The Heritage Foundation

Backgrounder #2318 on Health Care

September 21, 2009

September 21, 2009 | Backgrounder on Health Care

Sound Health Care Reform for a Sound Economy: A Response to the CEA Report

Key Points

  1. The CEA identifies numerous serious structural problems with the American health care system that are not addressed by current health care reform proposals.
  2. There are serious problems with current methods of pricing medical treatment, incentives for providers and consumers point in the wrong directions, the system is too fragmented, and much of the information needed for all parties to make good decisions is unavailable.
  3. The CEA correctly notes that fixing the problems with the American health care system would produce substantial economic benefits; however, none of the current proposals address those problems, so none would produce those benefits.
  4. Current health care reform proposals from the Administration and congressional Democrats address none of the serious problems that the CEA identifies and would in fact exacerbate some of those problems. They will, therefore, not lead to any of the economic benefits described in the CEA report and are more likely to harm both the health care system and the overall economy.

Abstract: The President's Council of Economic Advisers correctly notes that fixing several problems with the American health care system would produce substantial economic benefits. However, the current health care reform proposals from the Democrats would address none of these serious problems. In fact, their proposals would exacerbate some of the problems, without producing any of the economic benefits described in the CEA report.

The nation's health care delivery and financing systems have many serious problems, and President Barack Obama has reiterated his desire for major reform to fix them. Rapidly rising health care expenditures can be a manifestation of a troubled health care system, and devoting an increasing share of national income to health care might restrain growth in other sectors of the economy. The President's Council of Economic Advisers (CEA) released a report on June 2 that discusses these issues and how health care reform could strengthen the economy in the long run.[1]

The CEA's report offers much useful and insightful analysis of problems in the health care sector of the economy, but draws a rosy conclusion based on its stated assumption that health care reform will produce a best-case scenario. The report simply assumes, without giving any justification, that health care reform would "slow the annual growth rate of health care costs by 1.5 percentage points," which in turn "would increase real gross domestic product (GDP), relative to the no-reform baseline, by over 2 percent in 2020 and nearly 8 percent in 2030."[2]

The problem with this scenario is that none of the Democrats' current proposals for health care reform is capable of "slowing the annual growth rate of health care costs" because none of them would even address, much less solve, the most serious problems that the CEA identifies in the current health care system. All current reform proposals from the Administration and congressional Democrats would either increase health care spending even faster or reduce spending by limiting patient access to necessary care. Indeed, the CEA report conspicuously avoids discussing any specific reform proposals, either those currently before Congress or any others that the CEA might believe would achieve its best-case scenario.

While the CEA presents a tantalizing picture of the economic benefits that could theoretically accompany ideal health care reform, it gives no reason to believe that enacting any of the health care proposals before Congress would produce these benefits.

Health Care Reform and Economic Growth

If Americans could maintain the current level of health for a lower total cost, the resources saved could be used for some other beneficial purpose, and U.S. economic well-being would undoubtedly improve. This is the basic claim of the CEA report, and it is uncontroversial -- even tautological. Yet would health care reform lead to this idealized outcome?

Some experts claim, based on regional differences in Medicare spending[3] or the results of the RAND Health Insurance Experiment,[4] that health care expenditures could be reduced by up to 30 percent without any adverse health consequences. Notwithstanding the controversy surrounding these claims,[5] there are also many specific reasons to believe that America's health care system is inefficient, many of which the CEA details in its report. The U.S. health care system is clearly far from optimal, and reform, if done properly, would produce great benefits, both economic and otherwise.

The problem is that none of the reform proposals from the Administration and congressional Democrats would accomplish this goal. These reform proposals fall into two basic categories:

Reforms that would necessarily increase health care spending, with or without improving care, and

Reforms that would restrict access to care and very likely produce adverse health outcomes.

Sadly, the wrong reform could quite possibly do both.

