Answers to Questions for the Record Following a Hearing on Health Care Spending

On January 31, 2024, the Subcommittee on Health of the House Committee on Energy and Commerce convened a hearing at which Chapin White, the Congressional Budget Office’s Director of Health Analysis, testified about federal subsidies for health insurance and policies to reduce the prices paid by commercial insurers. After the hearing, Members submitted questions for the record. This document provides CBO’s answers. It is available at www.cbo.gov/publication/60073.

Representative Latta’s Question About Artificial Intelligence and Machine Learning

Questions. How is CBO thinking about advances in AI/ML (Artificial Intelligence/Machine Learning) as a tool to help with projections and modeling of cost impacts, especially in the context of health interventions that can prevent disease or preempt the progression of disease?

Answer. AI and ML tools might affect health care costs in the future in many ways, including by detecting illness earlier or identifying patients who might benefit from preventive interventions. Although some uses of those tools might reduce costs by preventing the need for costlier care or eliminating unnecessary care, others might increase costs by spurring the development of expensive new technologies with meaningful health benefits or by identifying additional patients who might benefit from certain medical services.

To date, the evidence on the usefulness of AI and ML tools is mixed. For example, research has found machine learning useful for predicting cancer mortality but less useful for predicting heart failure outcomes. Although there currently is not sufficient empirical evidence to project the overall effect of AI and ML on federal health care spending, CBO is continually evaluating new research and is particularly interested in evidence regarding interventions that might prevent the progression of disease or the need for more intensive care.

Representative Carter’s Questions About Pharmacy Benefit Managers and CBO’s Estimate for the Reconciliation Act of 2022

Question. We know that Medicaid managed care companies and their PBMs are using spread pricing to overcharge taxpayer-funded state Medicaid programs to the tune of hundreds of millions of dollars. For example, in the last 2 years, Centene has reached settlements with 18 states, totaling almost $1 billion, for overcharges to taxpayer-funded state Medicaid programs. Can you explain the negative consequences on the federal budget if MCOs and PBMs continue to overcharge Medicaid programs absent price-transparency reforms in the Lower Costs, More Transparency Act?

Answer. CBO estimates that provisions in the Lower Costs, More Transparency Act (H.R. 5378) addressing pharmacy benefit managers’ (PBMs’) ability to charge Medicaid more than they paid pharmacies for medications and addressing the lack of the transparency of drug costs would decrease federal outlays by $1.1 billion over the 2024–2033 period.1 Specifically, section 202 of the legislation would reduce spending in two ways. First, it would ban such spread pricing in contracts among PBMs, Medicaid’s managed care organizations, and other specified entities and states. Second, it would require retail community pharmacies to participate in a national survey of acquisition costs for drugs that pharmacies pay to manufacturers or wholesalers, which is administered by the Centers for Medicare & Medicaid Services (CMS). CBO estimates that both measures would help states gain a better understanding of PBMs’ practices and negotiate more favorable contracts than they are able to under current law, thereby lowering federal spending.

Question. Section 106 of the Lower Cost, More Transparency Act requires pharmacy benefit managers to report data on the cost of dispensed prescription drugs to group health plans “not less frequently than every 6 months (or at the request of a group health plan, not less frequently than quarterly, but under the same conditions, terms, and cost of the semiannual report under this subsection).” On December 8, 2023, CBO estimated that this provision will save the federal government $23 million over 10 years. What specific levers within the provision produce these savings? What behavior does CBO believe this will result in from sponsors of group health plans, particularly employer plan sponsors? Can group health plan sponsors currently access this information easily, prior to enactment of this legislation?

Answer. In CBO’s assessment, section 106 would increase transparency in group health plans’ contracts with PBMs and would help some employers negotiate more favorable contract terms than they are able to under current law. Most of the savings would be generated through lower costs for pharmacy benefits in employment-based plans, which would allow their sponsors to reduce premiums. Because more employee benefits would be provided as taxable wages, rather than nontaxable benefits, revenue collections would increase. Over the 2024–2033 period, the increase in revenues would be about $2.2 billion.2

Small and medium-sized employers would benefit more from the mandated disclosure required by section 106 and would therefore generate most of the savings in pharmacy benefit costs because larger employers are better able to compel disclosure of such information under current law. Over time, CBO expects, the utility of the information would decline as PBMs found ways to recoup the profits they would have otherwise realized.

