Interpreting The First Round Of Maximum Fair Prices Negotiated By Medicare For Drugs

On August 15, 2024, the Centers for Medicare and Medicaid Services (CMS) released the negotiated prices, or Maximum Fair Prices (MFPs), for the first 10 pharmaceutical products selected for negotiation under the Inflation Reduction Act’s Medicare Drug Price Negotiation Program. These prices will be effective in 2026. CMS stated that if the MFPs had been introduced in 2023, they would have resulted in a 22% decrease in Medicare Part D net spending, or $6B. CMS also reported differences between MFPs and list prices in 2023 for each of the products. It did not, however, release information on the extent to which MFPs compared to Medicare Part D net prices at the drug level. This distinction is critical because most of the negotiated products were subject to rebates before the program was implemented, so their estimated net prices after rebates were considerably lower than their list prices. However, the actual net prices are not known, as rebates paid by manufacturers are considered proprietary and kept confidential. 

In previous work, we leveraged a novel method developed by our research team to estimate net prices faced by payers – including Medicare Part D plans –after confidential rebates paid by manufacturers. This methodology overcomes important limitations of other data sources and allowed us to distinguish among four types of payments: rebates negotiated between manufacturers and payers; 340B discounts; Medicaid discounts; and Medicare coverage gap.

In this article, we apply this same methodology to understand how Medicare’s 2023 spending on the selected drugs at the negotiated price levels—the MFPs—would have compared to actual Part D net spending on these drugs in 2023. This enables us to estimate the extent to which each of the drugs contributed to the $6B figure mentioned by CMS and provide context for how to interpret this estimate.

We obtained 2023 estimates of gross Part D spending for each drug from data released publicly by CMS. Using the differences between MFPs and list prices announced by CMS, we estimated what Medicare Part D spending for each of the ten drugs would have been in 2023 if MFPs had been in effect. We updated our previous estimates of Part D post-rebate, net prices – which were based on 2021 data – to 2023 using CMS data. We compared the resulting estimates of 2023 Part D net prices to MFPs, expressing differences in both absolute and relative terms.

Maximum Fair Prices Versus Medicare Part D Net Prices

Based on our analysis, of the ten drugs selected in the first round of negotiation, Stelara presented the greatest relative price reduction, with an MFP 42% below 2023 Part D net price, followed by Entresto (36%), Enbrel (33%), and Imbruvica (30%) (Exhibit 1). With the exception of Entresto, we had previously anticipated that CMS would achieve the largest price reductions for these specific products, relative to the other drugs subject to negotiation. This is because, for these products, the ceiling of the MFP was set by minimum statutory discounts, as opposed to the Part D net prices. Under the Inflation Reduction Act, MFPs are capped at the lower of two different standards: 1) the net price paid by Part D plans before negotiation; or, 2) minimum statutory discounts (based on drug’s age). CMS could further negotiate prices below these ceilings based on how the net prices and clinical benefits of the selected drugs compare to those of their therapeutic alternatives, as well as additional factors specified in guidance.

The MFP of Eliquis, the product with the highest gross spending, was 20% lower than the net price. A similar reduction in magnitude was seen for the other negotiated anticoagulant, Xarelto (18%). The MFP of Jardiance was 9% lower than the net price in 2023. Januvia was the antidiabetic drug with the greatest relative price reduction, with an MFP 30% lower than the net price. For Farxiga and Novolog/Fiasp, the MFPs equaled their net prices in 2023.

Exhibit 1: Comparison of Medicare Gross Spending, Net Spending, and Spending at Negotiated “Maximum Fair Prices” for Ten Selected Drugs, 2023

Source: Authors’ analysis.

Notes: *Gross Part D Spending and MFP as Discount off List Price released publicly by CMS.

^Estimated Part D Rebates based on authors’ update of analysis in Hernandez et al, 2024.

