US HEALTH REFORM-MONITORING AND IMPACT Support for this research was provided by the Robert Wood Johnson Foundation. The views expressed here do not necessarily reflect the views of the Foundation. Capping Medicare Beneficiary Part D Spending at $2,000 Who Would It Help and How Much? Anuj Gangopadhyaya and Bowen Garrett March 2022 Introduction The Build Back Better Act, passed by the US House of Representatives in November 2021, included several provisions to lower high prescription drug costs.1 These provisions included allowing the federal government to negotiate lower prices for certain high-cost drugs in Medicare Parts B and D, limiting annual increases in drug prices in Medicare and private insurance, and capping out-of-pocket (OOP) spending on prescription drugs for Medicare Part D enrollees at $2,000. The Congressional Budget Office estimates that these drug pricing provisions could reduce the federal budget deficit by nearly $297 billion from 2022 to 2031.2 In this brief, we focus on the $2,000 cap on OOP spending for Medicare Part D enrollees, examining the number of people who might benefit from the provision and their characteristics, the resulting amount of savings for affected beneficiaries, and the provision's effects on Medicare spending. The cap on beneficiaries' Part D spending has also been proposed as a key feature of the Capping Drug Costs for Seniors Act of 2021, which three House members introduced in July 2021. Although the Build Back Better Act is unlikely to be passed in its current form, the cap and other drug provisions could remain in a potential slimmed-down budget reconciliation bill that might pass in both the House and the Senate. Overall, the Build Back Better Act's drug provisions could generate savings, and, as we discuss below, the $2,000 OOP cap would add little to new government spending.3 Such a cap would benefit enrollees who do not qualify for Part D cost-sharing protections under the Low-Income Subsidy (LIS) program, who we refer to as "non-LIS enrollees" throughout. We find the following:  In 2019, about 866,000 non-LIS Part D enrollees would have benefitted from the introduction of a $2,000 OOP spending cap for prescription drugs. This represents a small share of the 32.8 million non-LIS enrollees in 2019.  These enrollees have average total drug expenditures of about $19,800, of which they spent about $2,900 out of pocket. Thus, enrollees with spending above the proposed cap would save $900 on average.  Altogether, we show that a $2,000 OOP spending cap would have raised Part D expenditures by approximately $782 million in 2019. Because enrollee premiums finance 25.5 percent of the Part D program, we estimate that the spending cap would have required an additional $199 million in beneficiary premium spending and $583 million in Medicare program spending in 2019. This is equivalent to raising the premium for the standard benefit across the 45.8 million beneficiaries in 2019 by $4.35 annually and raising Medicare Part D expenditures by less than 1 percent of the $97.6 billion total Part D expenditures that year. The number of affected enrollees is, perhaps surprisingly, small because a large share of enrollees with high total drug expenditures already qualify for extensive benefits under the LIS program. Still, we find that among enrollees with Part D spending above the proposed $2,000 cap, about 30 percent had Part D spending above $3,000 and would receive substantial savings from this added protection. About US Health Reform-Monitoring and Impact With support from the Robert Wood Johnson Foundation, the Urban Institute has undertaken US Health Reform-Monitoring and Impact, a comprehensive monitoring and tracking project examining the implementation and effects of health reforms. Since May 2011, Urban Institute researchers have documented changes to the implementation of national health reforms to help states, researchers, and policymakers learn from the process as it unfolds. The publications developed as part of this ongoing project can be found on both the Robert Wood Johnson Foundation's and Urban Institute Health Policy Center's websites. Background on Medicare Part D The Part D program was created under the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 and first implemented in 2006. Enrollee premiums finance 25.5 percent of the Part D program's costs, and the Supplementary Medical Insurance Trust Fund, funded through general federal spending, finances the remaining 74.5 percent. The Part