U.S. health spending is slowing down

New data suggest a persistent decline in the share of the economy devoted to health. Recent policies and ongoing changes in this sector can help to sustain this trend.

Photo by National Cancer Institute on Unsplash

New estimates from the Bureau of Economic Analysis (BEA) show that although total health spending in the country increased in 2022, the share of Gross Domestic Product (GDP) spent on health declined for the second consecutive year. What contributed to the decline in the share of the economy devoted to health? The COVID-19 pandemic has disrupted the usual patterns of health care utilization, accelerated health care transformations that were already underway, necessitated an unprecedented temporary injection of federal funds, and contributed to high economy-wide inflation. Some of these changes are likely to be short-lived, while others might persist in the long term and affect the future trajectory of health spending in the United States.

The share of the economy devoted to health is declining

According to the new BEA estimates, the share of GDP spent on health in 2022 declined to 17.1 percent from 17.5 percent in 2021 and 17.9 percent in 2020.¹ This marks the second consecutive year of decline in health spending as a share of GDP. The estimated share in 2022 represents the smallest share of the economy spent on health since 2014 and a smaller share that was spent on health in 2019, before the start of the pandemic. It is also below the historic projections produced by the Centers for Medicare and Medicaid Services (CMS) (Figure 1).

Figure 1: Health spending as a percent of GDP, actual and projected

Several factors contributed to the recent decline

There are several possible reasons for the decline in the share of our economy that was devoted to health care in 2022.

First, there are factors – triggered by the pandemic – that affected the health sector itself, either by impacting prices or quantities (via changes in health insurance coverage as well as utilization intensity per person) of health care services and goods. To combat the pandemic and its impact, there was a significant increase in federal government spending that peaked in 2020 and continued at lower levels in 2021 and 2022. Some of this unprecedented increase in federal spending went directly into the health sector in the form of financial assistance to health care providers to make up for lost revenue and public health investments to support COVID vaccine development and testing. As a result, some of the reduction in total health spending in 2022 relative to 2021 and 2020 is due to the reduction in temporary federal pandemic-related spending.

The pandemic also disrupted the usual patterns of health care utilization. There was a significant decline in health care use during the first pandemic year when many people postponed outpatient visits and elective procedures. Since then, health care use rebounded but there is some evidence that utilization likely grew at a slower rate in 2022 relative to the pre-pandemic period. Hospital discharges and the share of adults with a doctor or a hospital emergency department visit remain below the pre-pandemic levels.

Historically, health care prices rose faster than the overall prices in the economy. However, this trend was reversed in 2021 and 2022, when health care prices grew at a much slower rate relative to economy-wide prices. During 2022, the average annual percent change in prices as measured by the Consumer Price Index for All Urban Consumers (CPI-U) for medical care was 4.1 percent, while for all goods and services, it was 8.0 percent. Prices in the health sector remained mostly in line with historic trends, while prices outside of the health sector grew significantly faster in recent years. This general trend continued well into 2023 with medical care prices growing significantly slower than all goods and services and is likely because medical prices are set in advance either administratively by the government through programs such as Medicare and Medicaid or through negotiations and long-term contracts between providers and private insurers. It has been documented that, even prior to the pandemic, there has been slower growth in relative medical prices since 2010 primarily due to slower growth in input prices of health care production relative to the overall input prices. This relative price effect is unlikely to continue in the long run as input prices tend to grow similarly across different sectors of the economy.

The gap between medical and overall inflation in recent years contributed to disproportional growth rates in the nominal GDP relative to the nominal total health spending. The nominal GDP grew at 10.7 percent in 2021 and 9.2 percent in 2022 – rates much faster than during the pre-pandemic period and faster than recent growth in nominal health spending. This reversal of growth rates between nominal economy-wide spending and health spending occurred after decades of health spending growing faster than the rest of the economy. Adjusting for differences in overall and medical inflation, the share of GDP spent on health in 2022 was largely unchanged from 2020 and 2021 (Figure 2). As such, the decline in health spending as a share of GDP in 2021 and 2022 was primarily driven by high inflation in sectors outside of health and a lagging medical inflation.

