Daring to be cautious?: Bigger steps needed for Medicare cost-sharing reform

Doctor and patient by Shutterstock.com

Article Highlights

  • Medicare cost-sharing reform will reduce program spending.

    Tweet This

  • Major-risk approach to Medicare cost-saving reform will yield most results.

    Tweet This

  • Adjust subsidies, not taxes, for Medigap coverage.

    Tweet This

Subscribe to AEI's health emails
Articles and events on health policy. Published approximately twice a month.

First Name:
Last Name:
Zip Code:

Finding economically feasible and politically safe options for cutting costs in Medicare has proven to be no easy task. A variety of government organizations, including the Congressional Budget Office and Medicare Payment Advisory Commission, and health policy researchers, such as MIT economist Jonathan Gruber, have put forth plans to improve the program, but they often diminish potential savings by playing it too safe. The best option for sustainable reform appears to be a major-risk approach toward restructuring cost-sharing requirements. This involves a higher coinsurance rate and an income-related stop-loss cap on participants’ annual cost-sharing liabilities. An additional key to subsidizing high-income seniors less is by restricting their use of supplemental insurance such as Medigap for early-dollar spending, rather than taxing the coverage itself.

Key points in this Outlook:

  • Reforming Medicare cost-sharing requirements is an important option for reducing program spending levels.
  • A major-risk approach to cost sharing, with higher coinsurance and annual caps tied to income level, could provide the fairest and most effective avenue toward the best results.
  • Taxes on Medigap coverage just complicate the tax code more without much precision in retargeting Medicare spending incentives; instead, higher-income seniors should be subsidized less and lower-income seniors subsidized more.


When health policy analysts join forces with budget-deficit hawks to search for remaining targets of belt-tightening in the Medicare program, they usually find that the policy reform shelves are relatively bare of politically “safe” options that can deliver early and significant cost savings.

Most premium support proposals generally would delay their implementation for at least a decade, and they stop short of seriously threatening the longevity of the traditional Medicare fee-for-service program. Although the Affordable Care Act (ACA) relies heavily on annual, across-the-board reimbursement cuts for health care providers, the sustainability of formulas that would eventually drive some Medicare payments down to below-cost Medicaid levels remains dubious. Raising the eligibility age for Medicare benefits in the near term also seems ahead of its time.

Repeatable and scalable results from a host of fiscal science-fair demonstration projects in the ACA are always somewhere over the budget window’s horizon. And the ability of pioneering accountable care organizations to breed in the regulatory captivity of Obamacare will require heroic assumptions about either asexual reproduction or politically assisted artificial insemination.

However, restructuring the splintered cost-sharing requirements of the traditional fee-for-service Medicare program’s separate silos for parts A and B (if not part D)[1] provides another potential policy reform tool that could achieve the twin goals of saving taxpayer dollars while improving risk-protection benefits for elderly beneficiaries. By increasing Medicare enrollees’ cost consciousness regarding more discretionary, first-dollar health care choices,[2] a coordinated set of changes in the traditional program’s deductible and coinsurance provisions could help reduce current and future levels of Medicare spending. At least some of those savings from greater sharing of health care costs between Medicare insurance benefits and enrollee’s out-of-pocket payments at the front end could be used to provide better “stop-loss” protection against larger catastrophic risks.

Past Proposals to Restructure Medicare Cost Sharing

Such reform of Medicare cost sharing is far from virgin territory in health policy circles. The historical legacy of Medicare’s original division between hospital-based care and other outpatient care (with two separate trust funds and sources of financing) has produced substantial initial cost sharing for the first day of hospital care in the form of a high deductible ($1,184 in 2013) in Part A, but a much lower deductible ($147 in 2013) for outpatient services spending in Part B.

