Prescription Drugs: Spending, Use, and Prices

Summary

Prescription drugs have become an increasingly important part of U.S. health care, as evidenced by the growth in nationwide spending on those drugs from 1980 to 2018.1 Over that period, such spending increased more than tenfold in real terms (that is, with the effects of economywide inflation excluded). This report by the Congressional Budget Office discusses trends in nationwide spending on prescription drugs in the retail market from 1980 to 2018. It also presents a detailed analysis of trends in spending, use, and prices in the Medicare Part D and Medicaid programs over the 2009–2018 period.

What Are Recent Trends in Spending for Prescription Drugs?

Nationwide spending on prescription drugs increased from $30 billion in 1980 to $335 billion in 2018. (All estimates of drug spending and prices in this report are expressed in 2018 dollars.) Over that period, real per capita spending on prescription drugs increased more than sevenfold: from $140 to $1,073.2 That increase in spending was driven by the development and use of many types of drugs that have yielded myriad health benefits. Because of those health benefits, some drugs, such as those that treat cardiovascular conditions, are associated with reductions in spending on services provided by hospitals and physicians.3 Other types of drugs, such as those that treat multiple sclerosis or cancer, may not offer such compensating savings, but they have improved the lives of those with chronic conditions and have also extended life.

Spending on prescription drugs increased particularly rapidly after the mid-1990s with an increase in the number of drugs that attained “blockbuster” status by generating at least $1 billion in sales annually. Those blockbuster drugs generally treat conditions, such as high cholesterol or high blood pressure, that affect a large segment of the population.

Aside from a pronounced increase between 2013 and 2015, spending on prescription drugs has grown more slowly since the mid-2000s. That slower growth in spending—and the accompanying reduction in per capita spending—is associated with the growing availability of generic drugs, which tend to have much lower prices than their brand-name counterparts. The brief, sharp increase in spending between 2013 and 2015 coincided with the introduction of a particularly expensive class of drugs that treat hepatitis C.

Although per enrollee spending in Medicare Part D—Medicare’s pharmaceutical benefit—was fairly stable from 2009 to 2018, per enrollee spending in Medicaid increased somewhat faster than nationwide per capita spending over that period. Per enrollee spending on prescription drugs in Medicare Part D averaged about $2,700 per year. Annual per enrollee spending in Medicaid increased over that period by 20 percent, from $445 to $530, whereas nationwide per capita spending increased by 10 percent.

Differences in average spending among enrollees in Medicare and Medicaid and the nation as a whole most likely stemmed from differences in the health profiles of their respective patient populations and statutory rebates in the Medicaid program. (Those rebates represent payments from drug manufacturers to “payers,” such as commercial and government-sponsored health insurance plans and the Medicaid program.) Medicare beneficiaries are more likely to be prescribed medications for various chronic conditions, whereas many Medicaid beneficiaries who have prescription drug coverage through that program are younger and healthier and therefore are less likely to have medications prescribed for them on an ongoing basis.

How Has the Use of Prescription Drugs Changed Over Time?

Nationwide per capita use of prescription drugs has increased in recent years. Per enrollee use of prescription drugs has also increased in Medicare Part D and Medicaid—from an average of 48 prescriptions per year in 2009 to 54 in 2018 in Medicare Part D, and from 7 prescriptions per year to 11 in Medicaid over that period. Increased use of prescription drugs is primarily associated with the increasing availability and use of generic drugs, along with the continued development of new treatments. In addition, the share of spending on prescription drugs that insurers cover has increased substantially: In 1990, consumers paid 57 percent of their prescription drug costs out of pocket, on average. By 2009, that share had fallen to 20 percent; it fell further, to 15 percent, in 2018.

The share of prescriptions for generic drugs has also increased substantially. Nationwide, that share increased from 75 percent in 2009 to 90 percent in 2018. In Medicare Part D, the share of prescriptions for generic drugs increased from 72 percent to 90 percent between 2009 and 2018; in Medicaid, it increased from 70 percent to 87 percent over that period. Increased use of generic drugs is attributable to several factors: their growing availability; their lower prices; and the lower out-of-pocket liability for consumers with health insurance compared with the amount people would pay for brand-name drugs. The share of prescriptions for generic drugs may be less likely to rise in the future, both because the 90 percent dispensing rate for such drugs is already high and because newer brand-name drugs tend to be more costly to manufacture and may be more challenging to replicate as generic drugs.

How Has the Average Price of a Prescription Changed Over Time?

The average price of a prescription has fallen in both the Medicare Part D and Medicaid programs in recent years: from $57 in 2009 to $50 in 2018 in Medicare Part D and from $63 to $48 in Medicaid over that period. Those decreases were largely driven by the increased use of generic drugs in those programs. The growing use of generic drugs has put downward pressure on the nationwide average price of a prescription in recent years as well.