D benefit is available to both enrollees with traditional, or fee-for-service (FFS), coverage and beneficiaries in Medicare Advantage (MA). Prospective Part D enrollees select private plans that are in turn reimbursed by Medicare; enrollees can either select a standalone plan or enroll in an MA plan that coordinates with a Part D plan 2 CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 (an MA-PDP). Beneficiaries with low incomes can qualify for Part D's LIS program, which provides support for Part D premiums, deductibles, and cost sharing. This brief focuses on enrollees who do not qualify for the LIS program but may benefit from an OOP spending cap. Under the program, enrollees first pay a deductible and then pay 25 percent of all of their drug expenses until their total drug spending crosses a threshold that initiates the coverage gap, formerly called the "donut hole." Originally, enrollees were fully responsible for their drug expenditures in the coverage gap until they reached a subsequent threshold (based on total OOP drug expenditures) that initiated the catastrophic range. Though the coverage gap was closed in 2019 and no longer exists from the perspective of enrollees, it still plays a critical role in the program in that it initiates requirements that drug manufacturers provide discount payments for brand-name drugs to help pay for the cost of services (i.e., until a beneficiary reaches the catastrophic range).4 In the catastrophic range, enrollees pay the greater of a 5 percent coinsurance or a small nominal copayment for prescription medicines, and Medicare covers the remaining drug costs. Each year, the Medicare program specifies the standard benefit's deductible and spending thresholds for the coverage gap and catastrophic range. However, the Part D program comprises privately run Part D plans, and these plans can adjust cost sharing and deductibles so long as average payments for enrollees are the same as would be expected under the standard benefit.5 The cost-sharing schedule is greatly simplified for LIS enrollees; they face a small copayment for prescription medicines until they reach the OOP threshold, and state Medicaid plans and the Medicare program pay for the remaining costs. Of the 61.5 million Medicare enrollees in 2019, 45.8 million received prescription drug benefits through Part D. Among Part D enrollees, 32.8 million (72 percent) were non-LIS enrollees who would be protected by an OOP spending cap for prescription drugs.6 For the standard benefit in 2019, non-LIS Part D enrollees faced a $415 deductible, reached the coverage gap when total drug spending reached $3,820, and reached the catastrophic range when the sum of total OOP spending and manufacturer discounts reached $5,100. Using 2019 data from the Part D denominator file, Cubanski, Neuman, and Damico (2021) estimated that 1.2 million non-LIS Part D enrollees had OOP drug spending exceeding $2,000 that year and would therefore likely benefit from a spending cap. The authors identified 154 drugs with annual average OOP costs exceeding $2,000. Many of these drugs were relatively rarely used, but the authors noted that the more frequently used of those drugs are for treating cancer, multiple sclerosis, lung disease, and hepatitis C. Our analysis provides new evidence on the number of non-LIS enrollees who may benefit from a $2,000 OOP spending cap, their spending on Part D services, their demographic characteristics, and their overall and OOP spending for other services covered under Medicare Parts A and B. Data and Methods We use the Urban Institute's Medicare policy simulation model, MCARE-SIM, to estimate 2019 Medicare and OOP spending for Part D services among non-LIS beneficiaries. MCARE-SIM uses data CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 3 from the 2015–2018 Medicare Current Beneficiary Survey (MCBS) and projects Medicare enrollment and spending estimates to 2019. The MCBS provides nationwide information on demographic characteristics, use of medical services, medical expenditures, health status, access to health care, and sources of supplemental insurance coverage for Medicare enrollees. We also use MCARE-SIM to describe the demographic characteristics and Parts A and B spending patterns of non-LIS enrollees with Part D OOP spending above and at or below $2,000. Though the MCBS reports the costs and use of Part D services and demographic characteristics for all enrollees, analyses of Parts A and B spending are limited to enrollees with