Health Spending Trends

V. Key Health Policy Issues >> B. Health Spending >> Health Spending Trends


Topic Outline

1. Key Questions
     a. How much does the government owe?
     b. What would it take to close the fiscal gap?
     c.  How much public debt is sustainable?
     d. Are current health spending trends affordable?
     e. Innovation and the future of medicine
     f. What are the key drivers of U.S. health spending?
     g. How does the U.S. compare to the rest of the world in terms of health care spending?
2. Links
     a.  Historical Trends in Aggregate Health Spending
     b. Historical Trends in Household Health Spending
     c.  Historical Trends in Medical Prices
     d. Key Drivers of Health Spending
     e. Health Spending Projections


Key Questions

How Much Does the Government Owe?

In the aggregate, the fiscal gap–namely, debts and unfunded liabilities across all levels of government–amount to $244 trillion as of late 2013 inclusive of federal, state and local debts and the federal, state and local fiscal gap. Debt represents money already spent and owed, whereas unfunded liabilities represent the difference between promised spending and projected revenues based on current policy, i.e., assuming no changes in commitments or taxes. Note that current debts are not included in estimates of unfunded liabilities since the latter occur in the future but they are included in what’s called the fiscal gap.

These are net present value figures measured in 2013 dollars, meaning that they are inflation-adjusted and discounted to reflect the time value of money. The discount rate used for such purposes is generally the long-term U.S. Treasury bill rate (usually 3%) since that is viewed as the rate of return on a “safe” investment; thus, it is the rate that U.S. citizens require to be paid to give up a dollar today to get a dollar in the future. Thus, a dollar today is worth $1 whereas $1 (of revenues or spending) is worth only $1/1.03. Another way of viewing the $244 trillion then, is that it is the amount the U.S. Treasury would need in the bank today–earning interest at 3%–to generate the stream of payments over future generations needed to finance all the promises of government above and beyond the realistically-projected government revenues that are expected during the same period. Put another way, it is the lump-sum annuity payment required to cover in perpetuity all additional government promises that cannot be paid under the current tax structure.

However, it’s worth noting that a recent paper shows that from 1836 to 2011,  U.S. Treasury bills (generally viewed as a risk-free asset) earned an average annual inflation-adjusted return of 1.0% (gold’s real rate of return was 1.1%). The return for long-term bonds was 2.9% and that for stocks was 7.4%. If unfunded liabilities were discounted using a 1% rate rather than 3%, they would be much higher.

Government Debt. The federal debt ($17.5 trillion in March 2014) and state and local government debt ($3.0 trillion in March 2014) are reported continuously at the U.S. Debt Clock. Technically, the federal debt includes a public component (e.g., what is owed to other nations and individual U.S. citizens who have purchased Treasury bonds) and an internal component, i.e., what the U.S. Treasury owes the Social Security and Medicare trust funds in principal and interest payments in light of the many decades of borrowing payroll taxes paid into those trust funds to instead pay for other activities of government. Given that the U.S. government is most unlikely to renege on its obligations to seniors, many analysts view the distinction as meaningless. At some point, all those debts will have to be repaid in some fashion which means that the taxes or borrowing used to do so cannot be spent on other government purposes.

Federal Fiscal Gap. At the end of 2013, Laurence Kotlikoff estimated that the federal fiscal gap–which includes the federal debt–was $205 trillion, calculated as the net present value of all projected federal revenues and expenditures under current policy. This is equivalent to 10.3% of the net present value of GDP over the same time period. The 75-year US fiscal gap is only 40 percent of the true $205 trillion fiscal gap. Kotlikoff argues that any clear picture of the gap requires an infinite horizon since otherwise a finite horizon measure will depend on labeling terms (p. 11).