Brand-name drugs, while accounting for a declining share of prescriptions, have experienced substantial growth in average prices. Over the 2009–2018 period, the average price of a prescription for a brand-name drug more than doubled in the Medicare Part D program and increased by 50 percent in Medicaid. Two key drivers of those increases were higher launch prices for new drugs and growth in the prices of individual drugs already on the market. The growing shift toward specialty drugs among new drug entrants was an important factor in the increased launch prices of new drugs. (Specialty drugs, which treat complex, chronic, or rare conditions—such as different types of cancer, rheumatoid arthritis, or multiple sclerosis—tend to be costly to manufacture, serve relatively small markets, and have high prices. They may also require special handling or patient monitoring.)4

Federal policies have also played a role in the pricing patterns for brand-name drugs. For example, Medicaid’s statutory rebates create an incentive for manufacturers to negotiate higher prices for commercial insurers as well as higher marketwide launch prices. (However, those rebates also create an incentive for manufacturers to increase prices more slowly over time, which probably mitigates the effect of higher initial prices.) In addition, the increase in the share of overall drug spending that is covered by Medicare and Medicaid may dampen the pressure on manufacturers to restrain prices because patients are more willing to purchase high-priced drugs when they have less exposure to those prices.

Unlike prices for brand-name drugs, average prices for generic drugs have fallen in recent years. From 2009 to 2018, the average price of a prescription for a generic drug fell from $22 to $17 in Medicare Part D and from $27 to $23 in Medicaid. Although the federal government and nearly all of the states have pursued legal action against several generic drug manufacturers for price fixing and other anticompetitive behavior, prices have probably increased for only a minority of generic drugs that represent a relatively small share of spending on prescription drugs.

Trends in Spending for Prescription Drugs

From 1980 until the mid-2000s, spending on prescription drugs increased steadily—both in dollar terms and as a share of overall health care spending. That growth was driven by increases in the availability and use of different types of new drug therapies along with increasing prices of brand-name drugs. However, after the mid-2000s, the increasing availability and use of generic drugs put downward pressure on spending growth. Nationwide, real per capita spending and the share of overall health care spending attributable to prescription drugs began to decrease in the mid-2000s, with the exception of a sharp increase from 2013 to 2015. That increase coincided with the introduction of a particularly expensive class of drugs that are used to treat hepatitis C. Per enrollee spending was relatively flat in the Medicare Part D program from 2009 to 2018, whereas Medicaid’s per enrollee drug spending grew over that period.

Nationwide Spending on Prescription Drugs

Since 1980, the share of nationwide spending on health care services overall that can be attributed to prescription drugs has nearly doubled, from about 5 percent to almost 10 percent in 2018. Through the 1980s and early 1990s, 5 percent to 6 percent of all spending on health care services and supplies was on prescription drugs obtained in the retail market (that is, from pharmacies—either in stores or by mail order). By 2018, that share was 10 percent (see Figure 1). In comparison, the share of spending on health care services and supplies that was attributable to hospital services fell from 40 percent in 1980 to 31 percent in 2018, and the share attributable to services provided by medical professionals and in clinical settings was about 20 percent over that period.

Nationwide per capita spending on prescription drugs increased from $140 in 1980 to $1,073 in 2018 (see Figure 2).5 (Those spending amounts are net of rebates and discounts. For background information on rebates, negotiations, and other attributes of pharmaceutical markets, see Box 1.) Per capita spending on prescription drugs roughly doubled every 10 years before slowing down in the mid-2000s.

Although this report focuses on prescription drugs obtained through the retail market, spending on prescription drugs purchased in nonretail settings—such as physicians’ offices, clinics, and hospitals—represents a substantial share of overall spending on prescription drugs. A recent estimate suggests that nonretail drugs represented approximately 30 percent of overall net spending on prescription drugs in the United States in 2016.6 The general pattern of total drug spending over the period studied in this report is similar when those other sources and types of prescription drugs are included.7

Trends in Spending Over the 1995–2013 Period. Spending on prescription drugs rose particularly rapidly after 1995 as a number of drugs reached blockbuster status.8 That wave of blockbuster drugs consisted of top sellers that were in high demand because they offered new treatments for relatively common health conditions. The most prominent of those drugs were statins for high cholesterol, ACE inhibitors for high blood pressure, proton-pump inhibitors for acid reflux and gastric ulcers, and antidepressants and antipsychotics for mental illnesses. When the patents on those drugs began to expire—an event often referred to as the patent cliff—lower-priced generic substitutes were introduced and gained market share.9 In the midst of that shift, the Medicare Part D program was created, which both increased access to prescription drugs for Medicare beneficiaries and contributed to the shift from brand-name drugs to generic alternatives.10 As a result, per capita spending on prescription drugs leveled off at about $940 in the mid-2000s and then fell to $900 by 2013, and the share of overall health care spending for prescription drugs peaked at 10.5 percent in 2006 and then fell to 9.2 percent in 2013.

Trends in Spending Since 2013. Spending on prescription drugs increased again over the 2013–2015 period before leveling off thereafter. A key factor in that increase was the introduction, at the end of 2013, of a class of specialty drugs that treat hepatitis C. The drugs for hepatitis C were introduced at particularly high prices.11 Insurers and patients may have been willing to pay those prices because of the enormous clinical benefits those drugs offer when compared with older therapies (they successfully treat about 95 percent of patients with chronic hepatitis C infection) and the stiff demand for that kind of therapy.12 However, the patient population for that class of drugs is larger than that for many specialty drugs, leading to a larger impact on spending than would have occurred for a similarly priced drug that treats a smaller patient population.

