Medicare Part D’s competitive bidding structure should serve as a model for future reforms
Medicare Part D cost under half what was originally projected, accounting for just 48% of original estimates
December 8 marks the 10 th anniversary of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (MMA), which established the Part D program and made available coverage for modern pharmaceutical therapies for all Medicare beneficiaries for the first time. The success of Part D over the past decade makes it a model for future entitlement reforms.
At the center of the Part D policy design is a reliance on market forces and competition. This approach raised three key questions: Would insurers and beneficiaries participate at high enough rates to create plan choice for beneficiaries? Would there be sufficient competition among insurers to result in affordability for beneficiaries? What would the long-term fiscal impact be on the federal budget?
Insurer participation proved to not be a concern. Indeed, the MMA provided for federally run “fallback” drug plans to be implemented if private sector plan sponsors did not materialize. No fallback plan has ever been needed.
Beneficiaries display great satisfaction with Part D and in 2013 have at least 23 different plans to choose from in every region. Moreover, the system in which plans bid against each other for beneficiaries and negotiate directly with drug manufacturers to achieve lower pricing has generated affordable premiums.
A second reflection of competitive pressures has been the downward trajectory of federal budget costs, in which actual costs for 2012 fell below every projection of those costs over the past decade.
The Medicare Prescription Drug Benefit has been an unmitigated success. A market-based approach to further federal entitlement program offers additional opportunities for controlling taxpayer costs and generating high levels of beneficiary satisfaction.
Medicare Part D is one of the few government programs that has actually cost far less than projected. In addition, its operations are directed in part by market forces, in which numerous privately operated Part D plans compete for the enrollment of millions of individual beneficiaries, on the basis of plan features and cost.
One might reasonably conclude that these two facts are connected, especially in light of the experience in other sectors of the economy, where competition is generally associated with lower costs, greater efficiency, and higher levels of consumer satisfaction. However, some observers have resisted this conclusion, attempting to attribute the unprecedented cost under-run to other factors, such as reduced cost drivers in the overall prescription drug market, such as the “patent cliff,” and have characterized the competition among plans as a source of “confusion” to beneficiaries, despite contrary evidence from available data.
In fact, the evidence shows that competition is a significant factor contributing to the success of Part D in reducing program expenditures and contributing to slower growth in the overall prescription drug market. Furthermore, there is evidence that increases in prescription drug utilization among Part D enrollees not only improves their health, but also reduces medical and hospital expenditures for that population. However, the lower level of spending does not come at the cost of beneficiary satisfaction. Survey data indicates that beneficiary satisfaction is at a very high level.
This paper will explore the history of the Medicare program as it relates to prescription drug coverage, the enactment of Medicare Part D and experience with the program ten years later, and how the competitive features of Part D have played a key role in the program’s success.
Background on Medicare and Prescription Drug Coverage
When Medicare was implemented in 1966 it included two main components, “Part A” for hospital costs (funded by a payroll tax), and “Part B” for physician services (initially voluntary, with a subsidized premium). While Part A covered drugs administered in an inpatient setting, and Part B included drugs administered by physicians in their offices, there was no coverage for prescription drugs obtained at retail pharmacies and self-administered by patients. This reflected the circumstances of the era, as outpatient drugs played a less prominent role in health care than they do now. Most private health insurance for the under-65 population, including employer-sponsored insurance, did not cover outpatient prescription drugs. Furthermore, substantially fewer treatments for chronic conditions using long-term maintenance medications were available than is the case now, and some conditions now treated with medications were at the time treated with surgery or other non-pharmaceutical treatments.
Over time, new pharmaceuticals were developed, and prescription drugs became a more significant component of medical practice, especially in the treatment of chronic conditions. These factors led to increased demand for prescription drug coverage in health plans. By the mid-1980s most private health insurance plans, including employer-sponsored plans, provided coverage for prescription drugs.
It took about two decades for Medicare to “catch up” with this change in medical practice. In the interim, Medicare beneficiaries were forced to obtain prescription coverage through other means or had no coverage at all. Until Part D was implemented, there was no standard outpatient prescription drug coverage that was part of the Medicare program and available to all beneficiaries.