However, coinsurance cost sharing of 20 percent for each additional dollar of Part B outpatient care above that program’s initial deductible amount is not capped, and it could potentially amount to even greater out-of-pocket liabilities for beneficiaries. For example, 6 percent of Medicare fee-for-service beneficiaries who enrolled in Part A and Part B for 12 months in 2009 had a cost-sharing liability of $5,000 or more (assuming no additional insurance coverage, such as an individual Medigap plan, employer-sponsored retiree coverage, or Medicaid supplemental benefits for low-income retirees).[3] Moreover, the probability of catastrophic health spending over time is higher than the probability within a single year would indicate.[4]

Hence, a majority of the members of the National Bipartisan Commission on the Future of Medicare in 1999 recommended that the separate deductibles for Part A and Part B be replaced with a single deductible of $400, which then would be indexed to growth in Medicare costs.[5]

In March 2011, the Congressional Budget Office (CBO) analyzed the effects of replacing Medicare’s mix of cost-sharing requirements with a single combined deductible of $550 (covering all Part A and Part B services), a uniform coinsurance rate of 20 percent for amounts above that deductible, and an annual cap of $5,500 on each enrollee’s total cost-sharing liabilities. CBO estimated that if this option took effect on January 1, 2013, with the various thresholds indexed to growth in per capita Medicare costs in later years, federal spending for Medicare would fall by about $32 billion over the 2012–21 period.[6]

Limiting the Spillover Spending Effects of Supplemental Coverage

If Medigap plans—private plans designed to supplement basic Medicare coverage and sold to individuals—were barred from paying the first $550 of an enrollee’s cost-sharing liabilities for calendar year 2013 and could cover only half of other Medicare cost sharing (equivalent to changing Part B’s 20 percent coinsurance rate to only 10 percent) up to the annual $5,500 cap on total cost sharing (and if the various thresholds were indexed to growth in per capita Medicare costs for later years), CBO estimated that, under this reform option, projected federal outlays would be reduced by roughly $93 billion over the 2012–21 period.[7]

Last June, the Medicare Payment Advisory Commission (MedPAC) proposed a slightly different approach to reforming Medicare’s benefit design. It recommended an annual deductible for Part A and Part B services equaling $500 (while leaving open whether it would be combined or separate for those categories).

However, MedPAC suggested that copayments (fixed dollar amounts), rather than coinsurance (a percentage of costs), should apply for cost sharing above the deductible amount and until a total annual, out-of-pocket $5,000 maximum is reached.[8] It further complicated and diluted the effects of this restructuring by insisting that beneficiaries’ cost-sharing liabilities in the aggregate should not increase in the redesign of the traditional Medicare fee-for-service program. MedPAC also proposed that copayment amounts may vary by type of service and provider, with the secretary of health and human services altering or eliminating cost sharing based on evidence of the value of particular services.[9]

MedPAC took a great leap of faith that the likely evidence base will be sufficiently robust for the secretary to tailoring cost sharing in an accurate and consistent manner. The Medicare advisory body then urged a different “tax” approach to discourage, or at least recoup, some of the added costs imposed on the basic Medicare program when supplemental Medigap coverage encourages greater spending. Instead of barring Medigap insurers from paying any of the initial costs falling within the unified deductible and limiting reductions in coinsurance liabilities above that amount to no more than half the standard 20 percent rate for Part B spending, MedPAC would allow beneficiaries to add costs to basic Medicare coverage through additional Medigap insurance but then charge them for this privilege with an extra 20 percent “excise tax” on the value of that supplemental coverage.[10]

The actual effects of this change in the treatment of supplemental Medicare coverage[11] would vary depending on the degree to which beneficiaries choose to retain their additional coverage and pay the tax (producing new revenue to offset some of the higher Medicare spending), as opposed to dropping that coverage and creating more budget savings through lower Medicare spending.