Implementation of the insurance expansions under the Affordable Care Act (ACA) also occurred during this period.13 Researchers have found that new take-up of Medicaid in states that expanded coverage under the terms of the ACA was associated with greater use of prescription drugs because those benefits reduced out-of-pocket spending for the previously uninsured and also tended to be more generous than previous sources of coverage.14 Take-up of marketplace health insurance options under the insurance expansions may have been associated with a similar response. However, overall use of prescriptions in the Medicaid program did not grow faster in the years immediately following the insurance expansions than it did in prior years, suggesting that the expansions were not a key driver of the increase in spending on prescription drugs over that period.15

Medicare and Medicaid Spending on Prescription Drugs

Measured in 2018 dollars, spending on prescription drugs grew from $74 billion in 2009 to $120 billion in 2018 in the Medicare Part D program and from $18 billion to $32 billion in Medicaid. Those programs account for a large share of all U.S. spending on retail prescription drugs. Together, beneficiaries in those programs were responsible for about 45 percent of nationwide spending on retail prescription drugs in 2018 as measured in the National Health Expenditure Accounts. The analysis in this section covers a shorter period than the period examined for nationwide spending because Medicare Part D was not implemented until 2006, and early data were not representative of the program’s spending because it took time for enrollees to adapt to the new program. This was particularly true for people who were dually eligible for Medicare and Medicaid and whose prescription drug coverage transitioned from Medicaid to Medicare Part D.16 CBO therefore decided to focus on the most recent 10-year period for which data were available for both programs. (Those totals reflect amounts spent by insurers and patients, less rebates and discounts for brand-name drugs.)

For the purposes of making comparisons between different populations, it is most meaningful to compare patterns in per capita or per enrollee spending. Changes in spending per person reflect a variety of factors, such as changes in average health status—both at the population level and the program level—and changes in prices. For example, the aging of members of the baby-boom generation most likely increased nationwide per capita spending over the 2009–2018 period because of the corresponding increase in the average age of the population.17 At the same time, that trend probably put downward pressure on per enrollee spending in Medicare Part D because it corresponded to a larger relative increase in the younger Medicare population. Changes in total spending are also affected by population and enrollment growth—and enrollment in Medicare Part D and Medicaid grew much faster than the nationwide population over the study period.18 For Medicare Part D, net per enrollee spending on prescription drugs remained relatively stable, averaging about $2,700 from 2009 to 2018. Medicaid’s per enrollee spending grew by about 20 percent over that period, starting from $445 per enrollee in 2009 and reaching $530 in 2018.19 Over the same period, nationwide per capita spending grew by nearly 10 percent—from $957 to $1,073.

The greater increase in Medicaid’s per enrollee spending may partly relate to new specialty drugs that treat conditions, such as HIV or hepatitis C, with a higher prevalence in the Medicaid population than in the Part D population.20 Another key factor is a greater increase in per enrollee use of prescription drugs in Medicaid than in Medicare Part D, as demonstrated below. The role of the Medicaid expansions in those trends is unclear, depending on the average usage patterns of the newly eligible Medicaid population compared with the previously eligible population. One study found that increases in prescription volume were similar to increases in enrollment in states that expanded Medicaid, suggesting that the impact on per enrollee spending depends on the average prices for drugs used by the newly eligible population compared with the previously eligible population.21

Differences in the amounts of per enrollee spending in Medicare Part D and Medicaid and in per capita spending in the United States as a whole are stark. They are most likely driven by a combination of differences in average health status and statutory rebates in the Medicaid program. Per enrollee spending in Medicare is much higher than the national average, probably because many Medicare beneficiaries have chronic health conditions and may fill several prescriptions per month. By contrast, lower per enrollee spending in Medicaid is probably attributable to a combination of the statutory rebates in that program—which lead to lower net prices—and the fact that many Medicaid beneficiaries with prescription drug coverage are relatively healthy adults or children. That offsets higher average spending by the less healthy disabled population in the Medicaid program.

Trends in the Use of Prescription Drugs

Utilization of prescription drugs nationwide has increased in recent years, both because the prevalence of chronic conditions has increased with the aging of the U.S. population and because new therapies and generic drugs have become available.22 Over the past several decades, consumers, health care providers, and insurers have witnessed the emergence of blockbuster therapies that treat common conditions, such as high blood pressure, high cholesterol, anxiety, and depression; and the emergence of generic alternatives for those drugs has improved access to such treatments by offering lower-cost options. In addition, consumers’ share of spending on prescription drugs has fallen substantially as the combined share of spending covered by federal programs and private health insurance has increased over time.

Use of prescription drugs among those enrolled in Medicare Part D and Medicaid increased as well. Administrative data about Medicare Part D show that from 2009 to 2018 the average number of standardized prescriptions per beneficiary rose from 48 to 54 per year—a 13 percent increase. (Standardized prescriptions are adjusted to 30-day equivalents for more than a 30-day supply.) According to administrative data about Medicaid, the number of prescriptions per person with Medicaid coverage for prescription drugs rose from an average of 7 to 11 per year over that same period—an increase of 57 percent. (The administrative data on Medicaid drug use and spending do not include information on days supplied and thus are unadjusted.) As with per enrollee spending, the variation in per enrollee use of prescription drugs between those two programs reflects differences in the health status of their beneficiaries.