Enactment of Medicare Part D
Eventually, a broad consensus developed – with the increased use of pharmaceutical treatment for both acute and chronic conditions, the absence of drug coverage from Medicare represented a significant shortcoming in the program. The result was the establishment of Medicare Part D as part of the MMA. Those who objected to the creation of Part D did so largely on the grounds of cost to taxpayers in an era of high government spending, or the policy architecture of the program, rather than on the merits of drug coverage as such.
Part D was designed such that private plans would offer drug coverage to Medicare beneficiaries subject to minimum benefit requirements while still allowing substantial flexibility in terms of cost-sharing structure and the ability to offer enhanced features. Plan sponsors would submit bids to CMS each year based on their expected costs for providing the benefit, and then a national average of submitted bids would be used to determine the amount of the government subsidy – a set percentage of the national average bid – as well as the monthly premium paid by beneficiaries. Beneficiaries were offered a choice among plans, which allowed plans to compete for enrollment based on benefit offerings and premiums.
Nothing quite like this had been tried in any federal program before, so there was understandably some concern as to whether it would work at all. Because the subsidy amount would be determined by bids, subsidy expenditures were not known with certainty at the time of enactment. In addition, there was no guarantee that private sector companies would step forward to offer plans, let alone that enough would do so to generate competition in all regions of the country. Furthermore, there was no guarantee that enough beneficiaries would enroll to make the program viable, especially given that some beneficiaries were enrolled in other sources of coverage, and others had low enough drug needs to feel comfortable going without coverage.
Experience with Part D
To address concerns about plan participation, the MMA provided for a federally-run “fallback” drug plan, to be implemented if an inadequate number of private sector plan sponsors materialized. The fallback plan could have been implemented for the entire country, or just for underserved regions. However, as there turned out to be not a single underserved area, the fallback plan was never needed. In fact, as of 2013, every beneficiary has the choice of at least 23 different plans, from multiple plan sponsors, with more options available in many regions.
There is also evidence that competition among many plans helps to reduce premiums. Regional data from 2007 to 2010 demonstrates that having more plan sponsors in a region is associated with lower premiums, and the effect of adding an additional sponsor is stronger when there are fewer plan sponsors to begin with in a region.
Concerns about beneficiary participation also failed to materialize. By the end of 2006, the first year of the program, 63 percent of beneficiaries who previously did not have drug coverage enrolled in Part D. As of 2012, almost 65 percent of Medicare beneficiaries were enrolled in Part D plans. Overall, 90 percent of Medicare beneficiaries are enrolled in Part D or some form of creditable coverage that is at least as generous as Part D’s standard benefit.
The ultimate cost of the new Part D benefit was also a fear. However, total program expenditures have come in far lower than initially projected. Part D's 10-year cost (starting in 2006) was projected to be $957.3 billion in 2004, after the MMA was passed but before the program started. By 2011, the combination of five years of actual data and five years of projections totaled $499.4 billion, for a cost under-run of $457.9 billion, or about 48 percent. The latest year for which the actual spending for Part D is available is 2012. The last CBO forecast for 2012 Part D spending made prior to implementation was in 2005, and the projected 2012 spending in that year was $126.8 billion. After the bids came in for 2006, the 2012 forecast was reduced to $110.2 billion. In all but one of the next six years, the forecast for 2012 was reduced further, with the final forecast being only $60.1 billion. The actual amount was $55.0 billion – about 57 percent lower than the original pre-implementation forecast. (See Figure 1 for how the cost estimate evolved over time.)
Even more, the Congressional Budget Office reduced its forward-looking 10-year projection for the Part D program by over $100 billion in each of the past three years. That is, even as the program gained experience over nearly eight years of operation, Part D continues to outperform spending projections.
It is often assumed that less spending means a lower quality program with fewer benefits and reduced overall satisfaction. However, as all available evidence indicates, Part D beneficiary satisfaction has been and continues to be consistently high. In a 2007 survey, 85 percent of seniors in Part D said they were “satisfied” with their coverage, with almost 60 percent saying they were “very” or “extremely” satisfied. Despite claims that beneficiaries would be “confused” by a wide variety of plan options, almost 80 percent of Part D enrollees felt they made a “good choice” when choosing a plan for 2007. 