FFS Medicare coverage. We calculate 2019 Part D spending and enrollment for non-LIS enrollees in three steps. We first reweight MCBS respondents to match non-LIS Part D enrollee counts reported in the 2019 Centers for Medicare & Medicaid Services Program Statistics.7 Second, we adjust MCBS Part D Medicare spending totals to align with per capita averages reported in the Medicare Trustees report;8 these adjustments account for differences in both drug utilization patterns and drug prices between each survey year and 2019. Third, given each enrollee's reported total Medicare spending, we calculate OOP spending and manufacturer discounts based on Part D cost-sharing rules for a standard plan in 2019. That is, we apply the 2019 average deductible ($415) and then 25 percent cost sharing to all drug spending up to the catastrophic coverage threshold. If an enrollee reaches the coverage gap,9 they continue to pay 25 percent cost sharing on all prescription medicines. However, we allocate the remaining spending to a mix of Medicare spending and drug manufacturer discounts determined by an individual's spending on brand-name drugs as a share of total prescription drug spending. In 2019, the catastrophic range began once the sum of an enrollee's OOP spending and drug manufacturer discounts reached $5,100. When enrollees reach the catastrophic range, we apply 5 percent cost sharing to all further drug expenditures and allocate the remaining 95 percent of costs to the Medicare program. We follow a similar approach to model FFS spending for Parts A and B services. We adjust MCBS- reported Medicare payments to match 2019 per capita spending for the following spending categories (each of which is reported for FFS enrollees in the Centers for Medicare & Medicaid Services Program Statistics):10 inpatient, skilled nursing facilities, hospice care, home health (reported separately for Parts A and B), and medical providers and other medical equipment. From Medicare totals, we impute all other spending based on 2019 program rules. For enrollees identifying supplementary coverage (i.e., Medigap, employer-sponsored insurance coverage, or Medicaid), we allocate all beneficiary spending to the secondary plan. For all other FFS enrollees, we allocate non-Medicare spending to OOP spending. Our approach has some limitations. First, our analysis does not incorporate behavioral responses to the introduction of an OOP spending cap for prescription medicines. An OOP spending cap may reduce the average price of drugs for beneficiaries with high spending, and this price reduction may lead beneficiaries to use more drugs than they otherwise would. Thus, we likely underestimate both the number of beneficiaries who would reach the $2,000 spending cap and the average reduction in OOP costs among those reaching the cap if this policy is enacted. Second, our estimate is based on 2019 Part D deductible and cost-sharing parameters. One provision of the Affordable Care Act limited the growth in OOP spending required to reach the catastrophic coverage range ($5,100 in 2019). However, this 4 CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 provision expired in 2020. Consequently, the OOP spending threshold increased substantially, and the rate of the increase persisted over subsequent years; the threshold was $6,350 in 2020 but has increased to $7,050 in 2022. Increasing the amount of OOP spending required to reach the catastrophic range will expose people with high drug spending to the higher 25 percent coinsurance rate for a larger share of their expenditures. Thus, this change could also result in many more enrollees reaching the proposed $2,000 OOP spending cap than our 2019 estimate shows. For these reasons, our 2019 estimates represent a lower bound of the number of enrollees who could benefit from a $2,000 OOP spending cap introduced in 2022 or later and a lower bound of the amount of Medicare spending required to protect beneficiaries.11 Findings Table 1 reports Part D spending amounts among non-LIS enrollees with OOP spending above the proposed $2,000 cap and with OOP spending at or below the cap. An estimated 866,000 Part D enrollees incurred more than $2,000 in OOP drug costs in 2019. (No LIS enrollees incurred OOP spending above $2,000, and we exclude such enrollees from the analysis.) Among those with spending exceeding the cap, total spending for Part D–covered drugs was about $19,800 on average. Of that total, the majority was Medicare program spending ($14,100), $2,700 