Underlying Drivers of Increased Use

The reduction in consumers’ out-of-pocket costs for prescription drugs is one key factor that explains the increased use of prescription drugs. In 1990, consumers’ share of spending on prescription drugs was 57 percent. By 2009, that share had fallen to 20 percent. It continued to fall thereafter, declining to 15 percent in 2018. That long-term decline is largely explained by a gradual increase in the share of spending covered by the Medicare and Medicaid programs, which grew from 13 percent in 1990 to 36 percent in 2018.23

Some of that increase is attributable to the creation of Medicare Part D in 2006. In that year, the share of spending covered by Medicare and Medicaid increased to 25 percent, up from 19 percent in 2005. That share has steadily increased since 2006. More recent increases were partly attributable to the increased generosity of the Part D benefit that was mandated by both the ACA in 2010 and the Bipartisan Budget Act of 2018, as well as to the Medicaid expansions that were encouraged by the ACA.24

The role of private health insurance in paying for prescription drugs has also increased since 1990. Its share of spending was 26 percent in 1990 and 44 percent in 2018, although the share covered by private health insurance was highest in the early 2000s, ranging from 47 percent to 50 percent. That share has since fallen.

Greater access to generic drugs in those programs may be another key factor that explains the increased use of prescription drugs: Lower-cost options make it easier for people to purchase their prescribed medications. Nationwide, the share of standardized prescriptions dispensed for generic drugs was 75 percent in 2009 and reached 90 percent by 2018.25 In Medicare Part D, the number of standardized prescriptions dispensed for generic drugs more than doubled from 2009 through 2018, whereas the number of standardized prescriptions dispensed for brand-name drugs declined by a third. In Medicaid, the number of generic prescriptions roughly tripled over that time, whereas the number of brand-name prescriptions was essentially unchanged (see Figure 3). As a result, the share of prescriptions for generic drugs in Medicare Part D increased from 72 percent in 2009 to 90 percent in 2018; that share increased from 70 percent to 87 percent in Medicaid over the same period. Increased use of generic drugs also helps explain why per enrollee spending in federal programs rose more slowly than the increase in overall use of prescription drugs in those programs.

Factors Contributing to Changes in the Use of Generic Drugs

One of the primary factors contributing to the increased use of generic drugs over the 2009–2018 period was the availability of generic equivalents for a growing number of brand-name drugs as their patents expired or were successfully challenged by manufacturers of generic drugs. That process accelerated in the first decade of the 2000s when the blockbuster drugs of the previous decade began losing their sales-exclusivity rights. In addition, insurers have used a variety of tools to steer patients toward generic drugs.

However, the rate of increase in the share of prescriptions for generic drugs has slowed in recent years. That reduced growth coincides with the leveling off of two former sources of growth: First, the share of prescriptions for which a generic option is available has equaled 92 percent since 2016. Second, since 2013, 97 percent of prescriptions that have both a brand-name option and a generic option have been dispensed as generic drugs.26 Also, going forward, further availability of generic drugs may be somewhat limited if newer and more complex brand-name drugs are less likely to attract generic competition. That could be the case if those drugs treat conditions that affect fewer patients and are more challenging to replicate.

Factors that Increase the Use of Generic Drugs. Health insurers use a variety of methods to encourage the use of generic drugs when they are available. A common tool is to charge lower out-of-pocket costs for generics than for brand-name alternatives. For example, employment-based insurance plans required a copayment—that is, a specified dollar amount that an enrollee pays at the time a drug is purchased—of $11, on average, in 2019 for prescription drugs in their first tier, which is usually largely restricted to generic drugs.27 Average copayments for drugs in the second and third tiers ranged from $33 to $59; those tiers tend to include preferred and nonpreferred brand-name drugs.28 Plans often require lower cost sharing for drugs with “preferred” formulary placement and higher cost sharing for drugs with “nonpreferred” formulary placement. (A formulary is the plan’s list of covered prescription drugs.) Plans typically require even higher cost sharing for specialty drugs, which are less likely to have generic alternatives.

Similarly, median copayments for generic drugs ranged from no cost for preferred generics to $3 for nonpreferred generics among stand-alone Medicare Part D plans in 2020. (The median copayment divides copayment amounts into two equal groups; that is, half of enrollees in stand-alone Part D plans paid nothing for preferred generics and no more than $3 for nonpreferred generics, and the other half paid no less than those amounts.) Copayments for generic drugs topped out at $10 that year, and the median copayment for preferred brand-name drugs was $42.29 In general, cost sharing for nonpreferred brand-name drugs and specialty drugs tended to be set at a percentage of retail prices—the prices that plans pay to pharmacies, which do not account for rebates provided by manufacturers.30

Other tools are used to manage utilization directly: For the most part, the Medicaid program requires that generic versions of a drug be dispensed when available, and most Medicare Part D plans exclude the brand-name version of a drug from its formulary when a generic alternative is available.31

When consumers pay the full amount for a prescription drug out of pocket, the difference in the amount they pay for a generic drug versus a brand-name alternative is generally larger than the differences described above. That leads to a greater incentive to choose a generic substitute over a brand-name drug. Consumers may have to pay the full amount for a drug for three reasons: They lack health insurance (or their plan does not include coverage for prescription drugs); their prescribed drug is not covered by their plan; or they have not met their insurance plan’s annual deductible (if any) for prescription drugs. (A deductible is the amount of spending an enrollee incurs before an insurer begins covering expenses.) In the first two cases, consumers may pay a pharmacy’s “usual and customary” price—which tends to be higher than the retail prices paid by other federal health care programs and commercial insurers—or they may pay a discounted amount using a coupon program such as GoodRx.32 In the third case, the consumer pays the retail price that the plan negotiated with the pharmacy. In 2019, 13 percent of people with employment-based insurance were enrolled in a plan with a deductible specific to prescription drugs, up from 10 percent in 2005.33

Constraints on Further Increases in the Use of Generic Drugs. The use of generic drugs may not increase much further for two reasons: First, generic drugs are used extensively. (Those drugs represent 90 percent of all prescriptions.) Second, newer drugs are more likely to be biologics (drugs that are produced from living organisms). Those drugs are more complex and harder to manufacture or replicate than small-molecule drugs. Consequently, there may be fewer generics developed from them when their patents expire.