The Gruber Plans: Now and Then

In February, MIT economist Jonathan Gruber offered his latest version of “Restructuring Cost Sharing and Supplemental Insurance for Medicare” as part of the Brookings Institution’s 15 Ways to Rethink the Federal Budget. He expressed concern with the affordability of revisions to cost sharing among seniors under some of the previous reform proposals, as well how proposed stringent regulation on supplemental Medicare plans would not allow the plans to “reflect diversity of elders’ tastes for supplemental coverage.”[12]

Gruber proposed an alternative to the previous CBO options (his paper references an older 2008 CBO analysis, rather than the more recent one in March 2011), based instead on an “income-related” out-of-pocket maximum. However, Gruber divides those stop-loss limits into just four income categories and then sets their maximums as respective fractions (1⁄3, 1⁄2, 2⁄3, and all) of the current-law health savings account stop-loss limit ($5,950) that also is used under the ACA’s rules for qualified insurance coverage. He also recommends reducing the unified deductible by half (to $250) for seniors below 200 percent of the federal poverty level.[13]

Gruber concedes that computing the amounts of such out-of-pocket protections will be administratively difficult and could raise privacy concerns if private insurers must know the incomes of individual Medicare beneficiaries. The greater irony is that Gruber admits that his plan to protect low-income seniors by lowering their income-related out-of-pocket maximums “by itself is unlikely to produce any budget savings.”[14]

But apparently, one more tax is always just around the corner to retrieve those dollars for the federal Treasury. Gruber proposes a tax on supplemental Medicare coverage—even higher than the one envisioned by MedPAC—to offset the higher Medicare spending supplemental coverage causes. Subject to political negotiations, he estimates that a tax rate of up to 45 percent on Medigap plan premiums and on the cost of employer-sponsored retiree coverage would be justified.[15]

"The major-risk approach to Medicare makes the most sense for most beneficiaries."Gruber concludes that the budgetary implications of this proposal are “difficult to infer.”[16] Actually, his proposal merely seems to reverse the order of usual liberal fiscal policy mantras. It is “spend and tax” rather than “tax and spend” but probably ends up in about the same place.

For a more effective and sustainable, but less equivocal, approach to slowing the growth of excessive Medicare spending while protecting beneficiaries against much costlier risks they cannot bear on their own, we could instead turn to a more astute health policy expert—Jonathan Gruber, version 1.0!

In a 2010 American Economic Review study with coauthors Amitabh Chandra and Robin McKnight,[17] Gruber recommended that increased cost sharing should be tied to a patient’s underlying health status (rather than just income), with chronically ill patients facing lower cost sharing. The authors found that higher copayments for office visits and prescription drugs reduced Medicare medical spending, with elasticities of demand similar to those reported in the RAND Health Insurance Experiment for the nonelderly.[18]

However, Chandra, McKnight, and Gruber also noted a significant offsetting rise in use of hospital care and overall hospital-based spending (reducing net budget savings from higher copayments on other services by about 20 percent) because of the higher copayments for outpatient care and prescription drugs. Moreover, they found large offsets for the sickest Medicare populations with chronic diseases, suggesting that higher copayments for that cohort of beneficiaries produced little net budgetary savings for the Medicare program.[19]

Hence, Gruber and his coauthors concluded that because the “mirror effect” of this relationship suggests that an increase in physician and drug spending arising from supplemental Medicare coverage is substantially offset (within the traditional Medicare program) by the fall in hospital costs, income-related out-of-pocket limits alone provide far-from-optimal health insurance (and Medicare cost savings). They recommend further specific targeting of copayments related to the underlying health status of chronically ill patients.[20] Apparently, those research findings did not make their way into Gruber’s purely income-based cost-sharing recommendations for the Brookings 15 Ways to Rethink the Federal Budget project.

Simpler and More Effective Cost-Sharing Reform

In any case, there is a better way to handle income-related limits on more unified Medicare cost sharing, again courtesy of a younger but wiser Jonathan Gruber. In 1994, Gruber and coauthor Martin Feldstein proposed “A Major Risk Approach to Health Insurance Reform.”[21] To reduce the economic dead-weight loss produced when low coinsurance rates (and low marginal costs of insured care) induce excessive consumption of health care and inefficient resource allocation, Gruber and Feldstein modeled a different type of health insurance plan. It would have a 50 percent coinsurance rate but limit out-of-pocket health spending to 10 percent of income.

They estimated that aggregate welfare gains (which also include reduced risk bearing for large health care costs) by switching to major-risk insurance for both private and public (Medicare and Medicaid) health coverage would range from $34 billion to $110 billion—in 1995 dollars.[22] Those estimates varied depending on the degree of risk aversion and price elasticity of demand, respectively, by health care consumers. For example, a higher degree of risk aversion and higher demand elasticity would produce a larger welfare gain.