represented manufacturer discounts incurred in the coverage gap that apply to spending toward the catastrophic range, and $2,900 was OOP spending. Forty percent of beneficiaries with more than $2,000 in OOP spending spent $2,000 to $2,500, 30 percent spent $2,500 to $3,000, 12 percent spent $3,000 to $3,500, and 18 percent spent more than $3,500. Thus, significant shares of enrollees in this high-spending group had Part D spending far exceeding $2,000 and would benefit substantially from the proposed cap. In contrast, most Part D non-LIS enrollees, more than 31.9. million, had OOP spending of $2,000 or less. Average total spending for these enrollees was around $800, of which about $400 was Medicare program spending and $500 was OOP spending (estimates do not sum to total because of rounding). More non-LIS Part D enrollees with high OOP spending are enrolled in FFS than in MA. Of the enrollees with more than $2,000 in OOP spending, 527,000 were enrolled in FFS Medicare and 339,000 were enrolled in MA; this split is consistent with the overall distribution of FFS and MA enrollees in Medicare in 2019. Total spending for covered prescription drugs was $21,500 for FFS enrollees and $17,000 for MA enrollees. On average, OOP spending was somewhat higher for FFS enrollees than MA enrollees in the high-spending group ($3,000 versus $2,800). Among enrollees with more than $2,000 in OOP spending, the share with spending exceeding $3,500 was higher among FFS enrollees than among MA enrollees (22 versus 11 percent). Table 1 indicates that among enrollees with OOP spending above the proposed cap, the average amount of OOP spending exceeding the cap was $900. Given that an estimated 866,000 beneficiaries had OOP spending above the proposed cap, implementing a spending cap in 2019 would have cost approximately $782 million. Because enrollee premiums finance 25.5 percent of the Part D program, we estimate that the $2,000 OOP spending cap would have required an additional $199 million in CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 5 beneficiary premium spending and an additional $583 million in Medicare program spending in 2019 (data not shown). For context, the change to beneficiary premiums is equal to raising the premium for the standard benefit across the 45.8 million beneficiaries in 2019 by $4.35 annually,12 and the change in Medicare program spending represents less than 1 percent of the $97.6 billion in total Part D expenditures that year.13 TABLE 1 Part D Spending Characteristics among Non-LIS Enrollees with Out-of-Pocket Part D Spending above and at or below $2,000 All Non-LIS Part D Fee-For-Service Medicare Advantage Enrollees Enrollees Enrollees Part D Part D Part D Part D Part D Part D OOP OOP OOP OOP OOP OOP spending spending of spending spending of spending spending of > $2,000 $0–2,000 > $2,000 $0–2,000 > $2,000 $0–2,000 Estimated number of enrollees in 2019 866,000 31,900,000 527,000 15,900,000 339,000 15,900,000 Average Part D spending ($) Total 19,800 800 21,500 900 17,000 800 Medicare 14,100 400 15,800 400 11,500 400 Manufacturer discounts 2,700 0 2,700 0 2,700 0 Out-of-pocket 2,900 500 3,000 500 2,800 400 Share with the following out-of-pocket costs (%) $2,000–2,500 40 42 37 $2,500–3,000 30 25 39 $3,000–3,500 12 11 13 Greater than $3,500 18 22 11 Source: MCARE-SIM 2019 estimates using 2015–18 Medicare Current Beneficiary Survey data. Notes: LIS = Low-Income Subsidy. OOP = out-of-pocket. Blank rows indicate the column heading does not apply. OOP spending does not include Part D premium payments. Estimates may not sum to total spending because of rounding. Table 2 shows the demographic, employment, and income characteristics of non-LIS enrollees with Part D OOP spending above $2,000 and at or below the proposed cap. The average age of enrollees is 70.9 among those with OOP spending above $2,000, compared with 73.8 for enrollees with spending at or below $2,000. The share of enrollees ages 45 to 64 (who qualify for Medicare through disability or end-stage renal disease) in the high-spending group was 18.4 percent, compared with 5.7 percent in the low-spending group. Enrollees in the high-spending group were significantly less likely to be in the higher age groups (ages 80 to 84 and ages 85 and older) than those in the low-spending group. 