For example, the Kennedy-Dodd bill in the Senate and the America's Affordable Health Choices Act (H.R. 3200) in the House of Representatives, focus on increasing health insurance coverage by expanding coverage to the uninsured and mandating increased coverage and lower out-of-pocket payments for those currently uninsured. While these may be admirable goals, especially expanding coverage, this approach would necessarily increase health care spending. In fact, that is their goal.

One problem these proposals seek to solve is that the uninsured and "underinsured" do not spend enough on health care. Covering the uninsured would improve their access to health care by enabling them to spend more. If reducing spending on health care improves economic growth, these proposals, whatever their merits, would not achieve that goal. Likewise, reducing or eliminating out-of-pocket payments, such as deductibles, co-insurance, and co-payments for those with insurance, will encourage those individuals to seek more health care and thus increase total spending.

Proposals centered on "cost containment" -- which usually really means expenditure containment -- work by limiting patients' access to health care. If the goal is to reduce spending without regard to patients' well-being, the government can easily contain health care costs by making higher spending illegal. The American Health Security Act (S. 703), introduced by Senator Bernie Sanders (I-VT) and Representative Jim McDermott (D-WA), would take this approach by establishing a Canadian-style system with a "global budget" and by banning private insurance and private health spending. S. 703 would explicitly limit total national health care spending to the 2008 level plus the GDP growth rate. It would prevent health spending from ever increasing as a share of GDP, except during a recession when the health budget would remain constant. Similarly, the United States National Health Care Act (H.R. 676), introduced by Representative John Conyers (D-MI), would have Congress annually establish a national limit on health care spending.

These proposals would likely succeed in limiting health care spending, but do nothing to ensure that health care outcomes would remain the same. On the contrary, with each state (and hospital) assigned a specific annual budget, patients would be turned away when the money ran out. Indeed, this is similar to the way the Indian Health Service -- a single-payer health care system run by the federal government -- already operates. The Wall Street Journal recently described the health care system "on reservations, where the common wisdom is 'don't get sick after June' -- the month when the federal dollars usually run out."[6]

Clearly, "health coverage" is not the same as "health care." Under this approach to cost containment, everybody would be covered, but everybody would be denied health care once the spending limits were reached. Nothing in this approach would make health care more efficient or more effective. The only goal is to limit spending, even if patients suffer. The reduction in spending would not be worth the reduction in health and longevity.

Sources of Current Problems

The CEA's excellent discussion of the problems with the current health care system is far more insightful than what is typically heard from most policymakers and pundits. The authors identify several key sources of inefficiency in the current system: structural features that lead to unnecessarily high expenditures for a given level of health care and features that reduce quality, but not costs. The problems include:

Because health care providers are paid for providing services, rather than for the effectiveness of those treatments, they have little incentive to avoid providing costly or excessive treatments.

Because most insured patients are largely insulated from the cost of care, they have little incentive to seek out the most cost-effective treatments.

Prices for health care services are usually determined by administrative procedures, which are based on imperfect historical measures of cost and, in the case of Medicare, are subject to intense lobbying by interested parties. As the CEA points out, these systems are slow to adjust to reductions in costs. Although not explicitly mentioned in the CEA report, by focusing on only cost of a service, these systems ignore the value of the service to the patient and thus blunt the incentives for patients and providers to choose treatments with the best value. On the contrary, the system discourages such choices by paying more for high-cost, low-value treatments than for low-cost, high-value treatments.

The health care system is highly fragmented. The lack of incentives for proper communication among different providers who are treating the same patient leads to higher utilization (for example, redundant tests) and poorer health outcomes. In addition, a diversity of billing systems increases providers' administrative costs.

Information about the effectiveness of and interaction among multiple treatments is difficult and expensive for providers to obtain. Providers have little disincentive to provide expensive treatments of marginal additional value compared to their less-costly alternatives.

It is difficult for providers to measure their own performance, and the payment system gives them little incentive to establish or implement systems to give them feedback.