An attractive dimension of the major-risk approach is its relative progressivity. Average out-of-pocket spending under the plan rises sharply as income rises because the stop-loss maximum rises with income. The major-risk plan reduces the total consumption of health care much more for high-income individuals. It also alters the risk distribution individuals face by increasing the risk of modest spending but limiting the maximum risk.

Gruber and Feldstein noted that relying on 50 percent coinsurance, rather than the equivalent amount of cost sharing (up to 10 percent of income) solely through a first-dollar deductible, extends the ability of the plan’s cost-sharing incentives to reduce dead-weight losses across a wider range of health spending. But it also limits the value of the increased risk to individuals through greater cost sharing. Of course, the various specific projections of welfare-gain effects in the 1994 study would need to be updated to align with current levels of health spending, and they remain sensitive to relative assumptions about the levels of demand elasticity and risk aversion for health care consumers.

Policy Priorities for Reform

The larger lesson from these analyses of Medicare cost-sharing restructuring involves the importance of setting clear policy priorities, avoiding trying to accomplish conflicting goals with the same policy instrument, and carrying out first what matters most.

Slowing the future rate of Medicare spending, or at least lowering its current projected baseline level, should be the primary reason for reforming the structure of cost sharing in the traditional Medicare program. And that means most beneficiaries (except for those provided separate special protection from this reform) will actually end up receiving lower levels of taxpayer subsidies and either pay more for the Medicare services they want or get fewer (or less expensive) services. The most appropriate area of Medicare spending to face those new cost-sharing incentives involves early-dollar, discretionary spending, rather than the costs facing beneficiaries with much more expensive or chronic medical conditions.

The major-risk approach to Medicare reform makes the most sense for most beneficiaries. It balances protecting them more effectively against catastrophic financial risks with increasing their cost consciousness for decisions involving health care costs they can manage better within the limits of their income. By relying on a higher percentage of coinsurance (rather than a large front-end deductible), this approach also produces the best mix of stop-loss protection and greater sensitivity to the noncatastrophic costs of covered services.

Trying to overcompensate and dilute the tension between most beneficiaries’ income constraints (including the opportunity cost of spending for other non-health-care wants and needs) and their initial layers of health care need will undermine the main purpose for taking on the difficult political challenge of increasing cost sharing for the traditional Medicare program.

The administrative complexities of income-related cost sharing can be managed through setting a reasonable range of annual income bands linked to proportionately related mixes of out-of-pocket maximum levels (rather than calculating them dollar for dollar at every level of reported income).[23] This should be supplemented with a narrow set of opportunities to appeal for exceptions based on unexpected hardship.

The cleaner and less complicated way to deal with the distortions of supplemental insurance for traditional Medicare enrollees is to set regulatory boundaries on what either individual Medigap plans or employer-sponsored retiree plans can cover. The current tax code already produces enough distortions in economic decision making without adding a new excise tax on supplemental insurance premiums to that list.[24] Instead of taxing a small slice of affluent seniors more and then recycling some of that revenue back to other lower-income seniors (with all the inefficient processing and extraction charges this political spin cycle entails), it would be far better simply to subsidize nonpoor seniors less.

We should acknowledge the limits of the relationship between personal income and the ability to handle much greater cost sharing for health care services but then develop a separate program of cash subsidies to lower-income Medicare beneficiaries facing more chronic health conditions, with such funding distributed directly to their (new) individual Medicare savings accounts based on a combination of their income and health risk scores. However, such subsidies should be limited in size and scope and then carefully targeted.

The distortions of supplemental insurance coverage primarily harm the traditional fee-for-service Medicare program, whether they originate from individual Medigap plans or employer-sponsored retiree coverage. These cost-sharing reforms are not needed for private plans in Medicare Advantage because the coverage scheme in those plans is integrated within a single insurer rather than spread across a taxpayer-financed primary insurer (traditional Medicare) and a secondary insurer.