The variation observed in price reductions below the ceiling of the MFP were likely the product of manufacturer responses in the offer / counteroffer process and external factors. The lack of price reduction observed for Farxiga—which explains the positive remarks of AstraZeneca’s CEO on the negotiation process—might reflect the generic competition that the product faces. In other words, it is possible that CMS did not negotiate a reduction beyond the current Part D net price for this product because a large share of its utilization will be eroded by generic competition before the MFP goes into effect. In addition, in 2024, Novo Nordisk voluntarily decreased the list price of Novolog by 75% in 2024. This new list price was however not reflected in the data used by CMS as part of the negotiation process, which preceded the price reduction. The negotiated price of Novolog/Fiasp (insulin aspart) closely resembles the new list price of Novolog after the 2024 voluntary price reduction.

The $6B Figure—What It Means And What It Doesn’t

The $6B figure announced by CMS is obtained by summing the differences between spending at 2023 Part D net prices and spending that would have been observed in 2023 if MFPs had been in effect. Eliquis was responsible for $2B, of the $6B total, followed by Entresto ($906M), Stelara ($747M), and Imbruvica ($641M). According to our estimates, if MFPs had been in effect in 2023, Farxiga and Novolog/Fiasp would have had similar spending as compared with 2023 spending estimated based on Part D net prices.

As with the $6B figure, our estimates of differences in 2023 spending at negotiated prices versus spending at estimated Part D net prices in the same year cannot be interpreted as actual savings that can be expected from the negotiation once the new prices are effective in 2026. There are two reasons why these numbers likely overestimate actual 2026 savings from negotiation:

First, net prices and utilization of products selected for negotiation likely have, and will continue to, change over the period 2023-2026, particularly for products facing generic or biosimilar competition. For instance, Farxiga is already subject to generic competition and a biosimilar for Stelara is set to launch in early 2025. The utilization and net price of the branded versions would have been expected to change by 2026, making estimates of net spending in 2023 a poor predictor of 2026 spending.

Second, starting in 2025, the new Medicare Part D benefit design incorporates mandatory discounts applied to list prices for all drugs in the initial (10%) and catastrophic phases (20%). These mandatory concessions replace the Coverage Gap Discount program, which was part of the benefit design until 2024. Drugs selected for negotiation will be exempted from these mandatory discounts, which will instead be paid by CMS. Thus, in order to estimate savings from negotiation, it is important to account for the costs associated with exempting drugs from these mandatory discounts. The extent to which these waived mandatory discounts offset negotiated price reductions from negotiation will depend on three factors:

• The share of drug claims filled in the catastrophic phase, where mandatory discounts are 20%, compared to 10% in the initial phase.
• The size of manufacturer rebates off of list price, prior to the negotiation program. The larger the rebate (i.e. the difference between list and net price), the larger the newly negotiated price reduction that waived mandatory discounts offset. For instance, for a drug with a 70% rebate prior to the negotiation program, waiving a 10% discount off list price offsets negotiated price reductions as large as 33% below the previous Part D net price (this is because 10% of a hypothetical list price of $100 = $10 = 33% of a net price of $30). However, for a drug with a 27% rebate pre-negotiation, like Entresto, the waiving of 10% discount off list price only offsets a reduction of 13.7% from the previous Part D net price (10% of a hypothetical list price of $100 = $10 = 13.7% of the net price of $73). In short, waiving mandatory discounts for drugs selected for negotiation will offset larger price reductions below the pre-negotiation net price for drugs that had larger rebates before negotiation.
• The extent to which, in the absence of negotiation, pharmaceutical manufacturers would have responded to potential increases in mandatory discounts by withholding rebates. This is relevant in the estimation of the counterfactual, particularly for drugs with limited competition and high utilization in the catastrophic phase. In these cases, in the absence of negotiation, it is likely that manufacturers would have responded to increased mandatory discounts by withholding rebates offered to Part D plans before the implementation of the new Part D design.

