Understanding the Bipartisan Senate Finance Prescription Drug Reform Package

Editor’s Note: This analysis is part of the USC-Brookings Schaeffer Initiative for Health Policy, which is a partnership between Economic Studies at Brookings and the University of Southern California Schaeffer Center for Health Policy & Economics. The Initiative aims to inform the national health care debate with rigorous, evidence-based analysis leading to practical recommendations using the collaborative strengths of USC and Brookings.

On July 25, the Prescription Drug Pricing Reduction Act (PDPRA) of 2019 was approved by the Senate Finance Committee by a 19-9 vote, with small modifications to the initial Chairman’s mark. Taken together, the legislation seeks to address many issues with prescription drug markets and coverage that affect Medicare and its beneficiaries, focusing on lowering drug prices, improving the benefit design of Medicare’s outpatient prescription drug benefit (Part D), and eliminating myriad impediments to competition.

The Congressional Budget Office (CBO) estimates that the package would reduce deficits by more than $100 billion over ten years, beneficiary cost-sharing in Part D by $25 billion, and Part D premiums by $6 billion, while also somewhat reducing the cost of commercial prescription drug coverage (due to spillover effects from the Part D inflation rebate policy).

After summarizing several issues with how prescription drugs are currently priced and sold, this blog describes how the PDPRA seeks to address these issues, including the important trade-offs that need to be weighed.

Underlying Issues with the U.S. Prescription Drug System

Today’s prescription drug system is characterized by several problems. The first—and the one that gets the most attention—is the high prices paid for prescription drugs, particularly those offering limited value, by both patients and health plans. Many studies have shown that other high-income nations achieve both lower list and actual (“net”) prices for almost all prescription drugs and limit the utilization of lower-value drugs (that is, drugs that are expensive but result in limited clinical impact compared to lower-cost competitors), resulting in significantly lower spending. As a result, the U.S. accounts for an estimated 64 to 78 percent of all worldwide pharmaceutical profits. This raises questions about the value of our drug spending, and high U.S. prices lead to access and affordability issues while resulting in significant costs to taxpayers.

Another critical issue is the design of the Medicare Part D benefit, which currently covers 47 million Americans, but was shaped by a political and fiscal compromise. As enacted in 2003, the benefit was designed with a coverage gap—often referred to as the “donut hole”: most Part D beneficiaries paid the full cost of drugs after an initial layer of insurance coverage but before reaching a catastrophic spending threshold (above which beneficiaries were responsible for 5% of drug list prices). The design, combined with how prescription drug spending has become over time increasingly concentrated among patients with high costs, has distorted incentives for health plans to manage drug spending while shifting costs to Medicare and imposing heavy financial burdens on those patients with the highest spending on drugs. For beneficiaries, Part D plans do not include an annual out-of-pocket spending limit, undermining a vital protection of insurance that is the norm in commercial insurance, leaving a small percentage of beneficiaries exposed to unlimited out-of-pocket spending for prescription drugs. In addition, the gap between list and net prices—caused by rebates paid after patients have purchased their prescriptions—has grown markedly over time, but Part D plans frequently base patient cost-sharing on high list prices, not the lower net prices plans that are actually incurred. Part D also includes a very generous federal reinsurance program, so the federal government (not the Part D plans) bears much of the cost of high cost patients, reducing plans’ incentives to manage prices and utilization.

The PDRPA attempts to address these issues and tackle other business practices that distort incentives and impede efforts to lower costs.

Prescription Drug Pricing Reduction Act

The PDRPA aims to address these intertwined problems with America’s prescription drug system. Modernizing Part D’s benefit design offers the potential to add much-needed beneficiary financial protections while correcting distorted incentives that inflate drug spending. First, the bill establishes a new $3,100 out-of-pocket cap on drug spending, a valuable consumer protection. While eliminating unlimited patient cost-sharing for the subset of beneficiaries incurring truly catastrophic costs arguably increases consumer demand and somewhat decreases downward demand-side pressure on drug prices, PDPRA’s benefit redesign includes two elements intended to lower prices for high-cost drugs. First, it mandates that brand drug manufacturers pay a new 20 percent rebate off the list prices for drugs utilized by beneficiaries who have reached their out-of-pocket limit—in place of the 70 percent rebate currently required in the donut hole portion of the benefit. Second, the bill reduces the amount of federal reinsurance paid to Part D plans for high cost patients, which increases plan financial responsibility for drug spending above the new cap, strengthening plan incentives to manage utilization and negotiate for lower prices.

