Health Affairs, 23, no. 1 (2004): 103-106 doi: 10.1377/hlthaff.23.1.103 © 2004 by Project HOPE	New Online   Senate Health Reform Bill   Rewarding Providers   Public Option Policy Brief   Health Reform & Abortion   Delivery System Reform

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Berndt, E. R. Search for Related Content PubMed PubMed Citation Articles by Berndt, E. R. Related Collections Medicare Pharmaceuticals Research And Technology

Medicare Challenge

PERSPECTIVE

Unique Issues Raised By Drug Benefit Design

Abstract

 
In this Perspective on the preceding paper by Joseph Newhouse, I point out a number of features of the pharmaceutical industry that differentiate it from other health care sectors. These differences help explain why it has proved to be so very difficult to construct policies that simultaneously contain health care costs, provide patients with high-quality care, and generate continued incentives for innovation. I then summarize Newhouse’s preferred Medicare prescription drug benefit program and the issues it raises.

As of this writing, it was still unclear whether Congress would be able to reach an agreement on rather different versions of legislation creating a Medicare prescription drug benefit, or whether a much less ambitious (and likely more defensible on economic grounds) "Plan B" would emerge that eliminates the infamous "doughnut hole" and instead offers primarily catastrophic benefit coverage, perhaps means-tested, with a relatively large deductible. Most of the issues raised by Newhouse, however, remain important, even if a much different version of the Medicare prescription drug benefit eventually is passed.

In terms of the "big picture," Newhouse states that "Medicare should not initially attempt to set prices, but to prevent abuses in pricing, Congress should allow cost to be considered in coverage decisions" (p. 89). Newhouse does not spell out why he qualifies his recommendation with the words "initially attempt," nor does he discuss what would likely follow after the initial scenario plays out. Presumably, given his own wealth of experience advising Medicare and his awareness of peculiarities of the health care industries, Newhouse foresees an iterative, evolutionary process of experimentation, evaluation, and revision as stakeholders accumulate experience. It would be useful to learn more from him regarding how such a process might best be structured and staffed. By stating that Congress should allow cost to be considered in coverage decisions, Newhouse is making explicit what Medicare and Medicaid have at times indicated they would like to do but have been prevented from doing because of existing law. It is a bold statement, understandable because the source is an economist who well understands what inevitably occurs when resources are scarce and demands are unlimited. But how cost-effectiveness will be assessed, and by whom, are critical issues that are regrettably well beyond the scope of his paper or this one.

Unique Issues In Benefit Design

Top Editor's Notes Unique Issues In Benefit... Newhouse’s Envisioned... Notes

 
Before summarizing some of Newhouse’s more detailed recommendations, I believe that it would be informative to reflect on ways in which pharmaceuticals raise different, or at least make more prominent, issues than those that are typically encountered in benefit designs for other health care services. This could help explain why it has proved to be so difficult to come up with policies that simultaneously contain health care costs, provide patients with high-quality care, and generate incentives for innovation.

High up-front R&D costs. First, although research and development (R&D) is very important in much of health care, for pharmaceuticals the ratio of up-front fixed or sunk costs to short-run variable costs of production is extremely high—making the first tablet that is safe and efficacious in treating an illness cost hundreds of millions of dollars but the second and subsequent tablets, typically dimes or quarters. As Newhouse and others have noted, this means that policies that enhance short-run or static efficiency (priced low, at short-run marginal cost) will typically conflict with those focusing on dynamic efficiency (use of the patent system and temporary market exclusivity to provide incentives to innovate, priced above short-run marginal cost). While the role of R&D and fixed/sunk costs is also important for, say, medical devices, surgical procedures, and dental procedures, for most drugs the post–discovery and development variable costs per unit are truly minuscule relative to up-front R&D spending.

It is worth noting, however, that this is not necessarily the case for related products, such as some biologics. For many of these products, such as those used in the treatment of anemia, hepatitis C, multiple sclerosis, rheumatoid arthritis, psoriasis, and treatment-resistant allergies, manufacturing is complex, not readily scalable, and costly. In terms of the short-run versus long-run efficiency trade-off, biologics are more like other medical products than pharmaceuticals. Thus, it is interesting that the example of cost-based Medicare pricing that Newhouse cites involves erythropoietin, a biologic. It is unclear how well this example generalizes to pharmaceuticals.

