medicare benefit to a top-up structure could generate substantial cost reductions and efficiency gains in the long term. The distributional implications of such a policy change would depend on the alternative uses to which the resources saved on public insurance would be devoted. Many European countries spend substantially more on other social insurance programs than the U.S., and some of those non-medical programs themselves are likely to yield health benefits. We show in the model that there exists a redistribution of the “Medicare dividend” that would raise wellbeing across all income groups.
While our model considers benefit design solely for people age 65 and over, the implications for Medicare benefit design are clearly amplified under proposals to expand the eligible population. For example, while the cost of “Medicare for All” proposals depends crucially on the details of eligibility, coverage, and provider payment rates, most proposals require additional tax revenues that would substantially raise marginal tax rates. The implication of our simple model is that a more basic public benefit – closer to “Medicaid for All” than to “Medicare for All” – with the option for individuals to top up to more generous private coverage, coupled with increased transfers to the poor, could prove to be a higher-value, more sustainable alternative to many proposals that seek to expand the current Medicare program.
From the Discussion and Conclusion
Means-tested in-kind transfers of housing, food, and health care are the predominant form of income redistribution to low-income households. Medicare is a prominent example of a uniform in-kind benefit provided to both high- and low-income populations. In this paper, we develop a model that allows us to gauge the tradeoffs involved in this uniform benefit design.
Using a stylized model, we show how tax distortions, income inequality, egalitarian preferences, and technology growth affect the efficient structure of the program. Our results suggest that in 1965 when Medicare was first created, its uniform generous structure was relatively well suited to the economic and technological environment. But by 2019 it has become much less efficient relative to a “top-up” health insurance program where more basic public coverage can be supplemented by private health insurance. Our results are consistent with policies seen in many other developed countries, which provide a basic universal public insurance plan and where many citizens take advantage of the opportunity to pay extra for amenities and additional services.
We explore the implications of an alternative “basic” form of public insurance that provides more restricted benefits with regulated prices and allows higher income households to top-up their coverage with privately financed plans. Under such a plan, lower income households would consume less health care than their higher income counterparts, and perhaps less care than they do now. This naturally raises concerns about equity. Many who support a uniform benefit structure point to the “right to health care” as a foundational rationale. It is worth noting, however, that a uniform benefit is likely to result in a substantially higher share of income being devoted to health care (rather than, for example, food, education, or housing) than the typical lower-income household might choose. It is also likely to result in fiscal pressures over time that not only raise taxes but also crowd out spending on other public goods and transfer programs. We demonstrate in our model that it is possible to offset some of the equity effects of the top-up redesign by appropriate redistribution of the taxpayer savings generated by scaling back the public benefit.
This top-up structure is not only similar to that seen in some other countries, but is also related to other proposals discussed in the U.S. context, including value-based insurance design and premium support plans.
Moving to a basic benefit-plus-top-up plan would of course pose both practical and political challenges because of the reliance on potentially controversial determinations of cost-effectiveness. Political pressures might, though, play out differently with different plans and different eligible populations.
As new technologies arrive with ever-larger price-tags, pressure will continue to mount on public budgets; and equality of access to care, rather than guaranteed access to a minimum level of care, will become increasingly costly. It is vital that policymakers consider how alternative program designs affect the overall wellbeing of households across the income distribution as they debate Medicare’s future.
Comment:
By Don McCanne, M.D.
This working paper is important because it comes from noted academics in the health policy community. The theme is not original. In a somewhat arcane economic disposition they rationalize a two-tiered or multitiered system of health care financing:
“We explore the implications of an alternative ‘basic’ form of public insurance that provides more restricted benefits with regulated prices and allows higher income households to top-up their coverage with privately financed plans. Under such a plan, lower income households would consume less health care than their higher income counterparts, and perhaps less care than they do now.”
In one of my earliest ventures into health policy, in response to the failure of the Clinton reform effort, I composed a health care proposal advocating for a basic plan for everyone and supplemental coverage for those who could afford it. I sent it in to PNHP, and Claudia Fegan was kind enough to return it, saturated in red ink. I learned two lessons. I realized how poorly informed I was on the topic of health policy, and it motivated me to study the topic more intensively, which I continue to do so to this day. More importantly, I developed a more acute sense of health care justice – that we can easily devise a health care system that ensures that everyone receives the health care they should have.
So why do so many in the policy community believe that lower income households should be restricted to only basic health care benefits while allowing higher-income households to top-up their benefits? I did further research to try to answer this question, and I found an article by the lead author of this working paper, Mark Shepard, titled, “Surprised by Jesus: how he found me at Harvard.” But the historical Jesus was a prophet of social change, which leads to the age-old question: What would Jesus have done?
Stay informed! Visit www.pnhp.org/qotd to sign up for daily email updates.