How the Gap is Calculated. These estimates were based on the Congressional Budget Office’s alternative fiscal scenario for the next 75 years. The alternative fiscal scenario (AFS) essentially reflects current policy as opposed to current law. For example, current law requires physician fees under Medicare to be cut drastically (24% in 2014), but every year for more than a decade, Congress has enacted a temporary “doc fix” that averted these steep fee reductions. Since these fixes are only temporary, the CBO in making its 75 year projections would have to assume  that a steep cut in fees was adopted as soon as the current fix expires–since that’s what “current law” requires. But this would result in a very unrealistic projection of Medicare spending: such drastic fee cuts would have such adverse consequences for the ability of Medicare recipients to find physicians willing to serve them that it is quite unlikely Congress ever will permit such cuts to take effect.

The AFS, in contrast, assumes Congress will continue doing what it has always done for this and other elements of policy. The alternative minimum tax is another illustration of where the law technically requires this to apply to many millions more taxpayers, but every year Congress enacts a  temporary reprieve, resulting in lower revenues than would otherwise be the case.  For spending beyond the 75 year window included in CBO’s annual projections of long-term federal revenues and spending, Kotlikoff assumed that annual non-interest spending, as well as taxes, would grow indefinitely by 2 percent a year beyond the point at which the CBO’s estimates end. Kotlikoff used the same method to calculate that by 2012, the federal fiscal gap had grown to $222 trillion–an increase of $11 trillion in just one year.  At least $122 trillion of this $222T is related to health entitlements, Medicare, Medicaid and new subsidies to be provided through health exchanges under the Affordable Care Act.

Trends. In 2011, the equivalently calculated fiscal gap was $211 trillion. In 2003, it was $60 trillion, but had risen to $175T by 2008 (p. 12). The 2013 tax hikes and budgetary sequestration represent significant policy changes. They shaved $17 trillion present value off the fiscal gap (p. 12).

Comparisons to Other Countries. Kotlikoff estimates the 2013 fiscal gap (10.3% of GDP) is essentially equivalent to that of Greece, Belgium and Japan. Finland (7%), Germany (5%) and Italy (5%) have lower fiscal gaps (Table 1).

State and Local Government Fiscal Gap. The equivalently-calculated figure of unfunded liabilities for state and local governments is $38 trillion. For example, as of fiscal year 2009, 61 major cities had promised at least $118 billion more than they had in hand to cover health care benefits for current and future retirees. Cities had set aside enough money to cover 6 percent of their promises, compared with slightly more than 5 percent in states.


What Would it Take to Close the Fiscal Gap?

Short Term. The CBO estimates that $2T in budgetary savings over the next 10 years would be required to keep the debt to GDP ratio constant at current levels. This would not eliminate current levels of debt, but it would prevent things from getting any worse during this period.

Long Term. The fiscal gap can be closed either through higher revenues or lower spending, or some combination. The figures below relate to close the federal fiscal gap of $205T. Thus, to address the entire $264T owed by governments at all levels would require these adjustments to be 29% larger.

  • Revenue Increases. Kotlikoff estimates that in late 2013, closing the federal fiscal gap ($205T) using taxes would require an immediate and permanent 57% increase in all federal taxes (in 2012, closing the $222T gap would have required a 64 percent increase in all federal taxes). If the decision were delayed, the required revenue increase would rise to 63.2% in 2023, 69,3% in 2033, and 75.9% in 2043 (Table 2).
  • Spending Cuts. Alternatively, the U.S. could cut, immediately and permanently, all federal purchases and transfer payments, including Social Security and Medicare benefits (i.e., all non-interest spending), by 37% (in 2012, this figure was 40 percent). If the decision were delayed, the required spending cuts would rise to 40.2% in 2023, 43.0% in 2033, and 46.3% in 2043 (Table 2).


How Much Public Debt is Sustainable?

According to Kotlikoff: “A country’s fiscal sustainability matters. It matters to a country’s growth path to its future tax rates, to its saving behavior, to its net domestic investment, to its labor supply, to its inflation rate, to its employment, to its wages, to its returns on capital, to the integrity of its financial markets, to the viability of its political institutions—indeed, it matters to virtually any question one might pose about a country’s economic future” (p. 5).