The PDPRA also includes multiple provisions intended to lower drug prices. These provisions include requiring price concessions (rebates) from manufacturers payable to the government if drug prices (specifically, list prices in Part D and average sales prices in Part B) grow faster than inflation, and modifications to Medicaid drug rebates. These proposals generate most of the savings projected by CBO for the federal budget, Part D premiums, and beneficiary cost-sharing. These savings primarily result from lowering drug prices for those existing drugs that face minimal competition whose prices would have increased faster than inflation under current law.

An important consideration in assessing the drug pricing debate concerns the impact of lowering drug prices on the development of new drugs. While existing evidence is clear that higher rewards to drug development in a therapeutic class tend to result in more new drugs in that class, little is known about the exact nature of that relationship or the social value of the newly developed drugs—and this relationship is likely characterized by diminishing marginal returns to investments in research and development. Apart from effects on total drug spending, another key consideration is how the PDPRA modifies incentives to develop different types of drugs–what has been termed the innovation-innovation tradeoff—which Scott Gottlieb, Benedic Ippolito, and Abigail Keller discuss at length here.

Part D Benefit Redesign

The donut hole benefit structure in Medicare Part D started out as a compromise that combined front-end coverage (as originally proposed by the Clinton Administration) and catastrophic coverage (added by the Republican Congress). Because the 10-year cost that was deemed politically feasible for Part D ($400 billion at the time of enactment) was insufficient to provide a continuous benefit, the donut hole coverage gap became the easiest way to achieve both front end and catastrophic coverage. Today, the donut hole has been filled in by requiring manufacturers to cover most of the costs. Specifically, the Part D benefit includes a deductible of $415 during which beneficiaries pay all costs; an initial coverage period that extends until total drug costs reach $3,820, during which the Part D plan pays an average of 75% of drug costs and beneficiaries pay 25%; the donut hole from $3,820 to approximately $8,140, during which plans pay 5%, manufacturers pay 70% and beneficiaries pay 25%; and catastrophic coverage above $8,140, after which a beneficiary pays 5 percent, the plan pays 15 percent, and federal reinsurance covers 80 percent. The very limited amount of risk faced by Part D plans above $3,820 in drug spending raises concerns of a concomitant lack of resources going to efforts to manage drug spending through techniques such as formulary development, prior authorization, and step therapy.

Changes in the Prescription Drug Market
Changes in the prescription drug market since the enabling legislation was enacted in 2003 have made the benefit structure more problematic. The combination of rapid growth in both high cost drugs, especially those known as “specialty medicines”, and of low-cost generic drugs, which now account for 90 percent of prescriptions (but 22 percent of total drug spending), has resulted in three-quarters of federal Medicare Part D spending for basic benefits being on catastrophic costs. While from 2012 to 2017, gross spending by Medicare beneficiaries on all drugs covered under Part D increased by 11.5 percent per year, spending on specialty drugs increased by 29.7 percent per year. In 2003, cost estimates for the Medicare Modernization Act projected that spending in the catastrophic phase of the benefit would only constitute 34 percent of Medicare drug spending. Instead, MedPAC calculates that Medicare reinsurance payments to plans increased from 31 percent of Medicare reimbursement for basic benefits in 2007 to 72 percent in 2017.

Need for Systemic Reform of the Part D Benefit
The combination of two sets of legislative changes plus the modest share of costs plans incur for expensive drugs, the low cost of many generic drugs, and post-purchase rebates that reduce costs for plans but not at the point-of-sale for patients create perverse incentives undermining the utilization of generic or biosimilar versions of drugs that would lower patient and Medicare costs. Recognizing the cumulative effects of changes in the prescription drug market and Part D benefits, the PDPRA provisions would address these issues systemically, rationalizing beneficiary cost sharing, reinsurance payments to plans, and manufacturer discounts. Specifically, the coverage gap would be eliminated (rather than “filled in” as it is today, financed by mandated payments from manufacturers). Instead, a uniform coinsurance averaging 25 percent would be established from the initial deductible until the catastrophic threshold, with the Senate Finance Committee bill eliminating beneficiary cost sharing above the catastrophic threshold, triggered when out-of-pocket spending equals $3,100 in 2022. (The threshold would subsequently be indexed by the growth in per capita Part D spending.) Manufacturers would no longer have to provide price concessions in the coverage gap but instead would have to contribute a discount equal to 20 percent of their drug prices during the catastrophic phase for all enrollees, including those receiving low income subsidies. At the end of a phase-in period, federal reinsurance for Part D plans during the catastrophic phase would fall from the current 80 percent to 20 percent for brand name drugs and to 40 percent for generics.