6 CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 TABLE 2 Demographic, Employment, and Income Characteristics of Non-LIS Enrollees with Part D Out-of- Pocket Spending above and at or below $2,000 Percent Part D OOP spending Part D OOP spending of > $2,000 $0–2,000 Age and gender Age 70.9* 73.8 < 45 1.4* 0.5 45–64 18.4* 5.7 65–69 19.5* 25.8 70–74 30.3 26.2 75–79 16.7 18.4 80–84 7.2* 12.1 85 or greater 6.6* 11.4 Male 50.0* 43.8 Race or ethnicity Non-Hispanic white 82.0 82.5 Non-Hispanic Black 3.0* 6.3 Hispanic 9.0* 6.4 Non-Hispanic, other race 6.0 4.7 Marital status Married 56.3 60.0 Divorced 12.9 13.7 Widowed 22.0 21.9 Never married 8.8* 4.4 Education Less than high school diploma 10.2 10.8 High school diploma 25.8 26.5 Some college 24.9 23.8 Bachelor's degree or more 31.7 31.1 Work and income Employed 9.1* 15.4 Income < FPL 5.1 5.5 100–200% of FPL 22.4 24.2 200–400% of FPL 33.7 33.9 > 400% of FPL 38.9 36.6 Source: MCARE-SIM 2019 estimates using 2015–18 Medicare Current Beneficiary Survey data. Notes: LIS = Low-Income Subsidy. OOP = out-of-pocket. FPL = federal poverty level. OOP spending does not include Part D premium payments. * Estimated differences in characteristics between enrollees with Part D OOP payments above versus at or below $2,000 are significant at the p < 0.05 level. CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 7 Differences in the racial and ethnic composition of the two spending groups are not large; non- Hispanic Black enrollees were less likely to be in the high-spending group than the low-spending group (3.0 versus 6.3 percent), and Hispanic enrollees were more likely to be in the high-spending group than the low-spending group (9.0 versus 6.4 percent).14 We find no substantial or statistically significant differences in the composition of the two spending groups by educational attainment or income level. Enrollees in the high-spending group were more likely to never have been married (8.8 versus 4.4 percent) and less likely to be employed (9.1 versus 15.4 percent). Aside from the differences described above, the overall pattern suggests all demographic and income groups are at risk of experiencing high OOP costs for prescription drugs. Table 3 shows what non-LIS Medicare enrollees spent in total and out of pocket for all Medicare services in 2019. The table also shows how total Medicare spending for the high- and low-spending groups was allocated across the various parts of Medicare. The OOP amounts are also shown in figure 1. Enrollees who spent more than $2,000 out of pocket for Part D services had total Medicare spending of $51,500, of which $7,100 was paid out of pocket. In contrast, enrollees who spent less than $2,000 out of pocket on Part D services spent far less ($13,800 in total and $2,100 OOP). Thus, enrollees with high Part D spending also had higher spending and incurred higher OOP costs in other parts of Medicare. Part A spending, which largely represents hospital services, was higher for enrollees with high drug spending than for those with low drug spending ($8,700 versus $5,200 total, $400 versus $300 OOP). Part B spending, which largely represents physician services, was more than 2.5 times higher for those with high drug spending than for those with low drug spending ($21,300 versus $7,700 in total, $3,700 versus $1,300 OOP). Capping OOP costs for those with high prescription drug spending would ease the financial burden for enrollees with very substantial OOP costs not only for prescription drugs but also for other services (primarily Part B physician services). TABLE 3 Total and Out-of-Pocket Spending for Parts A, B, and D Services among Non-LIS Enrollees with Part D Out-of-Pocket Spending above and at or below $2,000 Among fee-for-service enrollees only Part D OOP spending Part D OOP spending of > $2,000 $0–2,000 Total OOP Total OOP spending spending spending spending ($) ($) ($) ($) Total Parts A, B, and D spending 51,500 7,100 13,800 2,100 Total Part A spending 8,700 400 5,200 300 Total Part B spending 21,300 3,700 7,700 1,300 Total Part D spending 21,500 3,000 900 500 Source: MCARE-SIM 2019 estimates using 2015–18 Medicare Current Beneficiary Survey data. Notes: LIS = Low-Income Subsidy. OOP = out-of-pocket. OOP spending does not include Part D premium payments. Estimates exclude LIS enrollees and enrollees in Medicare Advantage plans. Estimates have been rounded. 