As of March 2013, U.S. public debt amounted to 74% of GDP. Public debt is the amount owed to others, as opposed to amounts owed by one part of government (e.g., Treasury Department) to another (e.g., Social Security Trust Fund.  There may be a moral obligation to pay back the Social Security and Medicare trust funds for past borrowing, but there is no economic obligation, i.e., interest rates will not rise were these amounts not repaid, whereas they assuredly would if the U.S. defaulted on its debts to other buyers of U.S. Treasury bonds. As of mid-2013, CBO projects that debt will continue to climb over the coming decades, reaching 100 percent of GDP by 2038, and twice the size of the economy by 2076. However, incorporating the negative effects of accumulating debt on the economy makes this picture worse.

A comprehensive cross-country study has shown that economic growth slows noticeably when a country’s public debt-to-GDP ratio exceeds 90%.  [Reinhart, Carmen M. and Rogoff, Kenneth, This Time Is Different: Eight Centuries of Financial Folly (2009)]. Thus, there is a large penalty to be paid for letting public debt grow beyond this limit since a slower economy means less government revenue available to repay the debt or interest on the debt.

Whereas individuals have to balance their budgets over their lifetimes (since creditors will not lend if they do not expect to be repaid), stable governments hypothetically can maintain some debt in perpetuity (i.e., creditors will keep lending since they know they will get repaid even if the government has to borrow again from someone else to make that happen). Generally, a sustainable level of debt is one where the debt-to-GDP ratio is kept stable. Under such circumstances–barring some catastrophe that independently drives up interest rates, such as a world war–creditors are assured of being repaid and hence long-term interest rates should remain stable.

But more precisely, a sustainable level of debt is one where the interest payments on the debt grow at or below the rate of growth in the economy. Interest payments on the U.S. debt currently are only a little more than 1 percent of GDP (out of government revenues that absorb less than one-fifth of GDP). If such payments grow substantially faster than the economy, then there would come a point at which interest payments could exceed the government’s ability to pay them. The political costs of financing debt payments–either by cutting spending or raising revenues drastically–might be too great, at which point defaulting on the debt may become the preferred option.

Either definition implies that debt can grow in absolute terms in perpetuity so long as this growth is at or below the rate of growth in GDP. But since the rate of economic growth itself can be affected by the level of debt, should it get too high, a country with a debt-to-GDP ratio below 90% can afford to let its debt grow faster than a country where the ratio exceeds 90%.


Are Current Health Spending Trends Affordable?

The Congressional Budget Office shows spending growth has exceeded real GDP growth. If these trends continue in the future, can the US afford to sustain their current spending habits? According to Micheal E. Chernew, our current trends are affordable only if excess health spending growth is limited to a 1% increase each year. He also illustrates how reform could keep things more affordable in the future. His insight has implications for many employers nation-wide whose employees are jeopardized by the rising health care costs.

A report prepared for the U.S. Department of Health and Human Services systematically examines The Effect of Health Care Cost Growth on the U.S. Economy. The report presents various perspectives on how growth in health care spending affects the US economy. The report summarizes a review of the literature, focusing on the mechanisms through which health care spending could affect the U.S. economy, including aggregate economic outcomes, employers, the government, households, and local economies. As well, available state-level data are used to analyze the effects of health care cost growth on aggregate economic indicators, industries, and state governments.


Innovation and the Future of Medicine

Medical innovation has the potential to either increase or decrease health spending trends relative to the economy. Eric Topol’s The Creative Destruction of Medicine: How the Digital Revolution Will Create Better Health Care argues that dramatic advances in genomics, personalized medicine, cellphone technology (mHealth), new imaging and new devices have the potential to fundamentally change how medical care is delivered. However, many of these digital medical innovations lie unused because of the medical community’s profound resistance to change.

What are the Key Drivers of US Health Spending? The document reviews up-to-date research about the key drivers of health care costs and a new policy framework for decreasing costs and enhancing quality of patient care. Americas Health Insurance Plans. 12/17/2013

In her issue brief, Cara S. Lesser addresses many problems of growing health care costs. Advances in technology and quality improvements in the health field are two key drivers of rising health costs. Paul B. Ginsburg notes that new medical technology has been the dominant driver of increases in health care costs and insurance premiums, and he discusses this implication in politics. The cost-effectiveness of new technologies raises a great debate. Furthermore, a peer group report of health care quality and cost indicated that more knowledgeable consumers today are demanding better quality of care, which in turn raises prices.


