Congressional Budget Office
Director's Blog

CBO issued a paper on health information technology last week; the blog entry on that paper is here. Unfortunately, when we released the paper, we erroneously left Laura Adams of the Rhode Island Quality Institute off the list of reviewers who were acknowledged in the paper. Laura provided extremely valuable comments, and we are very appreciative of the time and effort she put into helping us improve the quality of the paper.

On the broader topic of reviewers of CBO products, I want to emphasize two things.  First, we regularly seek rigorous outside expert review of our studies.  Second, reviewers are not responsible for what we say or don’t say.  As noted in the preface to the health IT paper (and other CBO papers), “The assistance of external reviewers implies neither responsibility for the final product, which rests solely with CBO, nor endorsement of the conclusions of CBO’s analysis.”  In other words, the fact that an expert is listed as outside reviewer does not mean that the reviewer agrees with all (or I suppose even any!) of what appears in the document.

CBO released an issue brief today on different approaches for splitting the projected growth in the costs of the major federal health care programs into “excess cost growth” and demographic effects (these two factors reflect, respectively, rising costs per beneficiary and the number and age of beneficiaries). For more information about CBO’s projections of long-term health care costs, see here.

The key points of the issue brief are:

• If health care costs per beneficiary grew at the same rate as per capita gross domestic product (GDP) and the age distribution of the population did not change, Medicare and Medicaid spending would remain a constant share of the economy. In reality, however, health care costs per beneficiary will grow more quickly than per capita GDP and the population will age.

• Although many observers portray aging as the dominant cause of future growth in federal spending on Medicare and Medicaid, most of the increase that CBO projects reflects rising costs per beneficiary rather than rising numbers of beneficiaries. The effect of population aging is smaller but still results in substantial spending growth. Rising costs and population aging also interact with each other: The rapid growth of health care costs is a more important factor when the population is aging over time; conversely, population aging looms larger when health care costs are rising over time.

• Understanding the relative contributions of those two factors to the growth in federal spending on Medicare and Medicaid is important. The aging of the population, which has a smaller impact on spending growth, cannot be easily influenced by policy changes, but efforts can be made to stem the rising costs per beneficiary relative to per capita GDP—by far the larger factor in spending growth in the two programs.

• To estimate the relative contributions of the two factors, CBO based its projections on three sets of reasonable assumptions. Regardless of the assumptions and methods used in the projections, the results were basically the same: More than half of the growth in federal spending on Medicare and Medicaid is attributable to health care costs per person growing more rapidly than per capita GDP. Depending on the approach used, by 2082 between 53 percent and 60 percent of the accumulated growth is attributable solely to cost growth, between 14 percent and 17 percent is attributable solely to aging, and the remainder (between 26 percent and 30 percent) is attributable to the interaction of those two factors as costs grow and the population ages at the same time. Over the next 25 years, aging will be relatively more important, accounting for between 27 percent and 35 percent of projected growth by 2032, but even during that period, CBO’s estimates suggest that more than half of the growth in spending will result from rising costs per beneficiary.

The issue brief was written by Noah Meyerson, who works in the long-term modeling group within our Health and Human Resources Division.

I am delivering a talk on behavioral economics and health care later today at the National Academy of Social Insurance. Today’s event is being held in honor of Peter Diamond’s winning the Robert Ball award from NASI. Peter is not only brilliant but also a wonderful friend and colleague, and I am thrilled that he has won this award.

Peter has been interested in behavioral economics, which combines insights from psychology with those from economics, for almost 40 years. The talk emphasizes my belief that effective policy design, including policies affecting health care, must reflect more of the insights from behavioral economics — that is, we need a bit more “Psych 101″ in addition to “Econ 101″ in the design of public policies.

CBO is often asked for our most recent estimates of before- and after-tax household income, along with our estimates of effective tax rates. We have therefore just created a special page on our website containing our most recent household tax and income data, to make it easier for people to find those numbers.

Many people believe that health information technology (health IT) has the potential to transform the practice of health care by reducing costs and improving quality. CBO just released a significant study, prepared at the request of the Chairman of the Senate Budget Committee, examining the evidence on the costs and benefits of health information technology.