More recently, MedPAC’s analysis of the 2010 Medicare Current Beneficiary Survey found that 94 percent of enrollees are satisfied with the Part D drug benefit and 95 percent were confident that their drug coverage meets their needs. Private sector survey data also indicates that in every year since 2007, a majority – 52 percent to 60 percent – of Part D beneficiaries were “very satisfied,” and 84 percent to 90 percent were either “very” or “somewhat satisfied” with their prescription drug coverage.
Finally, there is a growing body of evidence that shows increased use of prescription medicines reduces the use of other health care services, producing offsetting savings. For example, a study from Harvard researchers found that the implementation of Part D was associated with a $1200 average reduction in non-drug medical spending for Medicare beneficiaries who had limited prior drug coverage in each of the first two years of the program (2006 and 2007).  This achieved approximately $13.4 billion in overall savings during the first full year of the Part D program (2007). More recent studies have demonstrated the potential for further savings by improving adherence to prescribed medications, especially in managing chronic conditions like congestive heart failure. The Congressional Budget Office recently acknowledged this substantial body of evidence and changed their Medicare scoring methodology to account for the fact that increased prescription drug use in Part D leads to reductions in spending in Parts A and B.
Key Features of Part D’s Competitive Structure
Why has the Part D program been so widely successful? Several features of the program’s competitive structure have played a critical role, including the bidding process, beneficiary choice, robust private negotiations between Part D plans and drug manufacturers, and incentives that drive generic utilization.
Part D operates in a way that is different from other components of Medicare. In particular, Parts A and B’s prices are completely noncompetitive, Part C (Medicare Advantage) has competitive and noncompetitive components that figure into the pricing, and Part D pricing is almost completely competitive, even though the federal government provides a subsidy to cover a portion of the cost of the benefit.
Under Part D, plans submit bids for providing the basic prescription drug benefit to an “average” beneficiary. CMS calculates a “national average monthly bid,” which is a weighted average (based on enrollment) of all qualified bids. CMS then pays a set percentage (74.5 percent) of the national average monthly bid to plans on behalf of each enrollee, with adjustments for the relative health of each plan’s enrollees. CMS also provides reinsurance for cases in which a particular beneficiary's costs exceed a catastrophic threshold.
The remaining 25.5 percent represents the base monthly premium that is paid by beneficiaries. If a beneficiary enrolls in a plan that submitted a bid higher than the national average, the beneficiary pays the difference, in addition to the base premium. A beneficiary may also pay a higher premium if their selected plan offers enhanced coverage beyond the standard benefit (e.g. coverage in the “donut hole”). Part D beneficiaries who qualify for the full low-income subsidy (LIS) program pay no premium if they are enrolled in a plan that submitted a bid at or below the national average.
It is noteworthy that in the case of Part D, the CMS payment is derived from actual plan bids, rather than by administratively determined county rates derived from a legislative formula, as is the case for Medicare Advantage. In other words, Part D plans determine the CMS subsidy amount in a manner similar to that by which a competitive market determines the price of a good or service.
In addition, once submitted, the bids are binding on the plans if they choose to participate in the program that year. A plan cannot submit a high bid to increase the subsidy, and then sell at a lower price to attract more enrollees. Any difference between a plan's bid and the national average monthly bid translates directly into a price difference paid by beneficiaries. Therefore, plans have an incentive to submit lower bids so as to attract more enrollees. In fact, plans that submit bids below the benchmark in their bidding region are eligible to automatically enroll LIS beneficiaries. This incentive allows plans to increase enrollment volume without incurring the administrative costs of advertising and recruitment. Overall, this incentive has the effect of lowering the amount CMS pays for the subsidized portion of the premium for all beneficiaries.
Beneficiary choice among Part D plans further reinforces competition in the program. Beneficiaries have an incentive to select either a low premium plan, or a plan whose higher premium might be justified by additional benefits, a more preferable formulary, convenient pharmacy locations, better customer service, and/or lower cost-sharing.
