Congressional Budget Office
Douglas W. Elmendorf
Director’s Blog

I’m speaking today at a conference held by the Tax Policy Center and the American Tax Policy Institute on “Taxes and Health Insurance: Analysis and Policy.” The slides for my talk and the audio from the session are posted here. They summarize CBO’s analysis (based in part on revenue estimates from the Joint Committee on Taxation) of the Administration’s 2007 proposal to alter the tax incentives associated with health insurance. Our estimate of the budget impact of the Administration’s 2008 proposal (which is slightly different from the 2007 proposal) will be included in the summary of CBO’s analysis of the President’s budget that we will release on Monday morning; we will issue a more detailed report on March 19.

One point that has received little attention but that I will emphasize at the conference — and that is also noted in a paper by Jeff Liebman and Richard Zeckhauser that is being presented there — is that most workers seem to have little knowledge of how much after-tax wages they are forgoing in exchange for their employer-provided health insurance. (Liebman and Zeckhauser relate a story about how difficult it was, even for a motivated professor with a PhD in economics, for one of them to obtain that information from their employer, Harvard University.) One result of this is less pressure for efficiency in the health sector. Indeed, given the apparent size of inefficiencies in health care, it may turn out that one of the biggest distortions from the tax code occurs through this channel.

I spoke this past weekend about long-term care — a topic of particular concern to many governors given the role of Medicaid in its financing — at the National Governors Association Winter Meeting. My slides from the NGA talk are posted here, and the webcast from the session is here. CBO has done important work on long-term care; see for example the study authored by Stuart Hagen of our Health and Human Resources Division in 2004.

CBO released a report today on the deductibility of state and local taxes on federal income tax returns, which provides an indirect federal subsidy to the state and local governments that levy deductible taxes. The state and local tax deduction reduced federal revenues by an estimated $50 billion in fiscal year 2007.

CBO’s report, which was prepared at the request of the Ranking Member of the Senate Budget Committee, examines the justifications for the state and local tax deduction, how its benefits are distributed among different groups of taxpayers,  interactions between the deduction and the Alternative Minimum Tax (AMT), and options for modifying or eliminating the deduction.

The main points of the report include:

• Whether the deduction is an efficient use of federal resources depends on the nature of the benefits from any services at the state and local level that it subsidizes.
  • To the extent that state and local taxes are payments by residents of those jurisdictions for services that they themselves receive from their state and local governments, the rationale for a federal subsidy is weak.
  • In contrast, if state and local taxes pay for services that have spillover benefits that are regional or national in nature, then a federal subsidy may be desirable to ensure that an adequate volume of such services is produced.
• Some evidence suggests that state and local governments may respond to the taxes-paid deduction not by imposing higher taxes but by simply using deductible taxes in place of some nondeductible taxes.
• In 2004, slightly less than 35 percent of all taxpayers deducted state and local taxes they had paid, but whether a taxpayer claimed the deduction and the value of that deduction varies considerably according to income.   Taxpayers with incomes below $75,000 in 2004 accounted for more than 80 percent of all taxpayers but less than 20 percent of the total tax benefit from the deduction; taxpayers with incomes above $1 million accounted for 0.2 percent of all taxpayers and 16 percent of the total tax benefit from the deduction.
• In 2004, potentially deductible taxes accounted for 17 percent of state revenue and about 40 percent of local government revenue (in both cases excluding revenue received from another government or from government-run entities like utilities).
• Over the next several years, scheduled changes to tax law and the interaction of the regular income tax and AMT will change the number of taxpayers who claim the deduction and the associated loss of federal revenues.  The amount of that loss is projected to diminish through 2010, because a growing number of taxpayers will pay the AMT, which does not allow people to claim the deduction.  The scheduled expiration after 2010 of tax provisions enacted in 2001 and 2003 will boost income tax rates for many taxpayers, raising the value of the taxes-paid deduction for those who claim it and increasing the associated revenue loss for the federal government.
• With assistance from the Joint Committee on Taxation in estimating budgetary effects, CBO analyzed many options for changing the deduction with and without changes to the AMT.   For example, eliminating the deduction and indexing the AMT to inflation would, in combination, raise federal revenue by about $450 billion over the next decade.  Replacing the deduction with a 15 percent credit while indexing the AMT to inflation would reduce revenue by $330 billion from 2008 to 2017.  Other options are discussed and analyzed in the study.

The study was written by Kristy Piccinini of our Tax Analaysis Division. Kristy works in the areas of state and local taxation and tax-exempt bonds. She joined CBO in 2006 after receiving her Ph.D. in economics from the University of California Berkeley. Her dissertation examined how state tax legislation responds to changes in the budget balance.