Question. Recent CBO budgetary modeling of major health legislation and its impact on costs has been fraught with error. The CBO overestimated by 100% the cost of Part D, a Republican policy proposal that created a prescription drug benefit for seniors. CBO published a look back report 10 years later examining its mistakes, does CBO plan to do the same with the ACA now that we are a decade out? Will the CBO commit to doing the same for the IRA in a decade? If not, why? Isn’t CBO accountable to Congress and the American people?

Answer. A core part of CBO’s mission is to compare its projections with actual outcomes, learn from those differences, and incorporate those lessons into its future baselines and policy estimates. In 2003, the agency substantially overestimated the cost of the Part D program in Medicare and, in 2014, published an analysis detailing the reasons for that overestimate.3 In 2017, the agency published an assessment of the accuracy of previous projections of federal subsidies for the expansion of Medicaid and for the subsidies available through the health insurance marketplaces established by the Affordable Care Act (ACA).4 That assessment compared projected subsidies with actual subsidies through 2016 and identified reasons for the differences. CBO has also published an overarching analysis of the accuracy of its projections of outlays over the past several decades.5

The 2022 reconciliation act (sometimes referred to as the Inflation Reduction Act, or IRA) included several significant provisions affecting prescription drug prices and coverage under Medicare.6 CBO has published two working papers and a slide deck describing its analytical approach to estimating the effects of those provisions.7 To gain a better understanding of the effects of the legislation, the agency is closely tracking the prices paid for prescription drugs and investments in the development of new drugs.8

CBO’s goal is twofold: first, to take on information as it emerges and update the agency’s models accordingly, and second, to inform the Congress about those updates. Informing the Congress can take the form of regular reports on the accuracy of the agency’s baseline or reports that focus on a policy of particularly great interest. CBO will consult with the appropriate committees on the timing and format for presenting what the agency has learned about how the key prescription drug provisions of the 2022 reconciliation act have played out. Some outcomes will be relatively straightforward to measure and report, such as the prices that CMS negotiates and payments that drug manufacturers are required to make to Medicare if they raise the prices of certain drugs more than the rate of inflation. But given all of the interactions between the policies and the lack of a clear counterfactual, it will be difficult to translate any differences into budgetary effects. Other outcomes, such as the effect of the legislation on the pace of drugs’ development, are not directly measurable and thus cannot be clearly identified.

Representative Crenshaw’s Questions About Health Insurance and Small Businesses

Question. What are the projected economic outcomes or effects of further limiting small businesses’ ability to band together to purchase health insurance through AHPs? Specifically, how might this affect small business growth, employee compensation and employment trends, and small business contribution to the economy and percentage of GDP?

Answer. In CBO’s assessment, small businesses’ ability to purchase health insurance through association health plans (AHPs) affects how much they pay for premiums and whether they choose to offer health insurance. AHPs offer small employers one way to purchase health insurance outside the fully regulated small-group market. In the fully regulated small-group market, premiums are modified community-rated, meaning that they can vary only on the basis of enrollees’ age, location, and tobacco use. AHPs have the latitude to also adjust premiums on the basis of other factors related to enrollees’ health status. Consequently, through an AHP, employers with a healthier than average workforce will probably pay premiums that are lower than modified community-rated premiums. Although the availability of AHPs probably increases offers of health insurance coverage through employers, that effect is limited because, in CBO’s estimation, most enrollment in AHPs instead comes from employers’ switching from offering coverage through the fully regulated market to offering coverage through an AHP.

For small businesses making use of AHPs under current law, limiting their ability to purchase health insurance would increase the premiums they pay, as many would move to obtaining health insurance in the fully regulated market. That increase in premiums would also slightly decrease the number of employers offering health insurance and the number of employees taking up such offers. The magnitude of those effects would depend on the details of the legislation.9 CBO has not analyzed the effects that limiting access to AHPs would have on the growth of small businesses, overall compensation for employees, employment trends, or the overall economy and gross domestic product.

CBO’s analysis of special tabulations from the Insurance Component of the Medical Expenditure Panel Survey, administered by the Agency for Healthcare Research and Quality, suggests that in each year from 2019 to 2021, fewer than 1 million employees of businesses with fewer than 50 employees were enrolled in health insurance through an AHP, making up about 2 percent of all employees in businesses of that size.

Question. Could the Congressional Budget Office (CBO) elaborate via a report on the potential long-term effects on the economy and the labor market, particularly in relation to unemployment rates, labor participation, and inflation as a result of a continued drop in the percentage of small firms offering health care benefits to their employees due to limited choices?