“Health IT” generally refers to computer applications for the practice of medicine. (Those applications may include computerized entry systems for physicians’ orders for tests or medications, support systems for clinical decisionmaking, and electronic prescribing of medications.) Relatively few providers—as of 2006, about 12 percent of physicians and 11 percent of hospitals—have adopted health IT.

When used effectively, health IT can enable providers to deliver health care more efficiently. For example, it can:

• Eliminate the use of medical transcription and allow a physician to enter notes about a patient’s condition and care directly into a computerized record;

• Eliminate or substantially reduce the need to physically pull medical charts from office files for patients’ visits;

• Prompt providers to prescribe generic medicines instead of more costly brand-name drugs; and

• Reduce the duplication of diagnostic tests.

Indeed, many analysts and policymakers believe that health IT is a necessary ingredient for improving the efficiency and quality of health care in the United States. Research does indicate that in some instances, health IT appears to have reduced the cost of providing health care, helped eliminate inappropriate services, and improved the quality of care. In general, however, health IT appears to be necessary but not sufficient to generate cost savings; that is, health IT can be an essential component of an effort to reduce cost (and improve quality), but by itself it typically does not produce a reduction in costs.

The most auspicious examples involving health IT have tended to involve relatively integrated health systems. For providers and hospitals that are not part of integrated systems, however, the benefits of health IT are not as easy to capture, and perhaps not coincidentally, those physicians and facilities have adopted electronic health records (EHRs, the primary health IT package commonly purchased by a provider) at a much slower rate. For example, office-based physicians in particular may see no benefit if they purchase such a product – and may even suffer financial harm. Even though the use of health IT could generate cost savings for the health system at large that might offset the EHR’s cost, many physicians might not be able to reduce their office expenses or increase their revenue sufficiently to pay for it.

The search for improved efficiency in delivering health care has prompted numerous proposals for increasing the adoption of health IT. For example, a recent study by the RAND Corporation estimated about $80 billion in net annual savings that is potentially attributable to such technology. This study has received significant attention, but unfortunately it suffers from significant flaws and is therefore not an appropriate guide to estimating the effects of legislative proposals aimed at boosting the use of health IT:

• The RAND researchers attempted to measure the potential impact of widespread adoption of health IT, assuming that it was used effectively—rather than the likely impact, which would take account of factors that might impede its effective use. For example, health care financing and delivery are now organized in such a way that the payment methods of many private and public health insurers do not reward providers for reducing costs—and may even penalize them for doing so.

• The RAND study is based solely on empirical studies from the literature that found positive effects for the implementation of health IT systems; it excluded the studies of health IT, even those published in peer-reviewed journals, that failed to find favorable results. The decision to ignore evidence of zero or negative net savings clearly biases any estimate of the actual impact of health IT on spending.

• The RAND study was not intended to be an estimate of savings measured against the rates of adoption that would occur under current law, but rather, against the extent of adoption in 2004. That is, the study did not allow for growth in adoption even without a policy intervention, as CBO would in a cost estimate for a legislative proposal.

One significant potential benefit of health IT that has thus far gone relatively unexamined involves its role in comparative effectiveness research. Widespread use of health IT could make available large amounts of data on patients’ care and health, which could be used for empirical research that might not only improve the quality of health care but also help make the delivery of services more efficient. By making clinical data easier to collect and analyze, health IT systems could support rigorous studies to compare the effectiveness of different treatments for a given disease or condition. Then, in response to the studies’ findings, they could aid in implementing changes in the kinds of care provided and the way those services are delivered, and track progress in carrying out the changes. Such comparative effectiveness studies would, on average, probably lead to reductions in total spending for health care because of the tendency in the current health care system to adopt ever more expensive treatments even though rigorous evidence about their effectiveness is lacking. The likelihood of such reductions in spending would be higher if the studies’ findings were linked to the payments that providers received or the cost sharing that patients faced.

If the federal government chose to intervene directly to promote the use of health IT, it could do so by subsidizing that use or by imposing a penalty on failing to use a health IT system. From a budgetary perspective, the subsidization approach is less likely to generate cost savings for the federal government because it involves up-front costs. (It is also possible that, for any given underlying financial incentive, a penalty may be more effective at triggering adoption than a subsidy if a penalty carries a negative connotation that does not apply to failing to receive a subsidy.)

Stuart Hagen of CBO’s Health and Human Resources Division and Peter Richmond, formerly of CBO, prepared the report under the supervision of Bruce Vavrichek and James Baumgardner.