High drug resale values. Second, for many medical services (such as physical examinations, surgeries, doctors’ visits, and hospitalizations), resale and distant purchasing are impractical. A purchaser of a hospitalization cannot resell the service at a higher price to someone willing to pay more. One implication is that price differentials for these services cannot easily be arbitraged away. Moreover, even if hospitalizations and surgeries are much less expensive in Canada or the United Kingdom than in the United States, very few U.S. residents are likely to travel to Canada or the United Kingdom to be hospitalized to save costs. Geographically based price differentials for many medical services can persist. For pharmaceuticals, however, not only is resale possible (witness the sales of prescription drugs from Canadian pharmacies to U.S. consumers), but transportation costs are largely irrelevant, since shipping costs of drugs are trivial relative to "shipping" a patient for surgery. Hence, although pharmaceutical firms are vulnerable to having U.S.-Canadian price differentials arbitraged away, the forces of arbitrage are unlikely to put pressure on U.S.-Canadian price differentials for surgeries, hospitalizations, and laboratory services.

Intense adverse selection. Third, although adverse selection has long been recognized as an important phenomenon contributing to market failure in various health care markets, it is particularly intense for pharmaceuticals. The reason is that because many medications are used to treat chronic conditions, as Mark Pauly and Yuhui Zeng have documented, individuals’ pharmaceutical expenditures are highly autocorrelated and therefore predictable.¹ As noted by Pauly and Zeng, a market-based, stand-alone pharmaceutical insurance program is not sustainable in the long run, for it is vulnerable to adverse selection. Pharmacy benefit managers (PBMs) are therefore understandably reluctant to assume risk comparable to that assumed by other health care insurers.

Retail-sector dominance. Fourth, in most other sectors of health care products and services, the retail sector plays a relatively minor role. In pharmaceuticals, however, retail pharmacies dominate dispensing; hospitals are responsible for only about 10 percent of prescriptions; and while mail-order dispensing has been growing rapidly, it is still much smaller than retail.² Developments in information technology and telecommunications have affected the general retail sector dramatically and pharmacies specifically, reducing transaction costs and facilitating the inexpensive real-time monitoring of transactions. Mail-order dispensing is particularly cost-effective for medications for chronic conditions, but at least since the days of Sen. Hubert Humphrey, retail pharmacies have wielded very strong populist political clout. Newhouse does not discuss in detail how retail versus mail-order dispensing would occur. Would the characteristics of the dispensing medium be specified a priori by Medicare, or would they be a choice component of the PBM bidding process?

Potential redistribution of retiree benefits. Fifth, prescription drug benefits are already provided to a sizable portion of the Medicare population through private- and public-sector retiree benefits. How a new Medicare-funded drug benefit program would interact with existing retiree drug benefit programs raises enormously complex issues regarding employers’ honoring commitments to their retirees and to the government. Defense contractors, for example, have built expected retiree drug benefit expenditures into their contracts with the U.S. government and have already received funding for those expected future expenditures. Not only does the provision of a Medicare-funded drug benefit contain the possibility of a massive redistribution of retiree benefit obligations, but it is also likely to affect the labor market. One possible impact is to make the hiring of the cohort ages 50–65 more attractive to employers. Currently, employers may be reluctant to hire such people, because the present value of their retiree drug benefits is relatively high. With a Medicare drug benefit, the present value of the future drug obligation might be reduced, thereby making older people more attractive to hire. As best I can tell, there has been little discussion of the impact of a Medicare drug benefit on labor market outcomes for various working-age cohorts.

Incentives for price discrimination. Sixth and finally, although the previous five characteristics differentiate pharmaceuticals from other health care sectors to a considerable extent, one attribute is somewhat erroneously attributed primarily to pharmaceuticals, when in fact it occurs much more generally: differential pricing, or what economists call price discrimination. Undoubtedly the adoption of Medicare in the 1960s reduced physicians’ differential pricing on the basis of patients’ ability to pay, but even today hospitals are known to charge considerably different prices to various payers, as are physicians. Since most contracts between private payers and providers are proprietary and confidential, not much is publicly known concerning the extent of price discrimination in health care sectors outside of pharmaceuticals. Do hospitals, nursing homes, and physicians exercise more price discrimination than biotech and pharmaceutical firms? We do not know. One trend, however, seems quite clear. To the extent that scientific and commercial developments occur that facilitate the growth of personalized medicine, the possibilities for exercising what economists call first-degree price discrimination (charging each person exactly what he or she is willing to pay, rather than charging all members of a buying group the same price) could emerge. Although the issue clearly merits a thorough analysis, at first glance it appears that personalized medicine would create incentives for price discrimination that would be particularly strong in the genomics, biotech, and drug industries.