Let’s illustrate this offset with the example of Entresto, for which we calculated that total 2023 spending at the MFP would have been $906 million lower than total spending at the estimated Part D net price. Assuming, conservatively, that all utilization happened in the initial (non-catastrophic) phase of the benefit design, waived mandatory discounts would amount to $343M, or 10% of $3.4 billion in gross spending on Entresto in 2023. Thus, the net impact of negotiation would be $563M, or the difference between $906M and $343M. This estimate assumes, however, that Part D net prices would have remained stable for Entresto under the new Part D benefit design, if it had not been chosen for the negotiation program. It is possible that, if the $343M figure exceeded mandatory discounts under the old Coverage Gap (estimated at $300M for this illustration), the manufacturer may have tried to compensate by decreasing voluntary rebates to Part D plans. In other words, if the product had not been negotiated, Novartis could have responded to the increase in mandatory discounts (from $300M to $343M) by withholding $43M in rebates to Part D plans. Under this scenario, the net impact of negotiation would be $606M ($563M plus $43M), assuming constant utilization in 2023-2026 and no changes to list price or rebates beyond the $43M withheld.

This point brings attention to a relevant but often overlooked reform introduced by the Inflation Reduction Act—The likely increase in manufacturer mandatory concessions for drugs not facing negotiation associated with the new Part D benefit design. It should be noted, however, that Part D mandatory concessions may not increase universally, as drugs with a large share of spending in the coverage gap prior to the introduction of the Inflation Reduction Act may see decreases in concessions with the new design.

Estimating Savings Associated With The Negotiation Program

The example above illustrates how, for Entresto, waived mandatory discounts partially offset net price reductions from negotiation. The extent to which price reductions are offset by exempted mandatory concessions will differ across drugs. However, waived discounts will not be relevant for Part B drugs selected for negotiation in future years of the negotiation program, as they are not subject to mandatory concessions. This is one of the reasons why the first round of negotiation is unlikely to deliver a generalizable estimate of the impact of selection of a product for negotiation in the future.

The second and more obvious reason why the first round of negotiation is not representative of future waves of the process is the varied nature of drugs likely to be selected for negotiation each year. Many products anticipated to be selected for negotiation in the second round do not currently have the large Part D discounts that some of the first 10 drugs had. In other words, we anticipate that a larger number of drugs selected for negotiation in the second round will fall under the Stelara/Enbrel/Imbruvica category of products, where minimum statutory discounts - rather than Part D net prices – provide a started point for negotiated prices. This will also be the most likely pattern that Part B drugs will follow, as they traditionally have had lower rebates than Part D drugs.

Conclusion

Estimating the overall impact of Medicare drug price negotiation on spending is more complex than the simple comparison of MFPs to Part D net prices before negotiation. The evaluation of the negotiation process will require consideration of waived mandatory discounts, and isolation of the effects of concurrent changes to the Part D design and external market factors such as generic competition. Because of the context-dependent nature of the negotiation process under current guidance, estimates of the net financial impact of the first round of negotiation are unlikely to be generalizable to future rounds of the negotiation process. Only rigorous analyses assessing the financial implications of several rounds of negotiation will be able to answer the question that reporters, policymakers, industry stakeholders, and health commentators are asking themselves—what are the savings associated with Medicare Drug Price Negotiation?

Author’s Note

The authors report funding from the Commonwealth Fund (grant 23-23524). The authors are solely responsible for the accuracy of the information presented in this article. Any views or opinions expressed in this article are solely those of the authors, and do not necessarily represent those of the authors’ institutions or the Commonwealth Fund. Dr. Hernandez reported receiving consulting fees from Pfizer and Bristol Myers Squibb outside the submitted work. Dr. Wouters reported receiving personal fees from the World Bank and World Health Organization outside the submitted work. Dr. Sullivan reported receiving consulting fees from Sanofi, Pfizer, Neurocrine and Novo Nordisk outside the submitted work.