These changes should also have indirect effects on drug prices through market forces. By sharply increasing plans’ incentives to contain drug spending, manufacturers would find that higher drug prices would lead to more efforts to shift usage to less costly therapeutic alternatives.

Potential Changes beyond PDPRA Reforms
These provisions would address many issues with the Part D benefit design and reflect recent thinking in the health policy community—they are generally consistent with recommendations MedPAC issued in 2016 and 2019. Two additional issues, however, might merit attention. One is rebates—price concessions negotiated between drug plans and manufacturers to lower prices but which are paid after patients have purchased their drugs at higher (undiscounted) prices; rebates are negotiated in return for plans’ steering demand toward that drug. As indicated in the recent report of the HHS Office of Inspector General, rebates do lead to lower net prices to purchasers including Medicare. But rebates in Medicare are used to lower premiums rather than to reduce cost sharing for those beneficiaries using the rebated drugs, meaning that patients taking highly-rebated drugs are paying disproportionately high cost sharing at the point of sale, limiting the ability of the program to provide more adequate financial protection. Some commercial insurers have been taking steps to direct rebates to those using drugs whose list prices substantially exceed their net costs. Even though one consequence of a similar approach in Part D is that premiums would be higher, having Medicare do the same would target financial relief from rebates on those patients who incur high drug costs instead of lowering premiums for all beneficiaries—regardless of whether they have high drug costs or no drug spending.

The second issue reflects on the dramatic changes in the prescription drug market over the decade and a half since the enactment of Medicare Part D. Given the rapid advances in research, the market will continue to change, though not necessarily along the same trajectory. To be able to adjust policy more rapidly in response to market changes, Congress might want to increase the authority of the Secretary of Health and Human Services to update the benefit structure. For example, the Medicare Trustees (who include the Secretary) could be authorized to include in their annual April Report an analysis of changes in the prescription drug market and recommend, if appropriate, modifications. MedPAC could assess the Trustee’s proposals and offer its advice to the Congress.

Directly Reducing Drug Prices

PDPRA includes provisions that would directly reduce some drug prices; the provisions together account for the bulk of the bill’s budgetary savings as estimated by the Congressional Budget Office (CBO), which total roughly $68 billion over ten years.

Part D Inflation Rebate
Average rebates for brand drugs covered under Medicare Part D have increased significantly in recent years, from 11.7% in 2012 up to 25.3% in 2018, according to the Medicare Trustees’ Report, but rebates vary significantly across different drugs and Part D plans. While Part D plans and PBMs have tools to negotiate discounts and rebates for drugs facing competition, they have limited ability to negotiate price concessions for sole-source brand drugs or drugs in one of Medicare Part D’s six protected classes, in which plans must cover all available drugs. At the same time, list prices and rebates continue to rise notably faster than inflation.

The PDPRA would require drug manufacturers to pay a price concession in the form of a mandated rebate directly to the government (separate and apart from rebates paid to plans and/or PBMs that reflect negotiated agreements) whenever a brand drug’s list price—reflected by its wholesale acquisition cost (WAC, which is defined in statute as a list, not net, price)—grows faster than general economy-wide inflation (the CPI-U). Beginning in 2022, this rebate would be equal to the difference between the drug’s list price and what the list price would have been if it had grown with inflation since July 1, 2019. While the inflation rebate is based on list prices, this provision also is estimated to reduce net prices for many drugs (which would produce savings). CBO estimates that savings from this proposal would grow quickly, reaching $10 billion annually by the end of the decade, producing cumulative savings of $57 billion through 2029.

Net price reductions would largely arise from existing patent-protected brand drugs that, under current law, would not have offered significant rebates and would have been able to increase both their list and net prices faster than inflation. Evidence that such a proposal would save money comes from multiple studies finding significantly higher net drug prices in Medicare Part D than in the Medicaid program and attributing much of that difference to Medicaid linking rebates to inflation. Analyzing 2012 data, the Office of the Inspector General (OIG) for the Department of Health and Human Services (HHS) attributes more than half of the net price differences to Medicaid’s inflation rebate.