Manufacturing a biologic with the same active molecule—called a biosimilar—introduces an additional layer of complexity compared with small-molecule drugs. Biosimilars are necessarily created from different cell lines than the originals, so they are not identical at the molecular level. As a result—unlike for generic versions of small-molecule drugs—noninnovator firms (that is, manufacturers of generic or biosimilar drugs) typically need to run clinical trials to demonstrate that their biosimilars are not meaningfully different from the reference biologic product. For certain biologic drugs that have small markets, the difficulty that prospective imitators might face is compounded. The lower potential revenues from sharing a small market, at lower prices, may increase the risk that firms producing biosimilars will fail to recover their higher development costs from imitating a complex drug.34 As a result, future competition from noninnovators may be less robust in some cases as the patents for today’s newer drugs expire.

In addition, drug manufacturers have been able to partly neutralize insurance plans’ ability to steer their enrollees toward generic drugs by issuing coupons directly to consumers. Drug coupons make expensive therapies more affordable and increase manufacturers’ unit sales. However, coupons interfere with insurers’ cost-control efforts: Specifically, by covering some or all of an enrollee’s copayment or coinsurance (cost sharing in the form of a set percentage of the drug’s cost), coupons reduce or eliminate the cost difference between a more expensive drug and a cheaper generic or preferred alternative. When a coupon induces an enrollee to choose a brand-name drug over a generic, it increases the cost to insurers because they then must cover the more expensive brand-name drug for that enrollee. (Coupons also provide a discount for consumers who have not yet met their deductible or who lack insurance coverage.)

Coupon programs offered by manufacturers have become more prevalent over time: Whereas in 2009 manufacturers issued coupons for fewer than 100 brand-name drugs, by 2015 more than 700 drugs were covered by coupons.35 Some policymakers have taken steps to limit the use of those coupons. For example, California has banned their use for brand-name drugs that have generic equivalents. By one estimate, that ban affects about 20 percent of the drugs covered by coupons.36 Other states are considering similar legislation. In addition, coupons for brand-name drugs cannot be used by Medicare and Medicaid beneficiaries because they constitute a violation of the anti-kickback statute.37

Trends in the Average Price of a Prescription

Nationwide data on the average prices of prescription drugs are not readily available, but it is unlikely that the average net price of a prescription has increased considerably in recent years. Nationwide per capita spending on prescription drugs has generally held steady or declined since the mid-2000s—other than the increase from 2013 to 2015—whereas use of prescription drugs has most likely increased over that period. Further, a recent industry analysis shows that reductions in spending resulting from losses of exclusivity and generic pricing reductions nearly offset the growth in spending resulting from the entry of new drugs and price growth among other brand-name drugs.38

Changes in average prices in the Medicare and Medicaid programs also support that assessment. In Medicare Part D, the average net price of a prescription (measured in 2018 dollars) fell from $57 in 2009 to $50 in 2018. The decline was greater in Medicaid; in that program, net prices fell from $63 in 2009 to $48 in 2018 (see Figure 4).39

Changes in the average net price of a prescription are driven by two opposing trends: increases in the use of lower-cost generic drugs and increases in the prices of brand-name drugs. The share of prescriptions for generic medications that people have purchased at retail pharmacies has grown to 90 percent. That shift toward generic drugs has put considerable downward pressure on the average price of the prescription drugs that people have purchased. However, average prices of brand-name drugs—which constitute the remaining 10 percent of prescriptions—have increased considerably over time. Those increases in average prices represent the combined effect of price increases for drugs already on the market and prices for new drugs, which tend to be higher than prices for drugs already on the market.

Comparing Prices Among Payers

Underlying the overall trends are differences in the prices paid by various payers. (Payers are the entities that pay for prescription drugs, namely commercial insurers and federal health care programs, as well as individuals without prescription drug coverage.) One key factor that drives those differences is that manufacturers provide different rebate amounts to different payers for a given drug. Another key factor is that people with different sources of coverage tend to use different sets of drugs that have different average prices.

Comparisons Between Medicare and Medicaid. The similarity in the net prices for drugs covered by Medicare and Medicaid masks large differences in average retail prices—that is, the prices paid to pharmacies—for the drugs that beneficiaries of those programs purchase. The average retail price of a prescription covered by Medicaid is much higher than that of a prescription covered by Medicare. In 2018, that average price was $98 for Medicaid beneficiaries—about 40 percent higher than the average of $69 that Medicare beneficiaries incurred that year. Because retail prices for a given drug tend to be similar in Medicare and Medicaid, those differences primarily reflect differences in the mix of drugs used in the two programs.40

The differences in average retail prices are most likely attributable to differences in the two populations’ underlying health care needs. Medicare beneficiaries tend to use less complex drugs that treat chronic conditions. Those drugs tend to have lower retail prices. By contrast, the prescriptions that Medicaid-only beneficiaries fill are more likely to be costly drugs that treat complex conditions. (Such drugs include psychotherapeutic drugs and HIV treatments.) Nevertheless, per enrollee spending on prescription drugs is lower in Medicaid than in Medicare because Medicaid beneficiaries tend to use fewer drugs.