It is difficult for patients to obtain information about provider performance, and providers have little incentive to communicate to patients or prospective patients what information they have about their own performance, even if that information is positive.

In short, there are serious problems with the method of pricing health care services, incentives for providers and consumers point in the wrong directions, the system is too fragmented, and much of the information needed for all parties to make good decisions is unavailable. Discussion of these factors is too often missing from the health care debate, which seems at times to focus primarily, or even exclusively, on insurance coverage. Yet these problems would all remain even if the entire population had health insurance.

The only flaw in the CEA's diagnosis of the ills afflicting the American health care system is the extensive discussion of the supposed U.S. low performance in life expectancy and infant mortality. This is the standard conventional wisdom, supported by unverified data from the Organisation for Economic Co-operation and Development (OECD) and the World Health Organization (WHO). However, the data do not account for the different standards that countries use to report live births and, by implication, infant mortalities.

The United States has the most inclusive definition of live birth, and as a result reports a higher infant mortality rate than other countries would report in the same circumstances.[7] In other words, births that would be reported in other countries as stillbirths are reported as infant mortalities in the U.S. This also reduces the reported U.S. life expectancy, because the U.S. reports more "near-zero" life lengths in the data because of its inclusive definition of live birth.

Furthermore, U.S. infant mortality statistics and, consequently, life expectancy figures are sometimes adversely affected by factors reflecting improvements, not defects, in the health care system. For example, high rates of treatment for infertility reflect the widespread availability of advanced, expensive treatments that provide substantial benefit to many people who want, but would otherwise be unable to have children. However, they also result in a disproportionate number of high-risk pregnancies and infant mortalities. In addition, social factors also contribute to an increasing average maternal age, which increases the percentage of high-risk pregnancies and infant mortalities for reasons having nothing to do with defects in the health care system. Furthermore, attempts to save babies with conditions that are more frequently "treated" by abortion in other countries skew U.S. infant mortality rates compared to the rates in those countries.

In addition, the U.S. has a higher rate of accidental death than other advanced countries. While this is no comfort to Americans, it reflects many social and other factors besides the quality, efficiency, and accessibility of the U.S. health care system.[8] When adjusted for the higher rate of accidental death, the U.S. life expectancy is the highest in the world.[9]

All of these problems are distinct from the problem of insurance coverage, and these problems could conceivably persist even if every American had health insurance. Indeed, if the only achievement of health care reform is to provide health insurance for all Americans, these substantial systemic problems would remain, harming patients and increasing costs.

Economic Side Effects

The CEA report also discusses some consequences of the U.S. system of health insurance on the rest of the economy, particularly the effects of the link between health insurance and employment. For example, the report states that labor mobility is impaired by this link. In fact, the effect of health insurance on job-to-job mobility has largely disappeared since the Health Insurance Portability and Accountability Act of 1996 took effect, prohibiting pre-existing condition exclusions for those changing insurance plans. However, mobility from employment to entrepreneurship is still somewhat impaired.[10]

If health insurance were easier to obtain outside of the employment relationship, mobility into entrepreneurship and perhaps into small firms could be improved. Although labor market flexibility is certainly good for the economy, the CEA report calls this an "increase in labor supply," but this is not the proper way to characterize this improvement. For instance, when a person leaves one job to accept another, there is no net increase in the labor supply; it is just transferred from one employer to another. Furthermore, this characterization neglects the possible offsetting reduction in labor supply. After all, a significant number of people take jobs primarily to obtain health insurance. For example, many spouses of entrepreneurs obtain "minimal" jobs because obtaining health insurance as a teacher's aide, for example, is easier and often cheaper than purchasing insurance on the individual or small-group market as an entrepreneur. If health insurance were easier to obtain outside of the employment relationship, these individuals might devote themselves to different jobs, household production, or their own or their spouses' entrepreneurial enterprises.