Newhouse’s Envisioned Medicare Drug Benefit

Top Editor's Notes Unique Issues In Benefit... Newhouse’s Envisioned... Notes

 
Role of PBMs. With all of this as background, let me summarize Newhouse’s preferred Medicare prescription drug benefit program and the issues it raises. Newhouse envisages competition for two- or three-year contracts rather than competition for enrollees, because of the presence of strong adverse selection in pharmaceuticals. He envisages a single PBM for each of about sixty regions, with no single PBM serving more than twelve regions simultaneously. Although Newhouse would prefer to see PBMs assume some risk, he comprehends their limited ability to affect drug use in the Medicare context. In particular, as Newhouse states, "However difficult it is for a private insurer to exclude a drug because of cost, it is even more difficult for a public payer such as Medicare to do so" (p. 98). This reduces the bargaining power not only of Medicare but also of any PBM acting as a contractor to Medicare.

Formularies. In terms of formularies, Newhouse envisages annually updating them, although it would seem very useful to place generics on the formulary immediately following entry, provided their price is lower than that of the pioneer brand (this is not always the case, particularly in cases where a generic manufacturer has 180-day exclusivity because of a successful Paragraph IV certification).

Newhouse does not discuss issues of formulary response to switches involving prescription to over-the-counter (OTC) status, but such issues could be endogenous to the competitive bidding process. While formularies would be primarily open rather than closed, in terms of prices faced by beneficiaries, Newhouse prefers what he calls "incentive pricing." As I understand it, PBMs could have tiered pricing such as the common three-tier copayment benefit design for all drugs on the formulary. Beneficiaries wanting an off-formulary drug would pay 100 percent of the difference between the PBM’s price of the off-formulary drug and the reference drug in that therapeutic class. Newhouse also appears to be open to coinsurance schemes in which even on-formulary drugs would be subject to beneficiaries’ paying 100 percent of the difference. Drawing the boundaries of the therapeutic class would, in all cases but one, be a task assigned to PBMs, not Medicare.

The one important exception would involve cases in which a drug manufacturer develops a clinically important drug with no close competitor. If in addition the share of total U.S. utilization of that drug by Medicare beneficiaries was likely to be very high (what that threshold might be is unclear), Medicare might be unable to use private-sector insurer prices as benchmarks. In such a case, argues Newhouse, Medicare might take cost-effectiveness criteria into account. Whether it would actually do so is unclear, but Newhouse argues that the mere threat of such an action could serve as a deterrent to potential abusive pricing.

It is the possibility of abusive pricing in such cases that causes Newhouse to take a wait-and-see attitude, similar to President Reagan’s "trust but verify" game theory view of disarmament. Perhaps that is why the words "How Will the Game Be Played?" appear in the title of the initial article that Newhouse coauthored in 2000 on Medicare drug pricing.³ It is a high-stakes game indeed, with a very large number of players.

Editor's Notes

Top Editor's Notes Unique Issues In Benefit... Newhouse’s Envisioned... Notes

 
Ernst Berndt is the Louis B. Seley Professor of Applied Economics at the Massachusetts Institute of Technology (MIT), Alfred P. Sloan School of Management, and director of the Program on Technological Progress and Productivity Measurement, National Bureau of Economic Research (NBER), in Cambridge, Massachusetts.

The views expressed here are the author’s and not necessarily those of MIT or NBER.

Notes

Top Editor's Notes Unique Issues In Benefit... Newhouse’s Envisioned... Notes

 

1. See M.V. Pauly and Y. Zeng, "Adverse Selection and the Challenges to Stand-Alone Prescription Drug Insurance," August 2003, www.nber.org/books/garber7/pauly-zeng8-6-03.pdf (27 October 2003).
2. See E.R. Berndt, "Pharmaceuticals in U.S. Health Care: Determinants of Quantity and Price," Journal of Economic Perspectives 16, no. 4 (2002): 45–66.[Web of Science][Medline]
3. See H.A. Huskamp et al., "The Medicare Prescription Drug Benefit: How Will the Game Be Played?" Health Affairs (Mar/Apr 2000): 8–23.

                      What's this?