8 CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 FIGURE 1 Out-of-Pocket Spending for Medicare-Covered Services among Non-LIS Enrollees with Part D Out- of-Pocket Spending above and at or below $2,000 Among fee-for-service enrollees only Part D OOP > $2,000 Part D OOP of $0–2,000 $7,100 $3,700 $3,000 $2,100 $1,300 $400 $300 $500 Part A Part B Part D Total URBAN INSTITUTE Source: MCARE-SIM 2019 estimates using 2015–18 Medicare Current Beneficiary Survey data. Notes: LIS = Low-Income Subsidy. OOP = out-of-pocket. Estimates have been rounded. We conclude by presenting the estimated number of Part D enrollees who would have benefitted from alternative OOP caps of $1,000, $1,500, $2,500, and $3,000 in 2019 (table 4). We also show the average amount of spending above each of the proposed caps for these beneficiaries, the total cost of drug expenditures above each spending cap (i.e., the total amount of additional program and beneficiary premium spending required), and the implied change in annual Part D premiums for the standard benefit that enrollees will face assuming premiums continue financing 25.5 percent of program costs. In 2019, an estimated 1.3 million non-LIS enrollees had OOP spending above $1,000 and would therefore have benefitted from a spending cap at this amount. Average spending above a $1,000 cap would have been $1,300 that year. Altogether, this means a $1,000 OOP cap would have cost about $1.75 billion, or more than twice the cost of a $2,000 cap. To help finance this, beneficiaries would have spent about an additional $10 annually on Part D premiums. A $3,000 spending cap, on the other hand, would only have benefitted an estimated 255,000 enrollees in 2019. The estimated total cost would be about $226 million, which would have increased average premiums by slightly more than $1 annually. CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 9 TABLE 4 Estimated Number of Non-LIS Enrollees Affected and Associated Program Costs under Different Part D Out-of-Pocket Caps Number of Change in average beneficiaries with Average OOP annual per capita Part D OOP OOP spending spending above Part D premium spending cap ($) above cap cap ($) Total costs ($) spending ($) 1,000 1,322,000 1,300 1,747,000,000 9.72 1,500 922,000 1,300 1,231,000,000 6.85 2,000 866,000 900 782,000,000 4.35 2,500 516,000 800 400,000,000 2.23 3,000 255,000 900 226,000,000 1.26 Source: MCARE-SIM 2019 estimates using 2015–18 Medicare Current Beneficiary Survey data. Notes: LIS = Low-Income Subsidy. OOP = out-of-pocket. OOP spending does not include Part D premium payments. Estimates exclude LIS enrollees. The product of estimates may not equal the total reported in the fourth column because of rounding. The fifth column reports per capita premium rates based on (a) maintaining Part D financing, such that beneficiary premiums pay for 25.5 percent of program costs, and (b) Centers for Medicare & Medicaid Services Program Statistics reporting 45.8 million Part D enrollees in 2019. Discussion We estimate that about 866,000 non-LIS Part D enrollees would have benefitted from the introduction of a $2,000 OOP spending cap in 2019. This estimate is slightly smaller than the 1.2 million reported in Cubanski, Neuman, and Damico (2021). However, both estimates lead to similar conclusions: a $2,000 OOP spending cap in Medicare Part D would benefit a small share of Part D enrollees, and this small impact would likely carry a small price tag for the Medicare program (an estimated $779 million in 2019). Even more generous OOP caps would not affect a substantial number of beneficiaries; we estimate that lowering the OOP spending cap to $1,000 in 2019 would have (a) benefitted 1.3 million people overall (450,000 more than under a $2,000 cap), (b) saved the average beneficiary with OOP spending exceeding the $1,000 cap $1,300 (compared with $900 under a $2,000 cap), and (c) increased total Part D expenditures by $1.75 billion (compared with $782 million under the $2,000 cap). Finally, as Cubanski, Neuman, and Damico (2021) showed, patients with higher drug spending are more likely to have chronic conditions that will require sustained use of prescription medicines. Consequently, enrollees who incur high OOP drug spending in a given year will likely incur similar or greater expenses in subsequent years. A spending cap would help protect individuals and their families from withstanding high drug payments year after year. In doing so, a spending cap would enhance the overall benefit of the Medicare Part D program without substantially raising the program's costs. 10 CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 Notes 1 Build Back Better Act, H.R. 5376 (2021). 2 "Estimated Budgetary Effects of Title XIII, Committee on Ways and Means, H.R. 5376, the Build Back Better Act," Congressional Budget Office, November 18, 2021, https://www.cbo.gov/publication/57626. 