Answer. From tracking the share of small businesses offering health insurance, CBO has found the rates at which small businesses offer insurance to be relatively steady over the past decade. According to the Insurance Component of the Medical Expenditure Panel Survey, from 2014 to 2022 the percentage of businesses that offered health insurance fell from 25.7 percent to 24.9 percent among businesses with fewer than 10 employees, rose from 51.6 percent to 53.6 percent among businesses with 10 to 24 employees, and rose from 73.9 percent to 80.1 percent among businesses with 25 to 99 employees.

In an analysis of the effects of the ACA, CBO has previously examined linkages between health insurance and the labor market.10 In addition, for each update of its baseline budget projections, the agency assesses how economic developments affect the costs for health insurance and people’s coverage. The agency does not currently have plans to analyze the effects of changes in the share of small businesses offering health insurance on inflation and the labor market.

Question. In light of the recently proposed rescission of the 2018 Trump Administration Association Health Plan (AHP) rule, what alternative policy measures could be considered to support small businesses in offering competitive and affordable health benefits to their employees while also lowering health care costs? How might these alternatives compare in terms of economic and health coverage outcomes?

Answer. CBO makes no policy recommendations, but the agency has studied various approaches to reducing health care costs and premiums. In two 2022 reports, CBO analyzed the prices paid by commercial health insurers for hospitals’ and physicians’ services and policy approaches to reducing those prices, which would probably lower premiums as well.11 Possible broad approaches include promoting competition among providers (which would aim to reduce prices by targeting providers’ market power), promoting price transparency (which would aim to reduce prices by targeting consumers’ and employers’ price sensitivity), and capping the level or growth rate of prices (which would aim to reduce prices by regulating them).

CBO also examined design choices for a public option for health insurance in nongroup markets and noted that several legislative proposals would create a public option that would be available to employers.12 Depending on how those public options were designed, their availability could increase the offers of health insurance by small employers, whose decisions to offer insurance are more sensitive to premiums than large employers.13

CBO has been monitoring federal legislation and regulations relating to AHPs. In analyzing recent legislation and the proposed rescission of the 2018 rule on AHPs, CBO consulted with industry experts and stakeholders to better understand the types of employers that would be drawn to offering an AHP. From that analysis, CBO concluded that the primary motivation for businesses’ offering health insurance through an AHP rather than in the fully regulated small-group market was lower premiums and not the ability to offer a narrower set of benefits. Offering a competitive package of benefits including health insurance can help small businesses attract and retain employees. In examining previous proposals that would have expanded access to AHPs and other options, CBO found that if enacted, such proposals would tend to increase the number of businesses offering health insurance and reduce premiums for businesses that switched to those plans, but that premiums in fully regulated plans would probably increase.14

CBO is tracking closely the health insurance offerings of small businesses. Many small businesses that sponsor coverage offer plans that meet the requirements specified in the ACA, including modified community rating, risk adjustment (of plans’ costs based on enrollees’ health), guaranteed issuance, and coverage of a set of essential health benefits. Some small businesses continue to offer plans that were available before the implementation of the ACA’s rules affecting the small-group market. An increasing share of small businesses offer self-insured plans, often in the form of a “level-funded plan,” in which flat monthly payments cover expected health claims, premiums for a “stop-loss” policy that covers unexpectedly high costs for claims above a threshold, claims processing, and management of the provider network. In some states, AHPs, professional employer organizations, and a handful of other options are prevalent.

Representative Joyce’s Question About Consolidation Among Health Care Providers

Question. Dr. White, seeing how this continued underpayment has driven consolidation and required seniors to receive care in higher cost settings like hospital OPDs, how does CBO account for these increased government outlays from consolidation when weighing the direct budgetary cost of providing necessary relief in the physician fee schedule? Are there reforms Congress should be considering to help CBO better analyze these complex issues?

Links:

1. BHI Analysis: Hospital Outpatient Prices Far Higher, Rising Faster than Physician Sites
2. Avalere Analysis: CMS Site-Neutral Payments Affect Small Share of Spending

Answer. CBO’s baseline projections of Medicare spending and spending on commercial health insurance reflect the trend of more services being provided in hospital outpatient departments (HOPDs) rather than physicians’ offices. That long-standing trend has occurred partly because the vertical integration of hospitals and physicians has increased and partly because more independent physicians have chosen to provide more care in HOPDs.