"The cost-saving juice must be worth the political squeeze in undertaking reform of Medicare cost sharing."

Nevertheless, any remaining problems of mixed cost-sharing incentives and effects could be corrected in two ways: (1) require private insurers to offer only integrated coverage (current-law Medicare benefits plus any supplemental ones), with the separate price and taxpayer subsidies for the basic Medicare coverage, or its actuarial equivalent, determined through premium-support-style competitive bidding,[25] and (2) authorize the traditional Medicare program to offer enhanced catastrophic stop-loss protection, along with more flexibility in changing other cost-sharing provisions, as part of greater administrative flexibility needed to compete in such a bidding regime.[26]

Given the future budgetary stresses and broader sustainability challenges facing the Medicare program, this more aggressive approach to reform its cost sharing is long overdue. But taking two modest, but important, steps forward—a unified deductible and a stop-loss limit for traditional Medicare—will provide little progress if we then take two steps backward—diluting such cost sharing for most Medicare beneficiaries on the loose grounds that they may not be able to afford it or could refrain from seeking necessary health care services, and adding a new tax at a fill-in-the-blank rate on supplemental insurance plan coverage.

The 1984 Walter Mondale presidential campaign once was tagged with an uninspiring, but telling, slogan for such tactics: “Dares to be cautious.”[27]

The cost-saving juice must be worth the political squeeze in undertaking reform of Medicare cost sharing. Perhaps Jonathan Gruber will reacquaint himself and others with his older work and the ideas of his time-inconsistent younger self, which was less politically seasoned but more economically grounded and less irrationally addicted to income redistribution.[28] A major-risk approach to reform of cost-sharing and taxpayer subsidies for coverage offers more future reward, whether for just Medicare or also the rest of the private health insurance market (as Feldstein and Gruber originally proposed).