How Does the US Compare to the Rest of the World in Terms of Health Care Spending?

The Commonwealth Fund examined health care spending in the US and other OCED nations and found that U.S. health spending per capita significantly and consistently outpaces that of other industrialized nations. National health spending as a percentage of GDP is the highest in the US, thus health spending places more of a burden on the economy, says Uwe E. Reinhardt. This can partly be attributed to the fact that the US health system relies more heavily on specialty care than other developed nations. According to Leiyu Shi, the 70% of physicians in the United States are specialists, compared with 25-50% in other industrialized nations.

An NCPA report (2009) examines the evidence on Does the United States Spend Too Much on Health Care?




Historical Trends in Aggregate Health Spending

  • Congressional Budget Office, Growth in Health Care Costs, CBO Testimony before the Committee on the Budget United States Senate, January 31, 2008.
  • Getzen, Thomas, 2011. Long-run Forecasts of National Health Expenditure Growth. The “state of the art” in forecasting long run medical spending is assessed in models used by CMS, CBO, and the Society of Actuaries. Tracking medical expenditures by nominal dollar growth and real per capita spending are useful, yet focusing on the share (of wages, laborforce, or GDP) provides the perspective most immediately applicable to policy and capable of providing the most robust long run forecasts. Spending limits appear to be a variable result of politics and budgetary constraints more than morbidity and mortality. Although death and taxes may be certainties, “when” and “how much” are not.
  • Paul B. Ginsburg, High and Rising Health Care Costs: Demystifying U.S. Health Care Spending, Robert Wood
    Johnson Foundation. The Synthesis Project, October 2008.
  • Health Cost Index Report (Milliman). The Health Cost Index Report™ (HCIR) is a quarterly newsletter that presents the latest trends and forecasts from Milliman’s proprietary database of medical trends. These trends measure the market average rate of increase in medical costs for a typical $250 deductible Comprehensive Major Medical benefit package. The database measures the growth rate in medical consumption by measuring how fast provider net revenues increase. This inherently captures price, utilization and mix/intensity of service changes (technology). The HCIR presents trends by benefit and by region of the country. The quarterly reports also contain one year forecasts and Milliman’s latest research on medical trends.
  • Kaiser Family Foundation. Assessing the Effects of the Economy on the Recent Slowdown in Health Spending (April 2013). “about three-quarters (77%) of the recent decline in health spending growth can be explained by changes in the broader economy.”
  • Altarum InstituteHealth Sector Economic Indicators BriefsThe monthly HSEI briefs are designed to address significant shortcomings in the availability of timely economic data on the health sector, including employment, spending, and prices.  Official government estimates of national health expenditures are available annually only for the previous year (2009 estimates were released in January 2011). Examining updated health spending on a monthly basis significantly improves our ability to track trends and progress toward our goal of sustainable growth.


Historical Trends in Household Health Spending

  • Milliman Medical IndexThe annual Milliman Medical Index (MMI) measures the total cost of healthcare for a typical family of four covered by a preferred provider plan (PPO). The 2012 MMI cost is $20,728, an increase of $1,335, or 6.9% over 2011. Even though the rate of increase is slowing from prior years, it has taken fewer than nine years for such costs to more than double. In 2002, the cost of healthcare for the typical family of four was $9,235.
  • Auerbach, David I. and Arthur L. Kellermann, 2011. A Decade Of Health Care Cost Growth Has Wiped Out Real Income Gains For An Average US FamilyHealth Affairs 30(9).  Although a median-income US family of four with employer-based health insurance saw its gross annual income increase from $76,000 in 1999 to $99,000 in 2009 (in current dollars), this gain was largely offset by increased spending to pay for health care. Monthly spending increases occurred in the family’s health insurance premiums (from $490 to $1,115), out-of-pocket health spending (from $135 to $235), and taxes devoted to health care (from $345 to $440). After accounting for price increases in other goods and services, the family had $95 more in monthly income to devote to nonhealth spending in 2009 than in 1999. By contrast, had the rate of health care cost growth not exceeded general inflation, the family would have had $545 more per month instead of $95—a difference of nearly $5,400 per year. Even the $95 gain was artificial, because tax collections in 2009 were insufficient to cover actual increases in federal health spending. As a result, we argue, the burdens imposed on all payers by steadily rising health care spending can no longer be ignored.