Larger payments from both commercial insurers and Medicare for similar services provided in HOPDs (as opposed to physicians’ offices) are an important factor driving such changes, but they are not the only factor. Experts suggest that providers may also consolidate to increase their bargaining power during contract negotiations with private insurers, to control referral patterns, or to pool fixed administrative or technology costs, among other reasons. Some physicians may also prefer being employed by a hospital because that setting offers more predictable schedules, a team-based environment, and less financial risk than private practice. In addition, although the evidence supports the idea that payment policies contribute to increased consolidation and increased billing in higher-priced settings, it is not clear that payment reductions would reverse those tendencies in the market.

Given the range of factors contributing to the long-standing shift of services to higher-priced settings, it would be difficult for any single policy to meaningfully change the trend. Instead, the direct effects of policies that change payments to providers tend to swamp any possible secondary effects on the site of care. For instance, a policy to increase Medicare’s compensation for office-based physicians most meaningfully increases that program’s spending on office-based services, rather than shifting care across settings. Or a policy to give insurers more information about when they are billed for facility fees helps reduce spending on those fees but may not substantially affect the setting where care is provided. Unless a policy change would affect a large proportion of facility-based spending, it would be unlikely to alter the trend toward HOPD-based billing and thereby save money for the federal government.

Additionally, given that contractual relationships among hospitals, physicians, and insurers take a long time to develop, any policy changes that increased competition among providers probably would not lead to any meaningful reductions in the deficit within the 10-year budgetary window that is standard for CBO’s projections.

CBO uses information from a range of sources to analyze health care spending and competition. The agency relies heavily on other government agencies, including the Medicare Payment Advisory Commission, the Agency for Healthcare Research and Quality, and the Federal Trade Commission. The agency also frequently engages with academic and industry experts. In general, measures to make data on ownership and market structure more available to CBO and to the research community would be helpful in analyzing these issues.

Representative Allen’s Question About Health Care Costs and Transparency

Question. I have been asking experts in the health care field for years if they can break down the health care dollar and tell me where exactly it is being spent. No one has been able to give me a straight answer until last year when CBO Director Phillip Swagel was asked at an Energy and Commerce briefing.

As CBO’s Director of Health Analysis, can you explain the breakdown of the health care dollar in terms that patients and employers can understand, as well as the importance of price transparency for increasing efficiency in our healthcare system and how a bill like the Lower Costs, More Transparency Act can help in this effort?

Answer. People in the United States get health insurance coverage from a range of sources, including employment-based plans, Medicaid, Medicare, and plans obtained through the marketplaces established under the ACA. The federal government subsidizes that coverage in a number of ways by, for instance, excluding amounts paid for employment-based health insurance premiums from income and payroll taxes; paying for roughly two-thirds of the spending on state Medicaid programs and the Children’s Health Insurance Program (CHIP); subsidizing most spending in the Medicare program; and providing tax credits for people who purchase nongroup coverage through the ACA’s marketplaces.15 In 2023, by CBO’s estimates, those subsidies were $1.8 trillion, or 7 percent of gross domestic product. About half (47 percent) of those subsidies were for the Medicare program; 25 percent, for Medicaid and CHIP; and 21 percent, for employment-based coverage.

Roughly half of the people in the United States get their health insurance through an employer. Premiums for employment-based coverage are paid for jointly by employers and employees; in most cases, both the employer’s and the employee’s portions of the premiums are excluded from taxable income. On average, private insurers pay for about 80 percent of the total cost of covered care, CBO estimates. (The remaining 20 percent is generally covered by individuals out of pocket.) Recent estimates suggest that, in the past few years, private insurers spent 83 percent to 88 percent of the premium dollars they collect to pay for patients’ health care claims, with the remainder going to other activities, including administrative activities, taxes and fees, quality improvement, and profit.

According to CBO’s analysis of the national health expenditure accounts, private insurers’ spending for different types of care breaks down as follows: 45 percent for inpatient and outpatient hospital care, 37 percent for physicians’ services, and 18 percent for all other covered items, including prescription drugs.16

Price transparency is helpful for insurers and employers who are contracting with providers. It can also be helpful to consumers as they choose where to seek care. A lack of transparency, in CBO’s view, reduces the pressure on providers to compete. However, CBO’s review of the evidence suggests that the potential for transparency to reduce prices is limited for many reasons: Very few consumers make use of transparency tools when they are made available to them; the information provided in recent transparency regulations is often very difficult for consumers to use and not tailored to their plan; providers face limited competition, so consumers’ ability to shop for different services in some markets is curtailed; insurance makes consumers less sensitive to price information; health care is complicated, so patients and employers who particularly value quality may disregard price information; and finally, the exclusion of employment-based health insurance premiums from federal income and payroll taxes blunts employers’ and individuals’ responses to price information.