In addition, Medicaid beneficiaries tend to have low cost-sharing requirements—and sometimes none at all—and any cost sharing is generally not tied to the price of a drug. Therefore, Medicaid beneficiaries may be more likely to fill prescriptions for more expensive medications than Medicare beneficiaries, whose cost sharing is more directly tied to the price of a drug.

Another CBO report indicated that Medicare beneficiaries tend to use less expensive drugs within a therapeutic class than do Medicaid beneficiaries.41 Medicare beneficiaries also tend to use generic drugs to a slightly greater degree than Medicaid beneficiaries do. The remaining difference in retail prices stems from differences in average prices for the sets of brand-name and generic drugs used by the two populations.

Comparisons With Commercial Plans. Although CBO does not have data on per capita drug use or average prices for the rest of the U.S. population, it is plausible that both average use and average retail prices for that group are between the Medicare and Medicaid averages. The average age of people without public insurance (most of whom have commercial insurance) and the average health status of that population are probably between those of the Medicaid population (which has a large proportion of younger parents and children) and the Medicare population (which consists mostly of elderly people and disabled people). In addition, the fact that the nationwide share of prescriptions for generic drugs is about the same as the share of such prescriptions in Medicare suggests that the generic share among commercially insured people is closer to Medicare’s percentage (90 percent) and higher than Medicaid’s percentage (87 percent).

Average net prices for commercial plans most likely differ from net prices in Medicare and Medicaid not only because of differences in the set of drugs those plans’ enrollees take, but also because of differences in how those prices are determined. In CBO’s assessment, retail prices for a given drug are generally similar across payers. However, the net price of a given drug is generally lower for Medicaid than for commercial plans or for Medicare Part D because of rebates that manufacturers are required to pay to Medicaid plans—particularly for brand-name drugs. Medicaid programs are entitled by law to receive the greater of 23.1 percent of the drug’s average manufacturer price (AMP) or the largest rebate that the manufacturer gives to any payer (excluding certain government programs, such as Medicare Part D) for all brand-name drugs.42 The rebate also includes an additional component that is determined on the basis of growth in the drug’s AMP relative to inflation.43 For generic drugs, Medicaid’s rebate includes a basic rebate (equal to 13.0 percent of the drug’s AMP) and the same inflation-based component as Medicaid’s rebate for brand-name drugs.44

In CBO’s judgment, the average net price of a given brand-name drug is probably lower in Medicare Part D than in commercial plans. The Medicare Part D benefit is administered by private insurers who also provide commercial insurance plans, and rebate negotiations by both types of plans are handled in similar fashion, perhaps even by the same negotiators. Those negotiations operate under the same general set of incentives for both Part D plans and private plans in that the availability of therapeutic alternatives—and a plan’s willingness to steer enrollees toward a particular drug—increases the amount that manufacturers are willing to provide to that plan in the form of a rebate (see Box 1).

The primary difference in bargaining leverage between commercial plans and Medicare Part D plans is that the largest rebate that a manufacturer offers to a commercial plan also has to be made available to Medicaid, which reduces the size of the largest rebate that manufacturers would otherwise be willing to pay. In contrast, rebates that manufacturers pay to Part D plans do not directly affect Medicaid prices. Therefore, manufacturers may be willing to pay larger rebates to Part D plans than to commercial insurance plans.

However, another difference is that Part D plans must, by statute, cover all available drugs in each of six “protected” therapeutic classes.45 That puts Part D plans at a disadvantage when negotiating rebates on those drugs because their pharmacy benefit managers (PBMs) cannot credibly threaten to exclude such drugs from the Part D formularies they manage if the manufacturer does not offer a sufficient rebate.46 (PBMs are intermediaries who manage prescription drug benefits for insurance plans or other payers.) Although there may be some variation in the extent of price differences for a given drug across therapeutic classes, CBO believes that the link between Medicaid’s “best price” requirements and the rebates paid to commercial plans outweighs the impact of the protected classes in Part D because the best price requirement applies to all brand-name drugs.

Prices for Brand-Name Drugs

Net prices for brand-name drugs reflect the competitive landscape for a given drug. In cases in which therapeutic alternatives are limited, the manufacturer tends to have greater leverage, particularly for drugs that offer larger benefits to patients than other treatment options. In those cases, manufacturers have considerable monopoly power to exercise, particularly given that insured patients often pay a small share of the total price of a brand-name drug and that plans may feel considerable pressure to cover those drugs in order to retain market share. Similarly, as employers make decisions about the generosity of their employment-based plans, they may feel pressure to provide coverage for such drugs in order to retain employees. In situations in which there are therapeutic alternatives, payers and PBMs tend to have greater leverage to negotiate for lower net prices. In those cases, net prices would probably be set lower—or grow more slowly—because manufacturers typically accept lower prices in exchange for greater formulary access (or reduced formulary access for their competitors).