3 "Budgetary Effects of Title XIII, Committee on Ways and Means, the Build Back Better Act," Congressional Budget Office. 4 The Affordable Care Act included a provision to gradually close the coverage gap and to reduce enrollee cost sharing in the coverage gap to 25 percent by 2020. In 2018, the Bipartisan Budget Act accelerated the closing of the coverage gap; consequently, cost sharing in the catastrophic range had reached 25 percent by 2019, and the remaining drug costs in this range were covered by either the Medicare program or discounts provided by drug manufacturers. See KFF (2021). For generic drugs in the coverage gap, the Medicare program pays for the remaining 75 percent. For brand-name drugs, the legislation required drug manufacturers to provide discounts covering 70 percent of drug spending in the coverage gap, and the Medicare program then pays the remining 5 percent. Importantly, manufacturer discounts were deemed to also count as OOP spending toward the catastrophic range. Consequently, the rate at which enrollees reach the catastrophic range has increased over time. 5 For an overview of the early Part D design and implementation period, see Duggan, Healy, and Morton (2008). 6 Build Back Better Act. 7 See "Medicare Enrollment Section," Centers for Medicare & Medicaid Services, accessed February 1, 2022, https://www.cms.gov/research-statistics-data-systems/cms-program-statistics/2019-medicare-enrollment- section. MCARE-SIM reweights MCBS survey weights to match 2019 Parts A, B, and D enrollment separately for FFS enrollees and enrollees with MA and for those with and without LIS benefits for Part D. 8 See Table IV.B9 of Medicare Trustees (2020). To calculate per capita estimates for non-LIS enrollees, we take the sum of the direct subsidy average for all beneficiaries and 35 percent of the per capita reinsurance and risk- sharing spending. The estimate of the share of reinsurance and risk-sharing spending accruing to non-LIS enrollees is based on an estimate from a recent Medicare Payment Advisory Commission Part D status report indicating the share of enrollees reaching the catastrophic coverage range, in which reinsurance and risk sharing apply, by LIS status. See slide 14 of Suzuki, Schmidt, and Rollins (2021). 9 In 2019, the coverage gap began once total drug spending reached $3,820. 10 See "Medicare Utilization and Payment Section," Centers for Medicare & Medicaid Services, accessed February 1, 2022, https://www.cms.gov/research-statistics-data-systems/cms-program-statistics/2019-medicare- utilization-and-payment. 11 It is difficult to assess how low this lower bound is relative to the actual number of enrollees benefitting from a $2,000 OOP spending cap. Consumers are likely more price sensitive to prescription medicines than they are to other health care services. Although the RAND Health Insurance Experiment did not evaluate the effects of consumers' price responses to prescription medicines, some evidence from the introduction of the Part D program suggests arc elasticity estimates in the range of -0.3 (Einav, Finkelstein, and Polyakova 2016), meaning a $2,000 OOP spending cap could induce a sizable response from consumers. On the other hand, lengthening the range of the coverage gap over time (since 2020) may temper these behavioral responses by keeping consumers exposed to 25 percent cost sharing for a larger share of their drug expenditures, all else equal. 12 The 2019 enrollee estimate is based on the 2019 Centers for Medicare & Medicaid Services Program Statistics enrollment report. See "CMS Program Statistics," Centers for Medicare & Medicaid Services, last modified February 3, 2022, https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and- Reports/CMSProgramStatistics. Premiums for most LIS beneficiaries would be subsidized by Medicare. 13 See Table II.B1 of Medicare Trustees (2020). CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 11 14 Restricting this analysis to enrollees without LIS benefits excludes a large share of non-Hispanic Black and Hispanic Part D enrollees. About 57 percent of Black enrollees with Part D coverage have LIS benefits, and about 53 percent of Hispanic enrollees with Part D coverage have LIS benefits. References Cubanski, Juliette, Tricia Neuman, and Anthony Damico. 2021. "Potential Savings for Medicare Part D Enrollees under Proposals to Add a Hard Cap on Out-of-Pocket Spending." San Francisco: Kaiser Family Foundation. Duggan, Mark, Patrick Healy, and Fiona Scott Morton. 