CBO estimates that some provisions of H.R. 5378, the Lower Costs, More Transparency Act, would reduce prices by a small amount.17 Section 204, which would require a separate provider identification number for off-campus hospital outpatient departments, would assist insurers who have a policy of not paying for facility fees for off-campus providers and might also encourage more private health insurers to adopt such a policy. Providing insurers with more information about when they are being charged those fees would help them avoid paying them, resulting in savings for consumers.

CBO views some of the other transparency measures in H.R. 5378, such as section 101 through section 105, as largely codifying existing transparency requirements and therefore having no budgetary effect. Some parts of those provisions would also expand upon current regulatory requirements to address challenges that may be limiting the effect of the rules. CBO has yet to see evidence that the underlying regulations have reduced premiums for private health insurance, so the agency has not estimated an effect from the incremental policy changes.

Representative Kuster’s Questions About the Rehabilitation and Recovery During Incarceration Act

Question. Studies show that for every dollar we spend to treat substance abuse in our prisons, we can save up to $7 down the road from better health outcomes and lower recidivism rates. I believe we should be doing more to maximize effective, cost-saving treatments, which is why I introduced the Rehabilitation and Recovery During Incarceration Act. My bill, which has bipartisan support, would reform the Medicaid Inmate Exclusion Policy so that incarcerated individuals who are eligible for Medicaid otherwise would have mental health and substance use services coverage while they are incarcerated.

Dr. White, does the Congressional Budget Office consider downstream savings from better health outcomes in its analysis?

Answer. Yes. When assessing the federal budgetary effects of policies aimed at improving health, CBO generally accounts for any direct effects of such policies (for example, the costs that result from policies to expand the use of preventive medical services) and also considers effects related to improvements in health (the indirect effects) if they are supported by an evidence-based body of research. In a 2020 report, the agency described how it analyzes proposals to improve health through disease prevention.18 That report includes a discussion of the types of evidence CBO draws on to inform its analyses.

A 2012 CBO report on raising excise taxes on cigarettes illustrates the ways in which a policy aimed at improving health can affect the federal budget.19 CBO concluded that such a policy would reduce spending per capita on health care programs and would, over the long term, increase longevity. That increased longevity would increase spending on health care programs and Social Security. Similarly, CBO, in assessing legislation related to mental health care and substance abuse, would seek to include information on potential budgetary effects associated with better health outcomes and increased longevity, to the extent such effects were supported by a body of evidence.

Question. In addition to increasing access to care and lowering health costs, my bill would help states and counties with expenses associated with drug-related crime and criminal justice.

Dr. White, does the Congressional Budget Office consider savings for states and localities when analyzing health policy, or does it only consider the impact on the federal budget?

Answer. CBO’s primary responsibility under the Budget Act is to assist the House and Senate Budget Committees, and the agency’s analyses generally focus on the federal budget.20 When lawmakers consider health policies, their decisions depend on broader considerations, which can include effects on the budgets of state and local governments, people’s health status and mortality, views about the appropriate role of the government in influencing behavior, and the burdens that the policy might impose on people in different circumstances. Where feasible and appropriate, the agency provides supplemental information on the effects of policies beyond those on the federal budget, including impacts on state and local governments. For example, CBO’s 2022 Budget Options volume includes an option to establish caps on federal spending for Medicaid, and that write-up discusses how states would respond and how they could be affected.21

Representative Trahan’s Question About the Accelerating Kids’ Access to Care Act

Question. Dr. White, I have partnered with my colleague, Ms. Miller-Meeks, to introduce legislation called the Accelerating Kids’ Access to Care Act, which would streamline health care provider enrollment in Medicaid programs outside their state in order to expedite access to timely care for kids with Medicaid who need medical treatment outside of their own state. I have been encouraged that CBO has met with the Children’s Hospital Association and others to gather data that can inform a CBO score of this bill.

Could you provide the Committee with a timeline for when you expect to have a score for this bill that accounts for data like that provided by the Children’s Hospital Association?

Answer. Staff are reviewing the bill and working to develop an estimate for it. CBO anticipates having a preliminary estimate for the bill within the next few months. In the meantime, the agency welcomes any data and additional information you or your staff would like to share with us.

---