Since 2008, Medicare and Medicaid have both experienced substantial increases in the prices they pay for brand-name drugs. For Medicare Part D, the average net price of a prescription for a brand-name drug (measured in 2018 dollars) more than doubled between 2009 and 2018, from $149 to $353. For Medicaid, the average net price increased by about 50 percent, from $147 to $218 (see Figure 5). Retail prices have increased even more dramatically. Between 2009 and 2018, the average retail price of a prescription for a brand-name drug in Medicare Part D more than tripled, from $175 to $547. For Medicaid, the average retail price more than doubled, from $262 to $584. (For more detail on the divergence between retail and net prices, see Box 2.)

Factors Underlying Rapid Increases in Prices for Brand-Name Drugs. Growth in average prices reflects a combination of several factors. For example, the composition of brand-name prescriptions that people fill has shifted from less expensive drugs toward more expensive drugs. One key factor in the shift toward more expensive drugs is that newer drugs tend to be more expensive than older drugs. In addition, prices for drugs already on the market tend to grow faster than inflation. Finally, federal policies influence manufacturers’ pricing strategies and lead to higher prices in certain circumstances.

The Role of Launch Prices. Newer drugs are often launched at higher prices than those paid for drugs currently on the market. For example, in the Medicare Part D program, the average net price in 2015 for brand-name drugs that were launched after 2010 was nearly four times the average net price for brand-name drugs already on the market in 2010. And in 2017, the average net price for new drugs launched after 2015 was 12 times the average net price for brand-name drugs already on the market in 2015.47 As a result, the combination of take-up of new brand-name drugs and shifting from older brand-name drugs to generic alternatives has led to higher average prices for brand-name drugs.

The phenomenon of increasingly high launch prices for new drugs is partly driven by the rising number of specialty drugs. Specialty drugs tend to be more complex to develop and manufacture than nonspecialty drugs, and they generally have much higher prices because of the larger benefits to health and well-being that they tend to confer on their patients. In 2015, they accounted for 78 percent of spending on new drugs (launched after 2010) in Medicare Part D and 8 percent of prescriptions for new drugs. In 2017, specialty drugs accounted for 88 percent of that spending and 39 percent of prescriptions for new drugs (those launched after 2015). Although prices for a given drug vary across payers, the rising influence of specialty drugs on spending—and therefore on usage-weighted average prices—probably plays a role for all payers. An industry report supports that idea, noting that specialty drugs represented the largest share of new drugs over the 2014–2018 period and that specialty drugs have “far higher” costs per patient than nonspecialty drugs.48

The Role of Price Growth. Another key component of the growth in average net prices for brand-name drugs is year-over-year price growth for a given drug, though the importance of that factor may differ substantially among payers. Using a price index approach, CBO found that net prices for brand-name drugs increased by an average of 6.3 percent per year from 2010 to 2017 in the Medicare Part D program, after removing the effects of general inflation.49 In contrast, using a similar approach, recent research on nationwide drug spending patterns found that nominal net prices (that is, net prices with no adjustments for inflation) for brand-name drugs grew by an average of only 3 percent per year over the 2012–2017 period. CBO estimates that nominal price growth for brand-name drugs in Medicare Part D was an average of 7.3 percent per year over the 2012–2017 period, although the two calculations are not directly comparable because they consist of somewhat different sets of drugs. In addition, the net prices in the nationwide study are net of all discounts along the supply chain—including discounts provided to distributors or directly to patients as well as rebates that payers received—whereas the net prices in CBO’s analysis account only for the rebates that payers received.50

However, the large difference does suggest that prices paid by Part D over the 2012–2017 period grew more quickly than the prices paid by other payers, on average. That difference may have been driven by slower price growth in Medicaid, stemming from the statutory rebates that Medicaid receives. In addition, enrollees in commercial insurance plans may have been more likely to use drugs that face therapeutic competition than enrollees in Part D, which may have also led to slower growth in prices paid by commercial plans.

The Role of Federal Policies. Federal policies may have contributed to the growth in drug prices. Medicaid is entitled to the largest rebate that a manufacturer provides to any payer. (That requirement does not apply to the rebates provided to certain government programs, such as Medicare Part D.) That probably has increased average net prices for commercial payers more broadly. Similarly, the additional statutory rebate provided to Medicaid for drugs whose retail prices rise faster than inflation may have contributed to the higher launch prices of new drugs.51 However, that rebate requirement may also have dampened price growth over time; as a result, average prices may have approached the level they would have without the policy.

Likewise, Medicare’s minimum formulary requirements may contribute to higher net drug prices in that program. In addition to the requirement that Part D plans cover all drugs in the six protected classes, plans are also required to cover at least two drugs in all other therapeutic classes. Those requirements diminish the leverage that PBMs bring to their negotiations with manufacturers over drug prices for Part D plans.

Manufacturers may also feel less pressure to constrain prices because of the increase in the share of overall drug spending that is covered by insurers. That reduction in consumers’ exposure to high prices increases their willingness to purchase high-priced drugs. Manufacturers may interpret that phenomenon as a higher overall “willingness to pay” and set higher prices, or increase their prices faster, as a result. Policymakers and stakeholders have pointed to the high and rising prices of prescription drugs as a reason to question whether the profits of brand-name drug manufacturers are excessive. (For a brief overview of the profitability of the pharmaceutical industry, see Box 3.)

Factors That Constrain Growth in the Prices of Brand-Name Drugs. The availability of therapeutic substitutes provides insurance plans and PBMs with leverage to negotiate lower prices. When alternatives are limited, such as when a new drug is the first to treat a particular condition, then insurance plans and PBMs have limited leverage to negotiate lower prices. As competing products enter the market, payers gain the flexibility to exclude a given drug or to limit patients’ use of that drug through higher cost sharing or other utilization management tools.