1. Part A for Medicare covers the costs of inpatient care (primarily hospital spending). Part B covers physicians’ services and other outpatient care. Part D covers prescription drug expenses. Each part of the traditional Medicare program has its own sources of financing (a payroll tax for Part A, separate enrollee premiums plus general revenue support for Parts B and D).
2. Such discretionary spending decisions may involve either purchases within the scope and scale of first-dollar deductibles or other purchases subject to cost sharing (either coinsurance or copayments) before any stop-loss limit on out-of-pocket spending is reached.
3. Medicare Payment Advisory Commission (MedPAC), Report to the Congress: Medicare and the Health Care Delivery System (Washington, DC, June 2012), 5, www.medpac.gov/documents/Jun12_EntireReport.pdf.
4. Ibid., 10.
5. National Bipartisan Commission on the Future of Medicare, Building a Better Medicare for Today and Tomorrow (Washington, DC, March 16, 1999), http://rs9.loc.gov/medicare/index.html.
6. Congressional Budget Office, Reducing the Deficit: Spending and Revenue Options (Washington, DC, March 10, 2011), 49, www.cbo.gov/publication/22043.
7. Ibid., 49–50.
8. MedPAC, Report to the Congress: Medicare and the Health Care Delivery System, 20.
9. Ibid., 19.
10. Ibid., 21.
11. MedPAC would also apply the same reform to employer-sponsored retiree coverage.
12. Jonathan Gruber, “Proposal 3: Restructuring Cost Sharing and Supplemental Insurance for Medicare,” in 15 Ways to Rethink the Federal Budget, ed. Michael Greenstone et al. (Washington, DC: Brookings Institution, February 2013), 25, www.hamiltonproject.org/papers/restructuring_cost_sharing_and_supplemental_insurance_for_medicare/.
13. Ibid., 25.
14. Ibid.
15. Ibid., 25–26.
16. Ibid., 26.
17. Amitabh Chandra, Jonathan Gruber, and Robin McKnight, “Patient Cost-Sharing and Hospitalization Offsets in the Elderly,” American Economic Review 100, no. 1 (2010): 193–213.
18. The RAND Health Insurance Experiment remains the most extensive and important experimental study of the effects of cost-sharing and the design of health insurance benefits. Its results demonstrated that higher patient payments (cost sharing) significantly reduced use of medical care, without any adverse health outcomes for the average person. See, for example, Joseph Newhouse, Free for All: Lessons from the RAND Health Insurance Experiment (Cambridge, MA: Harvard University Press, 1993).
19. Chandra, Gruber, and McKnight, “Patient Cost-Sharing and Hospitalization Offsets in the Elderly,” 209–11.
20. Ibid., 212.
21. Martin Feldstein and Jonathan Gruber, “A Major Risk Approach to Health Insurance Reform” (working paper 4852, National Bureau of Economic Research, Cambridge, MA, September 1994).
22. Ibid., 26. A full and accurate update, in terms of current (2013) dollars is not readily available, because it involves more variables than simply the effects of health care spending inflation over time.
23. This involves balancing the unworkability of administering customized “personal” cost-sharing limits for literally millions of Medicare beneficiaries against dilution of relatively income-sensitive variations in cost sharing, as well as treating seniors with significantly different incomes too similarly. The issue comes down to how wide a dollar income range needs to be maintained between changes in stop-loss limits for different cohorts of Medicare seniors. Those cost-sharing brackets certainly should be more narrow than those used for the federal income tax’s rate brackets, but at least wider than every $10,000 in additional reported income. Congress has a good deal of experience in making similar judgments for the phase-out of income-related tax subsidies or phase-in of higher income-related taxes.
24. See, for example, Tom Miller, “Rethinking the ‘Other’ Payroll Tax,” Real Clear Markets, January 11, 2012, www.realclearmarkets.com/articles/2012/01/11/rethinking_the_other_payroll_tax_99454.html.
25. See, for example, National Bipartisan Commission on the Future of Medicare, Building a Better Medicare.
26. See ibid.; Thomas Miller, When Obamacare Fails: The Playbook for Market-Based Reform (Washington, DC: American Enterprise Institute, December 2012), 20, www.aei.org/papers/health/healthcare-reform/ppaca/when-obamacare-fails-the-playbook-for-market-based-reform/.
27. Mondale’s media director, Roy Spence, infamously praised his boss because he “dares to be cautious.” Richard Cohen, “Jackson Shows Mondale’s Caution,” Boca Raton News, January 6, 1984, www.news.google.com/newspapers?nid=1291&dat=19840106&id=5i5UAAAAIBAJ&sjid=-lwDAAAAIBAJ&pg=6804,1273090.
28. See, for example, Jonathan Gruber and Botond Koszegi, “Is Addiction ‘Rational’? Theory and Evidence” (working paper 7507, National Bureau of Economic Research, Cambridge, MA, January 2000).

Also Visit
AEIdeas Blog The American Magazine
About the Author


Thomas P.

What's new on AEI

AEI Election Watch 2014: What will happen and why it matters
image A nation divided by marriage
image Teaching reform
image Socialist party pushing $20 minimum wage defends $13-an-hour job listing
AEI on Facebook
Events Calendar
  • 20
  • 21
  • 22
  • 23
  • 24
Monday, October 20, 2014 | 2:00 p.m. – 3:30 p.m.
Warfare beneath the waves: The undersea domain in Asia

We welcome you to join us for a panel discussion of the undersea military competition occurring in Asia and what it means for the United States and its allies.

Tuesday, October 21, 2014 | 8:30 a.m. – 10:00 a.m.
AEI Election Watch 2014: What will happen and why it matters

AEI’s Election Watch is back! Please join us for two sessions of the longest-running election program in Washington, DC. 

Wednesday, October 22, 2014 | 1:00 p.m. – 2:30 p.m.
What now for the Common Core?

We welcome you to join us at AEI for a discussion of what’s next for the Common Core.

Event Registration is Closed
Thursday, October 23, 2014 | 10:00 a.m. – 11:00 a.m.
Brazil’s presidential election: Real challenges, real choices

Please join AEI for a discussion examining each candidate’s platform and prospects for victory and the impact that a possible shift toward free-market policies in Brazil might have on South America as a whole.

Event Registration is Closed
No events scheduled this day.
No events scheduled this day.
No events scheduled this day.
No events scheduled this day.