It seems clear that this proposal would reduce spending over time on the current crop of brand drugs, but its effects are less clear for drugs that are launched in the future. Knowing that they will be unable to increase a drug’s list price faster than inflation, manufacturers will be incentivized to launch at a higher price than they otherwise would have. There are additional factors that go into setting a launch price, however, that may constrain the ability of manufacturers to increase launch prices for new drugs in response to the proposed rebate policy. If launch prices for new drugs were to rise to the extent that they fully offset the (subsequent) effects of the Part D inflation rebate, then this policy would have reduced spending on current drugs but, under this eventuality, would not reduce the expected returns for newly developed drugs.

Because this policy is expected to slow the growth of list prices, it would also have the effect of lowering beneficiary cost-sharing in Part D, which is currently based on list prices. CBO estimates that it would save beneficiaries $5 billion in cost-sharing from 2019-2029. The impact on Part D premiums is more ambiguous because two countervailing forces are at play. On the one hand, the inflation rebate should lead to lower net prices and lower Part D plan spending for certain drugs. But the newly-mandated governmental rebates will also crowd out a portion of Part D negotiated plan rebates for many drugs, as manufacturers will be less willing to offer price concessions on top of federally-mandated rebates to the government—and it does not appear that the government rebates will factor into premium calculations. CBO estimates that the first effect—of lower net prices—will be the dominant factor, causing Part D beneficiary premiums to cumulatively fall by $5 billion over ten years.

Lastly, by slowing list price growth for drugs that would not under current law have offered significant rebates (to a level below the net price that would have been paid under current law for certain drugs), CBO also expects the inflation rebate proposal to somewhat reduce drug spending for commercial insurance plans.

Part B Inflation Rebate
Physician-administered drugs and biologics covered under Medicare Part B are reimbursed based on the drug’s average sales price (ASP)—which is net of discounts and rebates—plus 4.3 percent. Part B drugs have been the subject of many reform proposals recently, as the current administered pricing system has come under scrutiny due to the lack of competitive forces at play, along with high prices relative to international comparisons. The PDRPA would seek to reduce Part B drug reimbursement by mandating a rebate from manufacturers to the government if a brand drug’s ASP grows faster than inflation. By focusing on net prices, this policy would reduce federal spending for any drug whose ASP would have grown faster than inflation under current law, producing a cumulative ten-year savings estimated by CBO at $10.7 billion.

However, there are some potential drawbacks to this approach. The Part B inflation rebate somewhat weakens incentives to offer lower launch prices, larger rebates, or discounts for new drugs, even if that leads to lower volume, because manufacturers would be unable to raise post-launch net prices faster than inflation. There may be value to allowing manufacturers to set a lower initial price while establishing the efficacy of a drug and subsequently raise prices if evidence developed over time highlights the value of a drug, although it is not clear how often this sequence of events would occur in practice.

Spread Pricing, PBMs, and Improving Data on Acquisition Cost

Drug pricing remains extremely complicated and, frequently, opaque, with hundreds of billions of dollars flowing between manufacturers, wholesalers, health plans, PBMs, pharmacies and other providers. The Senate Finance Committee proposal contains provisions intended to improve reported prices and transparency while reducing gaming, such as a provision to eliminate so-called “spread pricing,” a practice in which PBMs retain a portion of the difference between what a Medicaid managed care organization pays to them (which is frequently based on list prices) and what the PBM, in turn, pays to a retail pharmacy. This provision limits the ability of PBMs to mask a source of their revenue, but other significant sources of pharmacy overpayment for dispensing generic drugs persist. CMS has required fee-for-service Medicaid programs to base reimbursement on the actual cost of acquiring drugs plus the cost of dispensing prescriptions. In theory, paying actual ingredient cost eliminates margins when purchasing drugs, and paying reasonable dispensing fees—which has generated sizeable increases—allow retail pharmacies to generate a profit. However, the common measures of acquisition cost, such as the National Average Drug Acquisition Cost (NADAC) published by CMS, significantly overstate the net acquisition cost paid by pharmacies when purchasing generic drugs because NADAC does not either capture “off-invoice” rebates that flow from generic manufacturers to pharmacies (or entities that purchase on their behalf) or prohibit “transfer pricing”  where a large pharmacy chain reports high invoice prices at its retail pharmacies. Reforming the reporting of generic drug acquisition costs could generate significant savings.


Steven Lieberman is the president of Lieberman Consulting, Inc, and routinely works with trade associations and private equity firms on general policy analysis on healthcare developments. Lieberman is currently a board member for Tuple Health, a healthcare design and technology company, and Primary Care Coalition of Montgomery County, a non-profit. The authors did not receive financial support from any firm or person for this article or from any firm or person with a financial or political interest in this article.