Because beneficiaries pay the difference between the base premium and their selected plan’s premium, one can expect that over time those beneficiaries who are particularly price-sensitive and less interested in additional benefits will gravitate toward lower-bid plans. This is borne out in a recent analysis that found 60 percent of Part D beneficiaries choose a plan whose premium is within $6 (per month) of the lowest-premium plan available to them, and many change plans when their plan's premium increases. In fact, recent evidence shows that twice as many Part D enrollees are switching plans in recent years compared to the early years of the program. Thus, plans that bid too high to justify their benefits will find it more difficult to attract enrollees, and will instead be motivated to decrease their bids in future years. These incentives have the effect of further lowering the CMS payment over time, by increasing the weight of lower-bid plans in calculating the national average.
A broad spectrum of choices also means that beneficiaries can select a plan that meets their health needs and budget. For example, consider the case of many brand-name drugs in the same therapeutic class for a chronic condition, with no generic available. If one patient finds that Drug A works best for her, and another finds that Drug B works best for him, and there are multiple competing plans with different formularies, then each can choose a plan in which his or her preferred drug is covered. This means both patients would experience lower costs that could not be obtained if they were forced into the same plan. Generalizing this to patients with multiple maintenance medications and preferences in each therapeutic class, it is clear that choices are necessary to make sure beneficiaries get the coverage they need at the lowest cost available.
Some have argued that since not all beneficiaries choose the lowest-premium plan, or change plans whenever their premium increases, we must conclude either competition does not work or beneficiaries are incapable of making rational choices. These conclusions are based on a misunderstanding of the competitive process and discount the fact that beneficiaries must make choices each year in the presence of uncertainty about future drug needs and based on various degrees of risk aversion.
In a competitive market, premiums do not simply “appear” or “change” on their own. Premiums are set by bids that are determined by Part D plans in response to market conditions. As mentioned above, if some beneficiaries are choosing a higher premium plan, they are likely doing so because of its enhanced benefits; the plan may also have better “intangibles” like customer service or convenient pharmacies. In such a case, it is entirely consistent with the idea of competition for a well-run plan to attract even more enrollees as word gets out, particularly if it further enhances benefits – even if the premium increases. For example, one “Preferred” plan that increased its premium by 15 percent found that its enrollment increased by 2 percent. In a market with differentiated products, this observation does not disprove the existence of competition in Part D plans.
Robust Negotiation between Plans and Manufacturers
Some observers incorrectly believe that currently, there is no drug price negotiation that takes place in Part D, but this is a misunderstanding of how the program operates. In order to compete with one another, Part D plans employ a variety of strategies to reduce the cost of providing drug coverage to their enrollees, and chief among them is robust negotiation with drug manufacturers for discounts and rebates on the medicines used by their enrollees. In turn, these savings are passed on to beneficiaries in the form of lower premiums and cost-sharing. Available evidence demonstrates that these private negotiations have resulted in lower prices (i.e. higher rebates) from manufacturers than was originally predicted. Furthermore, CBO has found that Part D plans have negotiated rebates that are slightly larger, on average, than the rebates that are negotiated in commercial health plans, and that “rebates negotiated by Part D plans on preferred brands appear to make the net prices approach the lowest prices obtained in the private sector.”
Program Incentives Drive Generic Utilization
In order to control costs and offer lower premiums, plans also have strong incentives to encourage their enrollees to use generic medicines when appropriate. Plans accomplish this by using tiered formularies which offer generics to patients at a lower cost-sharing level relative to brand medicines. These are basically the same tools available to private-sector, non-Medicare health plans covering prescription drugs.
Available evidence demonstrates the success of Part D plans in driving generic utilization. The portion of generic drugs that are dispensed in the Part D program has increased by 20 percentage points since the program’s first year such that now, about four out of five prescriptions dispensed in the program are for a generic. The use of generics in Part D is expected to continue growing in the future. 
Some observers have noted that a significant portion of the cost under-run can be attributed to higher levels of generic substitution than was initially predicted, and concluded that therefore competition is not as significant a factor. However, this generic substitution is a direct result of competition between Part D plans, just as in the private market it is the result of competition between private insurers. In this way, Part D leverages competitive forces that already exist in the broader marketplace. As explained above, competition between Part D plans provides incentives for the plans to reduce costs so they can compete through lower premiums, and one way to reduce those costs is to encourage patients to use generics when appropriate.