Figure 2: Health spending as a percent of GDP, with and without inflation adjustment

Recent policies and trends may sustain the decline in health spending share

There are several recent policies and ongoing trends in the health sector that will likely impact the future trajectory of total health spending in the United States by either impacting medical prices or quantities of consumed health care. Decreasing growth in nominal health spending relative to the growth of the overall economy will lead to a smaller share of our economy devoted to health, freeing up resources for other priority areas. Table 1 lists the main policies and trends expected to affect future health spending and their likely effects on prices, quantities, and net spending. Some of these factors are expected to affect health spending over a shorter, pre-specified period (such as the enrollment increase in the Affordable Care Act (ACA) Marketplace plans and the expiration of extra Medicare payments for COVID patient care), while other factors (such as Medicare prescription drug negotiation and greater emphasis on disease prevention) are more fundamental to how health care is delivered and, thus, will impact health spending over a longer period.

Immediate-term factors that will likely impact health spending

Policies that impact insurance coverage will likely impact the quantity of consumed health care over the next few years. Medicaid disenrollments are happening across the country following the expiration of the continuous enrollment provision implemented during the pandemic. So far, at least 9 million Medicaid enrollees have been disenrolled. Fewer Medicaid enrollees will result in lower Medicaid spending, mostly in the short term. Some of these dropped enrollees will obtain private health insurance through the ACA Marketplaces, while others will become uninsured. Per capita health spending among private payers is higher than among public payers. As a result, total health spending for those transitioning from Medicaid to private insurance will likely increase. But for those who become uninsured, total spending, even accounting for uncompensated care provided by providers and the government, will likely decrease (Finkelstein et al. 2018; Garthwaite et al. 2018; Finkelstein et al. 2019; Finkelstein et al. 2019). On net, it is likely that the ongoing Medicaid disenrollments will lead to a reduction in total health spending. The elimination of extra Medicare payments to providers to care for COVID patients will lead to a decline in total health spending. Similarly, and probably more consequential in the long term, the Inflation Reduction Act (IRA) will likely impact Medicare spending on prescription drugs by requiring drug price negotiation and limiting the drug price increases to the level of inflation. 

On the other hand, the increase in ACA Marketplace plan enrollments due to the enhanced premium subsidies available through 2025 will likely lead to a net increase in health spending in the short term as more people obtain health insurance coverage and increase their care utilization.

On net, these recent policies will likely result in lower health spending that would have occurred otherwise in the absence of any of these policies. This is primarily due to the IRA’s impact on Medicare spending.

Several long-term factors will likely reduce health spending

In terms of longer-term trends in health care, there are several factors that will likely lead to lower health spending. There will likely be an acceleration in the ongoing shift from inpatient care toward outpatient, at-home, and virtual care. As providing health care outside of hospitals is cheaper, this will likely result in lower health spending over time. Fewer barriers to accessing care could lead to an increase in utilization but the net effect is likely to be lower health spending. Greater emphasis on disease prevention in the future will allow individuals to prevent or catch diseases early thus reducing total health spending. There has also been a greater awareness and understanding of the impact of factors outside the health care system itself – so-called social determinants of health – on health care use and health outcomes. Greater investments in social determinants of health, such as affordable and quality housing, reliable transportation, and sufficient and healthy food, will result in improvements in health which will lower health spending.

The ongoing shift away from a traditional fee-for-service payment system toward systems focused on outcomes may improve outcomes, reduce inefficient and wasteful care, and lead to lower spending (Finkelstein et al. 2018; MedPAC Report to Congress 2020). Many value-based payment systems are currently either being tested or have already been implemented in some parts of the health care system. As more evidence on the effectiveness of these systems becomes available, there will likely be a significant expansion and scaling of these value-based payment systems in many parts of the U.S. health care system. Greater health data sharing and interoperability as well as the increased use of AI technology for both clinical and nonclinical functions will allow individuals and providers to catch diseases earlier, improve treatments, and reduce administrative waste, leading to increased productivity in the health sector and potentially lower spending.