Historical Trends in Medical Prices

  • Hospital, Employment and Price Indicators. Includes a) community hospital statistics; b) Medicare trust fund data; c) health sector employment, earnings and hours; d) medical prices; and e) HCFA projected price indices for hospital, nursing home, and home health.
  • AHRQ. Using Appropriate Price Indices for Analysis of Health Care Expenditures or Income Across Multiple Years. This document provides guidelines to help ensure consistency and avoid confusion about the use of price indices with MEPS expenditure or income data..
  • Bureau of Labor StatisticsConsumer Price Index. BLS tracks prices for a) Medical care, which is a composite of 1) Medical care commodities and 2) Medical care services.
    • Medical care commodities includes a) Medicinal drugs, which is a composite of 1) Prescription drugs; and 2) Nonprescription drugs; and b) Medical equipment and supplies.
    • Medical care services is a composite of a) Professional services, which includes 1) Physicians’ services, 2) Dental services, 3) Eyeglasses and eye care, and 4) Services by other medical professionals; b) Hospital and related services, which includes 1) Hospital services [which includes Inpatient hospital services and Outpatient hospital services], 2) Nursing homes and adult day services, and 3) Care of invalids and elderly at home; and c) Health insurance.
  • Bureau of Labor Statistics.  Producer Price Index. BLS tracks prices for a) Offices of physicians, b) Offices of dentists, c) Medical and diagnostic laboratories, d) Home health care services, e) Blood and organ banks, f) Hospitals, g) Nursing care facilities, and h) Residential mental retardation facilities.
  • Aizcorbe, AnaRalph BradleyRyan Greenaway-McGrevyBrad HeraufRichard KaneEli LiebmanSara Pack, and Lyubov Rozental, 2011. Alternative Price Indexes for Medical Care: Evidence from the MEPS Survey. Bureau of Labor Statistics, Washington DC. Spending on medical care is a large and growing component of GDP. There are well-known measurement problems that are estimated to overstate inflation and understate real growth for this sector by as much as 1-1/2 percentage points per year. Because of its size, this would translate into an overstatement of inflation for the overall economy of about ¼ percentage point with an equal understatement in real GDP growth. In this paper, we use data from the Medical Expenditure Panel Survey to obtain new, more comprehensive estimates for this bias and to explore a possible adjustment to existing official price indexes. The MEPS data show an upward bias to price growth in this sector of 1 percentage point, which translates into an overstatement of overall inflation of .2 percentage point and an understatement of GDP growth of the same amount. We also find that an adjustment recently used in Bradley et al provides a useful approximation to the indexes advocated by health economists.


Key Drivers of Health Spending

  • Chandra A, Skinner J., 2012. Technology growth and expenditure growth in health care. Journal of Economic Literature 2012; 50(3):645-80. In the United States, health care technology has contributed to rising survival rates, yet health care spending relative to GDP has also grown more rapidly than in any other country. We develop a model of patient demand and supplier behavior to explain these parallel trends in technology growth and cost growth. We show that health care productivity depends on the heterogeneity of treatment effects across patients, the shape of the health production function, and the cost structure of procedures such as MRIs with high fixed costs and low marginal costs. The model implies a typology of medical technology productivity: (I) highly cost-effective “home run” innovations with little chance of overuse, such as anti-retroviral therapy for HIV, (II) treatments highly effective for some but not for all (e.g. stents), and (III) “gray area” treatments with uncertain clinical value such as ICU days among chronically ill patients. Not surprisingly, countries adopting Category I and effective Category II treatments gain the greatest health improvements, while countries adopting ineffective Category II and Category III treatments experience the most rapid cost growth. Ultimately, economic and political resistance in the U.S. to ever-rising tax rates will likely slow cost growth, with uncertain effects on technology growth.
  • Clemens, Jeffrey, 2011. The Effect of U.S. Health Insurance Expansions on Medical Innovation, SIEPR Discussion Paper No. 11-016, Stanford University. I study the effect of health insurance expansions on medical innovation. Innovation by practitioners creates important roles for local patient flows and payment systems as drivers of medical technology development. I show that, over the 15 years following Medicare and Medicaid’s passage, U.S.-based medical-equipment patenting rose by nearly 50 percent relative to both other U.S. patenting and foreign medical-equipment patenting. Surges in medical-equipment patenting were largest in the states most significantly affected by the Great Society programs. No relative increase occurred among pharmaceutical patents, for which markets were not directly affected. The dynamic effect of U.S. insurance expansions may account for 25 percent of recent global medical-equipment innovation and 15 percent of the rise in U.S. health spending in hospitals, physicians’ offices, and other clinical settings from 1960 to 2010.