CBO released a new study today on regional variation in health care costs. As previous CBO documents have emphasized (and I walk around with a chart about it!), health care spending varies substantially across the nation.

This geographic variation matters mainly as an indicator of the efficiency of the health sector: Large differences across the country in spending for the care of similar patients suggest a health care system that is not as efficient as it could be, particularly since the higher spending does not appear to produce commensurately better care or improved health outcomes.

CBO’s study finds that:

• A substantial portion of the variation cannot be explained by the health status of the populations in different regions and other variables:
  • The severity of illness and the prices of health services across regions together account for less than half (and possibly much less than half ) of the geographic variation in spending.
  • Income and the preferences of individuals for specific types of care appear to explain little of the variation in spending.
• Some regions appear more prone to adopt low-cost, highly effective patterns of care whereas others are more prone to adopt high-cost patterns of care and to deliver treatments that provide little benefit or are even harmful.
• Geographic variation in Medicare spending has dropped sharply over the past three decades and recently has been slightly lower than the variation in total health care spending.
  • The coefficient of variation, a common way to measure dispersion, in Medicare spending across states fell from a peak of 0.20 in 1976 to 0.125 by 1991, and then rebounded in the early 1990s before resuming a sharp decline, ending at 0.11 in 2005.
  • That reduction may be the result of changes in Medicare’s reimbursement policies.
  • In contrast to Medicare spending, geographic variation in total health care spending per capita has trended upward in recent years.

• In recent years, geographic variation in health care spending has been much higher in the United States than in Canada, and somewhat higher than in the United Kingdom. Financing of health care in those countries is more centralized than it is in the United States.
  • From 1991 through 2004, the coefficient of variation in state-level health care spending per capita in the United States varied between 0.112 and 0.123.
  • Over the same period, the coefficient of variation in per capita spending by province in Canada (for public and private spending) varied between 0.059 and 0.088, with an increase in recent years.
  • In the United Kingdom, the coefficient of variation by region has varied in recent years between 0.091 and 0.107.

• In recent years, geographic variation in spending in VA health care system has been similar to that in Medicare, despite the fact that the VA system uses an explicit allocation formula to distribute funds to regions.

• The evidence suggests that efficiency gains in the health care system are possible: Spending in high-spending regions could be reduced without producing worse outcomes, on average, or reductions in the quality of care. Policies that reduce spending in high-spending areas, however, will not necessarily lead to increased efficiency unless the reductions target ineffective or harmful treatments. The report briefly explores policy options that could reduce geographic variation, including:
  • Increasing the “bundling” of services in payments to providers (such as those that have been implemented in the Medicare program for payment for hospitals), which could help to curb current incentives to provide more intensive services that produce only modest or no improvement in health.
  • Enhance incentives to provide care consistent with accepted guidelines for low-cost, highly effective care, thus helping to change patterns of medical practice in places that now are characterized by lower-quality, higher-cost care.
  • Generate more information about variations in practice patterns and the relative cost-effectiveness of different procedures for different populations as a way to help reorient inefficient practice patterns toward greater efficiency, especially if greater oversight or changed financial incentives led to increased pressure to use this sort of information.

Throughout this year, CBO is undertaking an expanded effort to examine options for modifying the health care system in the United States, which will ultimately result in two significant published volumes. The discussion of those options will include their potential impact on geographic variation.

The study was written by David Auerbach and Chapin White of our Health and Human Resources Division.

David Auerbach has been with the CBO since receiving his Ph.D. from Harvard University in 2002. His other work has focused mainly on issues related to health insurance coverage, including work developing a simulation model for analysis of proposals affecting the number of uninsured (described in a CBO background paper released in 2007). He has also worked extensively on topics such as Medicaid coverage, the individual health insurance market, and the effects of competition and tax incentives on health insurance premiums. He has master’s degrees in technology and policy from MIT and in chemistry from UC Berkeley and received his undergraduate degree in chemistry from MIT. He is also an expert and co-author of an upcoming book on the nursing workforce.

Chapin White joined CBO in 2004 and is currently a Principal Analyst in the Health and Human Resources Division. His work focuses on health care financing, health insurance markets, medical malpractice, and nonprofit hospitals. His CBO publications include The Impact of Medicare’s Payment Rates on the Volume of Services Provided by Skilled Nursing Facilities, Nonprofit Hospitals and the Provision of Community Benefits, and Medical Malpractice Tort Limits and Health Care Spending. His publications include articles in Health Affairs, Inquiry, Health Services Research, and the Health Care Financing Review. He received his A.B. in social anthropology from Harvard College, M.P.P. from Harvard’s Kennedy School of Government, and Ph.D. in health policy from Harvard University.