The authors discuss the underwriting environment that makes it difficult or impossible for many people with health problems to obtain insurance in the individual market, but fail to note that the tax code exacerbates this problem by penalizing those who do not or cannot obtain insurance through their employers. They also fail to note that the link between employment and health insurance, which is responsible for so many other problems, actually provides a solution for vast numbers of people with health problems (pre-existing conditions) that are expensive to treat and thus make insurance in the individual market expensive or unobtainable, but that do not render them unable to work.[11]

The report also neglects other important causes of uninsurance, such as state mandates for health benefits that increase premiums and make insurance unaffordable to many people[12] and community rating requirements that make individual insurance prohibitively expensive for many healthy people.

The CEA's Unrealistic Assumption

The health care system clearly has substantial inefficiencies, which means that, theoretically, the same level of health care could be delivered at a lower total cost. The key assumption behind the CEA argument is that not only is this possible in theory, but that health care reform will necessarily produce this result. The authors also assume that the savings be will manifested as a reduction in the growth rate of total health care expenditures, which will then be directly reflected in increased GDP in other sectors of the economy.[13] They simply guess at the savings that could be achieved from optimal reform -- whatever that might be -- and extrapolate the savings forward in time. They then divide those savings by arithmetically convenient multiples to "calculate" more conservative estimates.

Basically, the authors assume that health care reform would reduce the growth rate of total health expenditures by 1.5 percentage points. They explicitly state the assumptions that the improvement would mean "we can obtain the same health care outcomes" using fewer economic resources and that the resources not spent on health care would be spent producing other useful output instead.

The lack of conservatism in their assumptions lies not in the number they choose for the reduction in the rate of growth, but rather in the assumption that health care reform can, and necessarily will, achieve the same level of health at a lower total cost. As such, their estimates are necessarily "best-case scenarios" -- not numerically, in the sense that they assume some percentage improvement, but that a lower percentage would be more realistic -- but in the sense that whatever reform is implemented is assumed to achieve cost savings without degrading health care quality or health outcomes.

The problem with this assumption is that it does not reflect health care reform reality. Nearly all of the health care reforms suggested by the Administration or by congressional Democrats either explicitly increase spending, or decrease spending in a way that would effectively guarantee reductions in health care quality, access, or outcomes. Specifically, the CEA paper's results would not apply to any health care reform that uses:

Rationing or artificial resource limits of any kind;

Waiting or queuing of any kind;

"Global budgets" for the entire health care system or any part of it (including states and individual hospitals);

Mandated minimum benefit levels for health insurance, which would force some people to pay for services that they know in advance they would not use;

Restrictions on treatments based on age, remaining life expectancy, or a measure of cost-effectiveness that fails to take into account individual patient preferences;

Reductions in payment levels or working conditions that induce health care professionals to leave their professions through earlier retirement or career changes or that discourage workers from choosing a career in health care;

Restrictions in the choice of treatments, including limitations on the choice of prescription drugs;

Restrictions on labor markets, such as requiring a minimum percentage of physicians to specialize in primary care or limiting the number of residencies; or

Reduction in the rate of development and/or adoption of improved medical technology, such as new prescription drugs, new medical devices, and new diagnostic imaging modalities.

Every health care reform proposal discussed favorably by the Administration or congressional Democrats contains at least one and usually several of these features, any one of which would adversely affect health outcomes and reduce health care quality and access, leaving patients worse off than they would be without this sort of "reform."

The CEA paper does not discuss any specific health care reform proposals, let alone describe any reform that might achieve the results that it assumes are possible. The paper merely suggests that, if the best possible health care reform were discovered and enacted, substantial economic benefits would ensue. This is true, but unsurprising, uninformative, and almost tautological.

Useful, But Incomplete

The CEA report offers much useful and insightful analysis of problems in the American health care system and offers a tantalizing picture of the widespread economic benefits that might accompany solutions to these problems. However, the report fails to link any particular health care reforms with theoretically achievable economic benefits.