Health Spending Projections

  • Aaron, Henry. “There Is No Entitlement Problem,” Brookings Institution, February 23, 2009. “That the United States faces daunting long-term budget challenges is indisputable. But the very projections—those of the Congressional Budget Office—cited to document the long-term budget challenge, show that there is no general entitlement problem. Rather, the nation faces a daunting health care financing problem that bedevils private insurers and public programs alike.”
  • Altarum Institute. Altarum Health Sector Model (AHSM) forecasts health spending as captured in the National Health Expenditure Ac­counts (NHEA), based upon population demographics, disease prevalence, pre­vailing treatment patterns, insurance coverage/care access, and payment rates. This structure makes AHSM a powerful tool for integrating broad research strains related to health expenditure determinants. Altarum plans to create state-based ver­sions.
  • Altarum InstituteLong-Term U.S. Fiscal Crisis: Introduction to Altarum’s Triangle of Painful Choices. This short video describes a novel way to understand the long-term fiscal challenge the U.S. faces in light of choices regarding tax revenues, national health care spending, and federal spending on non-health items. It convincingly demonstrates that severe sacrifices are needed along all three dimensions if the country is to regain fiscal balance at reasonable debt levels.
  • Blahous, Charles.  CBO Explodes the Health Care Myth, Hudson Institute, June 30,2009, argues that 2007 CBO long term health projections report substantially understated the contribution of population aging on health spending trends.
  • Centers for Medicare and Medicaid Services, U.S. Department of Health and Human Services. NHE Historical and projections, 1965-2020 (ZIP, 22 KB).
  • Centers for Medicare and Medicaid Services, U.S. Department of Health and Human Services. Projections of National Health Expenditures: Methodology and Model Specification. July 28, 2011.
  • Congressional Budget Office, The Long-Term Outlook for Health Care Spending, November, 2007.
  • Congressional Budget OfficeLong-Term Budget Outlook, June 2010. This report concludes that through 2035, population aging would account for fully 64 percent of the cost growth in the major federal mandatory health programs and Social Security, with excess health cost inflation being a relatively smaller factor. By 2080, health spending is projected to account for 41% of GDP under the baseline scenario and 43% of GDP under the more realistic alternative fiscal scenario (p. 42).
  • The U.S. Healthcare System: Can This Patient Be Saved?” The Global Human Capital Journal, February 24, 2008. Provides assessments of U.S. health system from key experts, including Ian Morrison, Ph.D., healthcare futurist, Dean Harrison, CEO Northwestern Memorial Healthcare; William Novelli, CEO AARP; and Scott P. Serota, CEO BlueCross BlueShield Association.
  • Institute for the FutureWhat could health and health care in the US look like in 2020? This group postulates what health care could be like under 4 alternative scenarios: a) a growth scenario that manifests the results of current trends and conditions, extrapolated forward. This includes both positive and negative growth; b) a discipline scenario in which a core guiding value or purpose is used to organize society and control behavior (e.g., population control); c) a collapse scenario in which major social systems are strained beyond the breaking point, causing system collapse and social disarray; and d) a transformation scenario in which a fundamental reorganization of a society or system signals a break from previous systems (e.g., greater-than-human machine intelligence).