CBO released a new macroeconomic forecast today, in advance of a new set of budget baseline projections we will release in conjunction with our analysis of the President’s budget.  (We have occasionally issued these types of revisions before but they are relatively uncommon.)  The revision is motivated by three recent developments: new data about the weakness of the economy, actions by the Federal Reserve, and the stimulus package passed by the Congress and signed into law by the President.

CBO’s previous forecast, which was embodied in budget projections released in January, was finalized in early December 2007.  Data released since then––especially regarding the labor market––indicate that economic conditions are weaker than previously projected, and conditions in some segments of financial markets remain worrisome. Other indicators––such as production indices and information on retail sales and sales of new homes––also suggest a slowing in economic activity. At the same time, changes in monetary policy have been more substantial than CBO assumed in December, and fiscal policy stimulus has been enacted. The Federal Reserve reduced the target for the federal funds rate by 125 basis points in January, and financial markets anticipate further easing in the near future. In addition, the Economic Stimulus Act of 2008 will provide about $150 billion in tax rebates and business tax deductions in fiscal year 2008. CBO anticipates that the recent monetary and fiscal policy actions will provide significant support to the economy in 2008. The net effect of those developments since CBO’s previous set of projections is slightly stronger projected economic activity for 2008 (because the impact of monetary and fiscal policy stimulus slightly outweighs the deterioration in economic conditions absent those policy changes) and slightly weaker projected economic activity for 2009 (in part because the withdrawal of fiscal stimulus temporarily reduces economic growth). CBO’s projections are similar to the most recent Blue Chip consensus forecast, an average of the estimates of about 50 private-sector forecasters. Although CBO’s projections do not show the slowdown in economic growth becoming severe enough to meet the economic definition of recession, the risk of a recession remains elevated, and economic activity will remain subdued for some period as the economy continues to work through the effects of problems in the housing and financial markets and the high price of oil.  More specifically, CBO now forecasts that real GDP will grow by 1.9 percent in calendar year 2008 and 2.3 percent in 2009.   The previous projections had been 1.7 percent and 2.8 percent respectively for 2008 and 2009.

CBO released a new report today on policy options for reducing carbon emissions.

Global climate change represents one of the nation’s most serious long-term problems. Rising concentrations of CO2 and other greenhouse gases are gradually warming the Earth’s climate, and some risk exists that the buildup of those gases could trigger abrupt changes and extreme damage. Substantially reducing emissions of those gases, however, could impose significant costs on the U.S. and global economies.

Minimizing the costs involved in reducing emissions would require a reliance on incentive-based policies that provide flexibility about where and how such reductions will take place, rather than more restrictive command-and-control style policies (such as technology standards).

In this study, CBO examined a variety of incentive-based policies for reducing CO2 emissions, including a tax and a cap-and-trade system:

• A tax would set an upper limit on the cost of emission reductions—firms would undertake reductions that cost less than the tax—but would leave the amount of emissions uncertain.
• An inflexible cap-and-trade program would set an upper limit on the amount of emissions but would leave the cost of reducing emissions uncertain.
• A flexible cap-and-trade program would maintain the structure of a cap-and-trade program, but would include features designed to limit the cost of meeting the cap. Specifically, a cap-and-trade program could include one or more of the following:
  • A price ceiling (often referred to as a safety-valve) and/or a price floor;
  • Provisions that permit firms to “bank” unused allowances in one year for use in a future year and/or “borrow” future allowances for use in an earlier year;
  • Provisions to make the cap less stringent if the price of allowances rises beyond an agreed upon amount. A “circuit breaker” would directly modify the cap. Alternatively, the government could indirectly modify the cap by changing the terms under which firms could use borrowed allowances.

The study examines the relative efficiency and administrative issues surrounding these approaches. It should be noted that other criteria could be of interest to policymakers in determining how best to address concerns about climate change. For example, economic efficiency addresses how well policies might function to minimize the cost of reducing emissions over a period of several decades; however, policymakers may choose to place more emphasis on providing certainty about the amount of emissions at specific points in time. Similarly, policymakers may also wish to focus on how different policy designs affect different segments of society.