In fact, Part D has a large enough share of the prescription market that generic substitution and other cost reductions in Part D are a significant driver in constraining growth in overall drug spending growth at the national level.
The Medicare Part D program is one of the few government programs that have achieved significant savings relative to its projected costs. It is also one of the few government programs in which competition between providers and beneficiaries is a significant factor in the program design. As a result, Part D harnesses competitive incentives to reduce costs, saving money for both taxpayers and Medicare beneficiaries. These savings have not come at the cost of program quality, as the overwhelming majority of enrollees report year after year that they are satisfied with their Part D plans. Rather than dismissing the role of competition in Part D’s success, it would be better to learn from the program’s experience and apply these lessons in the context of other government programs. In short, Part D's success should make it a model for future programs and future reforms.
 HannsKuttner, Introductory Essay in HannsKuttner and Tevi Troy, eds. “The Medicare Drug Benefit Five Years Later: Is it working?” Hudson Institute, December 2011.
 Terence Kealey, The Economic Laws of Scientific Research. St. Martin's, 1996, gives the example of histamine-2 blockers replacing surgery for the treatment of gastric ulcers, pp. 226-228. This is just one example of a larger trend. One more recent example is that of infliximab (Remicade), which reduces the percentage of ulcerative colitis patients requiring a colectomy from 17% to 10%. Sandborn, W.J. et al. “Colectomy Rate Comparison After Treatment of Ulcerative Colitis With Placebo or Inﬂiximab” Gastroenterology, 2009; 137: 1250 –1260.
 Medicare Payment Advisory Commission, “Report to the Congress: Medicare Payment Policy,” Chapter 15, March 2013, p. 344.
 Andrew Stocking, “Competition and Bids in Medicare's Prescription Drug Program,” Congressional Budget Office, June 23, 2013, p.8-9
 Amy J. Davidoff, Bruce Stuart, Thomas Shaffer, J. Samantha Shoemaker, Melissa Kim, and Christopher Zacker, “Lessons Learned: Who Didn’t Enroll In Medicare Drug Coverage In 2006, And Why?” Health Affairs. 29:6, pp. 1255-1263, June 2010.
 Medicare Payment Advisory Commission, “Report to the Congress: Medicare Payment Policy,” Chapter 15, March 2013, p. 333-334.
 See CBO, “Preliminary Analysis of the President’s Budget for 2012,” March 18, 2011, p. 12. http://www.cbo.gov/ftpdocs/121xx/ doc12103/2011-03-18-APB-PreliminaryReport.pdf and CBO, “Updated Budget Projections: Fiscal Years 2012 to 2022,” March 2012, p. 9. http://www.cbo.gov/sites/default/ files/cbofiles/attachments/March2012Baseline.pdf. See also CBO Part D Baselines for Part D Mandatory Outlays for 2010 through 2013, available at www.cbo.gov.
 Report on Access, Satisfaction, and Cost,” AARP, November 2007
 Medicare Payment Advisory Commission, “Report to the Congress: Medicare Payment Policy,” Chapter 15, March 2013, p. 345.
 Kuttner, op. cit. . citing KRC Research, “Seniors' Opinions About Medicare Rx: Sixth Year Update,” October 2011.
 J.M. McWilliams et al. “Implementation of Medicare Part D and Nondrug Medical Spending for Elderly Adults with Limited Prior Drug Coverage,” Journal of the American Medical Association. July 2011.
 Afendulis&Chernew, “State Level Impacts of Medicare Part D,” The American Journal of Managed Care. October 2011.
 See Dall et al. “The Economic Impact of Medicare Part D on Congestive Heart Failure,” The American Journal of Managed Care, May 2013.
 Congressional Budget Office, “Offsetting Effects of Prescription Drug Use on Medicare’s Spending for Medical Services,” November 2012, available at http://www.cbo.gov/publication/43741.
 Andrew Stocking, “Competition and Bids in Medicare's Prescription Drug Program,” Congressional Budget Office, June 23, 2013. (Figure cited is for 2008.)