I am testifying this morning on climate change before the Senate Committee on Energy and Natural Resources. Reducing greenhouse-gas emissions would provide benefits to society by helping to limit the damage associated with climate change, especially the risk of significant damage. The webcast of the hearing is posted here. The testimony focuses on ways to reduce the economic cost of achieving any given greenhouse gas emissions target. In particular:

• Market-oriented approaches to reducing carbon emissions, such as a cap-and-trade program, would reduce emissions more cheaply than would command-and-control approaches, such as regulations requiring across-the-board reductions by all firms. Those market-oriented approaches are relatively efficient because they create incentives and flexibility for emission reductions to occur where and how they are least expensive to accomplish.

• The cost of meeting an emission target with a cap-and-trade program could be reduced, potentially quite substantially, by providing firms flexibility in the timing of their efforts to reduce emissions. In its most inflexible form, a cap-and-trade program would require that a specified cap on emissions was met each year. That lack of flexibility would increase the cost of achieving any long-term goal because it would prevent firms from responding to year-to-year differences in conditions that affected costs for reducing emissions — such as fluctuations in economic activity, energy markets, the weather, and the technologies available for reducing emissions. In contrast, because of the long-term nature of climate change, the key issue from an environmental perspective involves the long-term emissions and concentration paths of greenhouse gases, not the year-to-year fluctuations in emissions. The most cost-effective cap-and-trade design would thus encourage firms to make greater reductions when the cost of doing so was low and would allow them leeway to lessen their efforts when the cost was high. Providing firms with such flexibility could also prevent large fluctuations in the price of allowances that could be disruptive to the economy.

• One option for allowing firms flexibility in determining when to reduce emissions while also achieving compliance with a cumulative emissions target would be through setting both a ceiling—typically referred to as a safety valve—and a floor on the allowance prices each year. The price ceiling would allow firms to exceed the annual target when the cost of cutting emissions was high, while the price floor would induce firms to cut emissions more than the annual target in low-cost years. The price ceiling and floor could be adjusted periodically to ensure that emission reductions were on track for achieving the long-run target; such a dynamic price system could substantially reduce the cost of a cumulative emissions target.

• Policymakers’ choices about whether to distribute the allowances without charge or to auction them — and if auctioned, how to use the proceeds — could also have a significant effect on the overall economic cost of capping emissions. Evidence suggests that the cost to the economy of a 15 percent cut in U.S. emissions (not counting any benefits from mitigating climate change) might be half as large if policymakers sold the allowances and used the revenue to lower current taxes on capital that discourage economic activity, rather than giving the allowances away to energy suppliers and energy-intensive firms or using the auction proceeds to reduce the costs that the policy could impose on low-income households. Using the allowances’ value to lower the total economic cost could, however, exacerbate the regressivity of the policy change.

Under current law, rising costs for health care and the aging of the population will cause federal spending on Medicare, Medicaid, and Social Security to rise substantially as a share of the economy. At the request of the Ranking Member of the House Budget Committee, CBO released a letter examining the potential economic effects of (1) allowing federal debt to climb as projected under the alternative fiscal scenario presented in CBO’s December 2007 Long-Term Budget Outlook; (2) slowing the growth of deficits and then eliminating them over the next several decades; and (3) using higher income tax rates alone to finance the increases in spending projected under that scenario.

How Would Rising Budget Deficits Affect the Economy? Sustained and rising budget deficits would affect the economy by absorbing funds from the nation’s pool of savings and reducing investment in the domestic capital stock and in foreign assets. As capital investment dwindled, the growth of workers’ productivity and of real (inflation-adjusted) wages would gradually slow and begin to stagnate. As capital became scarce relative to labor, real interest rates would rise. In the near term, foreign investors would probably increase their financing of investment in the United States, but such borrowing would involve costs over time, as foreign investors would claim larger and larger shares of the nation’s output and fewer resources would be available for domestic consumption.