Consolidation within the pharmacy-benefit-management industry has also increased the leverage that PBMs wield in negotiating on behalf of their client insurers. In the 1980s, the industry consisted of many small PBMs, which, for the most part, simply processed pharmacy claims. Since then, the industry has evolved and now develops and administers many insurance plans’ drug formularies. In addition, the PBM industry has come to be dominated by just a few large firms. A 2015 merger between the third- and fourth-largest firms left the three largest PBMs handling more than 70 percent of the industry’s prescription drug claims. That market power helps those large PBMs negotiate bigger rebates and steeper discounts because they represent a larger share of the patient population. However, their market power remains limited when negotiating net prices of drugs without direct substitutes.

Nevertheless, greater consolidation within the PBM industry also has implications for plans’ costs. For example, PBMs probably use their increased leverage when negotiating with insurance plans over contract terms to charge higher fees to plans (which could take the form of PBMs keeping a larger fraction of rebates). The insurance plans then pass some share of those higher costs on to consumers in the form of higher premiums. In addition, over time, there has been an increase in consolidation between PBMs and insurers, as well as between PBMs and pharmacies, including mail-order and specialty pharmacies. Although that type of consolidation can increase efficiency and reduce plans’ costs, the extent to which reduced costs are passed on to enrollees depends on the amount of competition in the market. In addition, to the extent that insurers, PBMs, and pharmacies negotiate with entities outside their ownership group, there may be an incentive to charge higher fees to rival companies.52

Prices for Generic Drugs

As with brand-name drugs, the competitive landscape for generic drugs has a great deal of influence over prices for those drugs. Unlike brand-name drugs, however, generic drugs often face direct competition because the same drug is often manufactured by several companies. Although new generic drugs may have higher prices when they first enter the market, those prices tend to fall as competitors enter the market.

As a result, in contrast with the growth observed in the average prices of brand-name drugs, average prices of generic drugs have tended to fall in real terms in recent years. In Medicare Part D, the average price of a generic prescription was $22 in 2009 and gradually fell to $17 in 2018.53 For Medicaid, the average price of a generic prescription fell from $27 in 2009 to $23 in 2018 (see Figure 6). A 2016 federal review found that, from mid-2013 to mid-2014, nearly 65 percent of prescriptions for generic drugs in the Medicaid program were for drugs whose prices had declined over that time, even before accounting for rebates to the Medicaid program. That review, by the Department of Health and Human Services (HHS), also cited a finding from Express Scripts, a PBM: Specifically, Express Scripts indicated that for the top 80 percent of generic drugs (by unit sales), the average prescription price fell in 2014 by 20 percent. HHS concluded that generic prices are not a key driver of high spending on prescription drugs.54 Other recent evidence suggests that year-over-year prices for generic drugs have fallen in the past several years, on average.55

Even so, price increases for some generic drugs have raised concerns for policymakers. For example, there are ongoing criminal and civil proceedings related to anticompetitive behavior that led to higher prices.56 In addition, HHS has found that for about one-fourth of generic drugs with the highest spending in the Medicaid program, prices in recent years have increased faster than the rate of inflation (before accounting for Medicaid’s rebates). However, HHS also found that the set of drugs with particularly large price increases represents a very small share of the market and that those price increases did not have a sizable impact on overall spending.57 More generally, increases in the prices of specific generic drugs—whether facilitated by anticompetitive behavior or by the exercise of market power, or simply reflecting shifts in supply or demand—affect total drug spending only to the extent that those drugs are sold in large quantities. Nevertheless, those price increases could have a substantial impact on consumers who use drugs that have increased in price, especially if those consumers’ out-of-pocket costs are based on the price of the drug.

---

About This Document

This report was prepared at the request of the Chairman of the Senate Committee on Finance. In keeping with the Congressional Budget Office’s mandate to provide objective, impartial analysis, the report makes no recommendations.

Tamara Hayford and David Austin prepared the report with guidance from Joseph Kile, Lyle Nelson (formerly of CBO), and Julie Topoleski. Ru Ding contributed to the analysis. Anna Anderson-Cook (formerly of CBO), Colin Baker, Elizabeth Bass, Julia Christensen, Ryan Greenfield, Stuart Hammond, Lara Robillard, Asha Saavoss, Ellen Werble, Chapin White, and Kate Young provided useful comments. Joshua Varcie fact-checked the report.

Amitabh Chandra of the Kennedy School of Government and Harvard Business School, Sean Keehan of the Centers for Medicare & Medicaid Services’ Office of the Actuary, Chris Park of the Medicaid and CHIP Payment and Access Commission, and Rachel Schmidt and Shinobu Suzuki of the Medicare Payment Advisory Commission commented on an earlier draft. The assistance of external reviewers implies no responsibility for the final product; that responsibility rests solely with CBO.

Mark Doms, Jeffrey Kling, and Robert Sunshine reviewed the report. Loretta Lettner edited it, and Jorge Salazar created the graphics and prepared the text for publication. The report is available at www.cbo.gov/publication/57050.

CBO seeks feedback to make its work as useful as possible. Please send comments to communications@cbo.gov.

Phillip L. Swagel

Director

January 2022