Furthermore, the conditions required to achieve these economic benefits are conspicuously absent from current health care reform proposals. On the contrary, most reform proposals from the Administration and congressional Democrats contain provisions that would preclude the achievement of these widespread economic benefits, either by mandating lower levels of health care services, by requiring higher levels of involuntary spending, or both.

To achieve the CEA's optimistic scenario, the U.S. needs health care reform that overcomes the fragmentation of the current system, enhances patient choice, and provides patients and providers with the information to make better-informed choices. This would harness market-based incentives to reduce prices and costs and, therefore, insurance premiums by encouraging both health care providers and patients to seek out the most beneficial and cost-effective treatments and preventive care.

Health care reform is important to America's health and the health of the national economy, but that reform must address the most serious problems in the existing health care system and should be based on sound analysis rather than rosy assumptions. Otherwise, the U.S. will not achieve better health at lower costs for more people or give patients greater control over the health care decisions that affect them.

Robert A. Book, Ph.D., is Senior Research Fellow in Health Economics in the Center for Data Analysis at The Heritage Foundation.

Appendix: Measuring Health Care's Contribution to GDP

The CEA report notes[14] that there is a conceptual problem with how health care is counted in GDP. GDP is normally measured as the total market value of all goods and services produced in a country. It would be convenient to regard a particular industry's portion of GDP as a reasonable approximation of its contribution to the well-being of the population. However, in the case of the health care industry, the value (even the "market value") of the output is difficult to measure directly, or even to define, because its contribution to the well-being of the population -- "improved health" and "elimination of disease" -- is not something that can be bought and sold directly.

Instead, unlike most industries, the health care industry's contribution to GDP is in practice calculated as the value of its inputs rather than its outputs: the total amounts spent on physician services, hospital treatments, drugs, and so forth. This means that if it became possible to achieve the same health outcomes by spending less, it would reduce health care's contribution to measured GDP, even though it would not reduce the well-being of the population, which is what we would like GDP to measure.

For example, if a new technique were developed that allowed heart attacks to be treated at half the cost with the same level of effectiveness, ideally, the contribution of "heart attack treatment" to GDP should remain the same because the "output" of that activity is the same. However, because health care is measured based on expenditures on inputs, that contribution to measured GDP would be cut in half, making the health care industry appear less productive even though it became more productive.

Under the CEA's assumption that health care savings would be entirely redirected to other industries, measured GDP would not change. The moneysaved treating heart attacks, for example, would be spent on something else besides health care. Therefore, the increases in measured GDP in other sectors of the economy would exactly offset the reduction in measured GDP in health care. The benefit to the population would be reflected in the value of other goods obtained with the savings achieved in the health care sector. The result would be no change in measured GDP or economic growth -- even though inefficiencies in the health care sector would have been eliminated and the well-being of the population would thereby have been improved. To get around this problem, the CEA report introduces the idea of "conceptual GDP," which allows health care's contribution to GDP to remain constant even as less is spent in that sector, thus allowing total conceptual GDP to increase when health care spending is redirected to other sectors.

While this is admirable in principle, the authors' application of this notion introduces a serious inconsistency. The authors ask that we compare measured GDP before reform to conceptual GDP after reform. This is not a valid "apples-to-apples" comparison. Furthermore, it is impossible to obtain the necessary data in practice with current data and measurement methods. In other words, even granting all of the assumptions and projections in the CEA paper, we are still asked to simply believe that conceptual GDP would increase.

If the authors wish to make a valid comparison of this type, they should provide a precise definition of conceptual GDP and method for calculating it -- and most importantly, apply that method consistently -- both to actual GDP before reform and to projected GDP after reform. Such an endeavor is possible in principle and, if implemented properly, would provide a better basis for measuring the contributions of the health care sector to society.[15]

About the Author

Robert A. Book, Ph.D. Senior Research Fellow in Health Economics
Center for Data Analysis

Related Issues: Health Care

Show references in this report

[1]Executive Office of the President, Council of Economic Advisers, "The Economic Case for Health Care Reform," June 2009, at http://www.
whitehouse.gov/assets/documents/CEA_Health _Care_Report.pdf
(June 23, 2009).