But several long-term factors will likely increase health spending

On the other hand, there are several factors that will likely lead to higher health spending. Due to the inadequate supply of health care providers to meet the growing demand for health care, the gap between the supply of and demand for physicians and nurses may result in higher wages and lower utilization due to worsening access to care. On net, it will likely push health spending up. As the higher overall inflation of 2021 and 2022 filters through to all sectors of the economy, there will likely be increased input costs in the health sector. Higher input costs may lead to increased provider reimbursement rates leading to higher aggregate spending. Additionally, recent rising interest rates have led to increased credit stress and higher financing costs for many health care companies. Continuation and further acceleration of provider and insurer consolidations and corporate investment in health care (for example, private equity investments) lead to higher costs with small, if any, improvements in outcomes (Kessler and McClellan 2000; Cooper et al. 2019). As the COVID pandemic enters its endemic phase, utilization of care will increase during future waves of the virus leading to higher health spending during those periods. Finally, demographics also play a major role. As the population ages, demand for health care services and goods increases.

It is hard to predict changes in the rate and impact of medical technology innovations, especially as new and more powerful AI tools enter our world. Some innovations may be expensive but lead to significant improvements in life expectancy and quality of life. Others may also lead to reductions in health spending, offsetting the cost of the innovation.

What do all these policies and trends mean for future health spending in the United States?

While it is hard to make precise predictions, it seems plausible that the Inflation Reduction Act’s impact on Medicare drug spending, the acceleration of shift away from hospital care, greater emphasis and investments in social determinants of health, modification of provider incentives to prioritize disease prevention, and improved data and technology may outweigh other factors and lead to slower health spending growth relative to past projections.

What share of our economy is spent on health will ultimately depend on the relative growth in nominal GDP and health spending. Returning to the pre-pandemic trend of higher medical inflation relative to the economy-wide inflation might increase health spending share of GDP, all else equal. CMS projects that the decline in the health spending share of GDP is temporary and that it will increase over the next decade as health spending growth will outpace growth in the overall economy. If recent health care trends and future policies lower future health spending more than currently anticipated by CMS – a plausible assumption given the consistency of lower actual spending relative to projections, then the recent decline in health spending as a share of GDP may continue.

Table 1: Recent health care policies and trends and their likely effects on future health spending

Source: The author’s summary of an existing body of empirical evidence and personal judgment. Note: An up arrow indicates a likely increase in total health spending, a down arrow indicates a likely decrease in total health spending, a question mark indicates an unknown impact on total health spending due to a high degree of uncertainty, and a null symbol indicates likely no major impact on total health spending. Double arrows indicate the likely magnitude of an impact on price (or quantity) relative to an impact on quantity (or price) and the strength of evidence of an impact on net spending.

Changes in national health spending must be examined jointly with changes in outcomes

Next month CMS will release the official estimate of U.S. health spending in 2022. It will likely show a continued reduction in the share of the economy devoted to health care in 2022. While there are several reasons for the decline, it seems likely the decline was primarily driven by high inflation in sectors outside of health and lagging medical inflation.

Several recent policies and ongoing trends in the health sector will likely impact the future trajectory of total health spending in the United States. Relative growth in nominal GDP and health spending will ultimately determine what share of our economy is spent on health.

It is important to evaluate changes in the country’s health spending jointly with resulting changes in health care access, costs, health outcomes, and health equity. In 2021 – the latest year of data available – life expectancy in the United States continued its second-year decline. The pandemic played a major role in this decline but, in most comparable countries, life expectancy started to rebound in 2021. In the same year, maternal mortality increased by 40 percent from 2020. Racial and ethnic disparities in health insurance coverage persisted in 2021. On the other hand, the percentage of adults reporting not getting needed medical care due to cost declined during 2020-2022 relative to 2019, the pre-pandemic year. The share of individuals without health insurance continued to decline during 2021-2022, reaching an all-time low in early 2023. Gains in health insurance tend to translate to improvements in access to health care and subsequent health outcomes. Any changes in national health spending will need to be examined jointly with these and other access and quality metrics.

1. The official measure of the country’s total health spending – the National Health Expenditure Account (NHEA) data – is released by the Centers for Medicare and Medicaid Services (CMS) with approximately a one-year lag, and in December of this year, CMS will release health spending estimates for 2022. Recently released BEA estimates now allow researchers and policymakers to see how the country’s aggregate health spending changed during 2022. BEA and CMS use different methodologies and some data sources differ when estimating total national health spending. However, the two estimates have been consistently similar over time, with the only exception of 2020 and 2021 when an unprecedented one-time increase in government spending related to the COVID-19 pandemic was counted differently between the two measures. It seems likely that the BEA estimates will continue to track the official CMS estimates of health spending in the post-pandemic period.