2008. "Providing Prescription Drug Coverage to the Elderly: America's Experiment with Medicare Part D." Journal of Economic Perspectives 22 (4): 69–92. https://doi.org/10.1257/jep.22.4.69. Einav, Liran, Amy Finkelstein, and Maria Polyakova. 2018. "Private Provision of Social Insurance: Drug-Specific Price Elasticities and Cost Sharing in Medicare Part D." American Economic Journal: Economic Policy 10 (3): 122– 53. https://doi.org/10.1257/pol.20160355. KFF (Kaiser Family Foundation). 2021. "An Overview of the Medicare Part D Prescription Drug Benefit." San Francisco: KFF. Medicare Trustees (Boards of Trustees, Federal Hospital Insurance, and Federal Supplementary Medical Insurance Trust Funds). 2020. 2020 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds. Washington, DC: Medicare Trustees. Suzuki, Shinobu, Rachel Schmidt, and Eric Rollins. 2021. "The Medicare Prescription Drug Program (Part D): Status Report." Presentation given virtually January 15. About the Authors Anuj Gangopadhyaya is a senior research associate in the Health Policy Center at the Urban Institute. His research focuses on the impact of safety net programs on health and well-being, family income, and education achievement outcomes for children in low-income families. He has focused on the impact of Medicaid eligibility expansion on children's education achievement, maternal and child health effects of the earned income tax credit program, and the impact of the Affordable Care Act Medicaid expansion on adult labor supply and fertility rates of women of reproductive age. He also helps lead Urban's Medicare simulation model (MCARE-SIM), estimating potential impacts of proposed policy changes on program spending, beneficiary spending, and use of services. Gangopadhyaya received his PhD in economics from the University of Illinois at Chicago. Bowen Garrett is an economist and senior fellow in the Health Policy Center. His research focuses on health reform and health policy topics, including health insurance and labor markets, Medicare's prospective payment systems, and evaluation of the Strong Start for Mothers and Newborns program. He leads the development of the Urban Institute's Medicare policy simulation model (MCARE-SIM). Previously, Garrett was chief economist of the Center for US Health System Reform and McKinsey Advanced Health Analytics at McKinsey and Company. Garrett received his PhD in economics from Columbia University in 1996 and was a postdoctoral research fellow in the Robert Wood Johnson Foundation's Scholars in Health Policy Research Program at the University of California, Berkeley, from 1996 to 1998. 12 CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 Acknowledgments This brief was funded by the Robert Wood Johnson Foundation. The views expressed do not necessarily reflect the views of the Foundation. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Funders do not determine research findings or the insights and recommendations of Urban experts. Further information on the Urban Institute's funding principles is available at urban.org/fundingprinciples. The authors thanks John Holahan and Stephen Zuckerman for providing feedback on the brief, Nikhil Rao for research assistance, and Rachel Kenney for editing. ABOUT THE ROBERT WOOD JOHNSON FOUNDATION The Robert Wood Johnson Foundation (RWJF) is committed to improving health and health equity in the United States. In partnership with others, we are working to develop a Culture of Health rooted in equity, that provides every individual with a fair and just opportunity to thrive, no matter who they are, where they live, or how much money they have. ABOUT THE URBAN INSTITUTE The nonprofit Urban Institute is a leading research organization dedicated to developing evidence-based insights that improve people's lives and strengthen communities. For 50 years, Urban has been the trusted source for rigorous analysis of complex social and economic issues; strategic advice to policymakers, philanthropists, and practitioners; and new, promising ideas that expand opportunities for all. Our work inspires effective decisions that advance fairness 500 L'Enfant Plaza SW and enhance the well-being of people and places. Washington, DC 20024 Copyright © March 2022. Urban Institute. Permission is granted for www.urban.org reproduction of this file, with attribution to the Urban Institute. CAPPING MEDICARE BENEFICIARY PART D SPENDING AT $2,000 13