[2]Ibid., executive summary.

[3]John E. Wennberg, Elliot S. Fisher, and Jonathan S. Skinner, "Geography and the Debate over Medicare Reform," Health Affairs Web Exclusive, February 13, 2002, pp. W96-W113, at http://content.healthaffairs.org/cgi/content/full/hlthaff.w2.96v1/ DC1 (September 8, 2009).

[4]Willard G. Manning, Joseph P. Newhouse, Naihua Duan, Emmett B. Keeler, Arleen Leibowitz, and M. Susan Marquis, "Health Insurance and the Demand for Medical Care: Evidence from a Randomized Experiment," American Economic Review, Vol. 77, No. 3 (June 1987), pp. 251-277.

[5]Richard A. Cooper, "States With More Health Care Spending Have Better-Quality Health Care: Lessons About Medicare," Health Affairs, Vol. 28, No. 1 (2009), pp. w103-w115, at http://content.healthaffairs.org
/cgi/content/full/28/1/w103
(September 11, 2009).

[6]Terry Anderson, "Native Americans and the Public Option," The Wall Street Journal, August 28, 2009, at http://online.wsj.com/article/
SB10001424052970203706604574376981533298534.html
(September 11, 2009).

[7]For example, see Bernadine Healy, "Behind the Baby Count," U.S. News & World Report, October 2, 2006, at http://health.usnews.com/
usnews/health/articles/060924/2healy.htm
(September 8, 2009); Gabriel Duc, "The Crucial Role of Definition in Perinatal Epidemiology," Social and Preventive Medicine, Vol. 40, No. 6 (November 1995), pp. 357-360; and Barbara A. Anderson and Brian D. Silver, "Infant Mortality in the Soviet Union: Regional Differences and Measurement Issues," Population and Development Review, Vol. 12, No. 4 (December 1986), pp. 705-738.

[8]Clearly, accidental death and homicide rates are not completely independent of the quality of a country's health care system. A high-quality health care system can certainly reduce the fatality rates from injuries and assaults. However, if the underlying rate of injuries and assaults is higher in a particular country, then that country's life expectancy will be lower even with an equally good (or perhaps even better) health care system.

[9]Robert L. Ohsfeldt and John E. Schneider, The Business of Health: The Role of Competition, Markets, and Regulation (Washington, D.C.: AEI Press, 2006), pp. 19-23.

[10]Paul L. Winfree, "Does Employer-Sponsored Health Insurance Reduce Job Mobility?" Heritage Foundation White Paper, February 26, 2009, at http://www.heritage.org/Research/HealthCare/wp032609a.cfm.

[11]For example, in the case of someone with Type I (juvenile) diabetes, the disease is usually diagnosed at an early age before the patient enters the work force. Treated properly, the result is a long life free of employment-impairing disabilities, but with substantial health care costs. Such a person would find it difficult or expensive to purchase insurance in the individual market, but could obtain job-based insurance as an employee or dependent on the same basis and at the same price as anyone else.

[12]Victoria Craig Bunce and J. P. Wieske, "Health Insurance Mandates in the States 2009," Council for Affordable Health Insurance, 2009, at http://www.cahi.org/cahi_contents/resources/pdf/HealthInsuranceMandates
2009.pdf
(September 11, 2009).

[13]For a more detailed discussion of issues that arise in measuring the health care sector's contribution to GDP, see the Appendix.

[14]Executive Office of the President, "The Economic Case for Health Care Reform," pp. 22-24.

[15]For a similar and credible comparison, see Kevin M. Murphy and Robert H. Topel, "The Value of Health and Longevity," Journal of Political Economy, Vol. 114, No. 5 (October 2006), pp. 871-904.