How much would the deficits projected under the alternative fiscal scenario presented in the December 2007 Long Term Budget Outlook affect the economy? For its analysis, CBO used a textbook growth model that can assess how persistent deficits might affect the economy over the long term. According to CBO’s simulations using that model, the rising federal budget deficits under this scenario would cause real gross national product (GNP) per person to stop growing and then to begin to contract in the late 2040s. By 2060, real GNP per person would be about 17 percent below its peak in the late 2040s and would be declining at a rapid pace. Beyond 2060, projected deficits would become so large and unsustainable that the model cannot calculate their effects. Despite the substantial economic costs generated by deficits under this model, such estimates greatly understate the potential loss to economic growth because the effects of rapidly growing debt would probably be much more disorderly and could occur well before the time frame indicated in the scenario.

How Would the Slowing the Growth of Deficits Affect the Economy? The minority staff of the House Budget Committee provided CBO with a target path slows the growth of budget deficits. In evaluating the economic effects of the target path, CBO did not examine how specific policies to achieve that path would affect the economy; instead, CBO limited its attention solely to examining how the deficits produced by the target would affect the economy, assuming that such effects would play out as they have in the past. (CBO has not evaluated either the political feasibility or the economic effects of reducing spending sufficiently to accomplish this path for the deficit. Furthermore, the spending and revenue targets provided by the Committee staff are not the only way to achieve a sustainable budget path. Alternative policies will have different effects on the economy, and changes in taxes and spending can exert influences on the economy other than the effects of reducing budget deficits.)

Under the target path, federal outlays excluding interest (that is, primary spending) would rise from 18 percent of GDP in 2007 to 20 percent in 2030 and then decline to 19 percent in 2050 and 13 percent in 2082. For almost all years, revenues would remain at 18.5 percent of GDP. Under those assumptions, the budget deficit would gradually increase to about 6 percent of GDP in 2040 but then would decline to almost zero in 2075. By 2082, the target path would generate a budget surplus of about 2 percent of GDP. Under this path, real GNP per person would continue to grow over the entire projection period, rising from about $45,000 in 2007 to about $165,000 in 2082 in inflation-adjusted dollars. By 2060 (the last year for which it is possible to simulate the effects of the alternative fiscal policy using the textbook growth model), real GNP per person would be about 85 percent higher under the target path than under the alternative fiscal scenario.

How Would Increasing Income Tax Rates to Finance the Projected Rise in Spending Affect the Economy? How would the economy be affected if the projected rise in primary spending under CBO’s alternative fiscal scenario (from about 18 percent of GDP in 2007 to about 35 percent in 2082) was financed entirely by a proportional across-the-board increase in individual and corporate income tax rates? Answering that question is difficult because the economic models that economists have developed so far would have to be pushed well outside the range for which they were initially developed.

Nonetheless, tax rates would have to be raised by substantial amounts to finance the level of spending projected for 2082 under CBO’s alternative fiscal scenario. Before any economic feedbacks are taken into account, and assuming that raising marginal tax rates was the only mechanism used to balance the budget, tax rates would have to more than double. Such tax rates would significantly reduce economic activity and would create serious problems with tax avoidance and tax evasion. The letter provides more details about possible scenarios. (Raising revenue in ways other than increasing tax rates would have a less marked effect on economic activity.)

Conclusion. The United States faces serious long-run budgetary challenges. If action is not taken to curb the projected growth of budget deficits in coming decades, the economy will eventually suffer serious damage. The issue facing policymakers is not whether to address rising deficits, but when and how to address them. At some point, policymakers will have to increase taxes, reduce spending, or both.

Much of the pressure on the budget stems from the fast growth of federal costs on health care. So constraining that growth seems a key component of reducing deficits over the next several decades. A variety of evidence suggests that opportunities exist to constrain health care costs both in the public programs and in the health care system overall without adverse health consequences, although capturing those opportunities involves many challenges.

Health Affairs held a briefing on the future of health care reform earlier this week. A theme that emerged during the session I participated in with Jason Furman of Brookings, and that will be the focus of an upcoming speech, is the need for more attention to behavioral economics and psychology in health policy. The audio from the session is posted here, and the video is available here.

I am testifying this morning on SCHIP before the Subcommittee on Health of the House Energy and Commerce Committee. The testimony is very similar to testimony delivered before the Senate Finance Committee in April. The audio Webcast of the hearing is posted here.

Last month, I participated in an event on mortgage and housing markets with Alan Blinder of Princeton and Zanny Minton Beddoes of The Economist magazine. The event was co-hosted by the Woodrow Wilson School and The Economist, and video from it is now available here .