The study finds that:

• A tax could achieve a long-term emission reduction target at a much smaller economic cost than an inflexible cap.
  • Provided that the tax was set equal to the expected benefit of reducing a ton of CO2, a tax could thus result in substantially greater net benefits (benefits minus costs) than a comparable cap-and-trade program.
  • The advantage of a tax stems from the long-term nature of climate change (which depends on the build-up of emissions over many decades, but is not sensitive to the amount of emissions in any given year) and the uncertain and variable nature of the cost of reducing emissions (which will vary from year to year based on the weather, conditions in energy markets, and the availability of new technologies).
  • An inflexible cap-and-trade program would provide more certainty about annual emissions than would a tax; however, that certainty would come at a cost: The cap would require too many reductions when the cost of achieving them was high and would mandate too few reductions when the cost was low.
• Flexible cap-and-trade programs could achieve some, but not all, of the efficiency-improving/cost-minimizing advantages of a tax:
  • Out of the flexible cap designs that CBO considered, a cap-and-trade program that included both a safety valve and either a price floor or banking provisions could offer the greatest potential to minimize the cost of meeting a given long-term target.
  • Including a circuit breaker, or altering the extent to which firms could use borrowed allowances, could help prevent the price of allowances from going higher than policymakers wanted. Either approach, however, would be less direct, and less effective than including a safety valve.
• Either a tax or a cap-and-trade program could be relatively easy to implement.
  • Some flexible design features, such as banking, borrowing or a safety valve, would be straightforward to implement.
  • In contrast, price volatility in the allowance market could make it difficult for the government to know when to implement a circuit breaker (or to change the terms associated with borrowing allowances).
• Minimizing the cost of reaching a global emissions target would entail undertaking the lowest-cost emission reductions regardless of where in the world they were located.
  • If coordinated among emitting countries, a tax would help minimize the cost of achieving any given target.
  • Linking the cap-and-trade programs of various countries could help minimize global costs, but could create some significant concerns:
    • Countries would give up sovereignty over the price of allowances traded in their programs.
    • Poor monitoring or enforcement in any one country could undermine the integrity of the allowances traded throughout the whole system.
    • Flexible design features, such as a safety valve, banking, or borrowing, would become available to all regulated entities in the linked system.
  • Major emitting countries could help minimize global cost of reducing emissions by establishing national cap-and-trade programs that each included a safety valve set at roughly the same level.

The report was written by Terry Dinan, who joined CBO in 1989.  She is a member of the Microeconomic Studies Division and serves as CBO’s Senior Advisor for Climate Policy.  She has written numerous CBO papers examining the effects of alternative policies for reducing carbon dioxide emissions as well as addressing other environmental issues, including CAFE standards, drinking water regulations, and the control of ground-level ozone.  She has published in a variety of journals, including the Journal of Environmental Economics and Management, the National Tax Journal, the Journal of Urban Economics, the Journal of Regulatory Economics and the Natural Resources Journal.  She served as an associate editor for the Journal of Environmental Economics and Management and was on the board of directors for the Association of Environmental and Resource Economists.  She has a Ph.D. in economics from Iowa State University.  Terry drives a Prius and rides her bike to work when her schedule permits.

Today CBO released a new study analyzing the increases in funding for military activities in Iraq, Afghanistan, and elsewhere in the war on terrorism over the past several years.

• The United States began combat operations in Afghanistan in fiscal year 2002 and in Iraq in fiscal year 2003. To finance those operations (and others related to the war on terrorism), the Congress provided $18 billion and $76 billion in emergency appropriations in those years, respectively.
• With the exception of a slight decrease in 2004, to $74 billion, funding has increased steadily each year, to a total of $165 billion for 2007.
• If the Administration’s request for 2008 is funded in full, appropriations for military operations in Iraq and Afghanistan, and for activities elsewhere in the war on terrorism will rise to $188 billion this year and to a cumulative total of $752 billion since 2001.
• Most of the spending is concentrated in two categories — operation and maintenance (O&M), which has roughly doubled from 2004 to 2008, and procurement, which has increased tenfold over that period.
• Prior to 2005, funding for military operations was largely limited to the incremental amounts needed to mobilize and deploy troops, transport equipment and supplies, and purchase additional quantities of consumables such as fuel, repair parts, and munitions. War funding also paid for an increase in the number of service members on active duty. About 60 percent of appropriations provided during this period went to O&M accounts and 20 percent went to military personnel accounts.
• Beginning in 2005, as part of its request for war funding, DoD asked for appropriations to “reset” equipment, that is, to repair or replace worn or damaged equipment. Those efforts include major overhauls that restore the item to “like new” condition. At the same time, DoD often added major upgrades to repaired items, returning equipment to the field with significantly enhanced abilities; these upgrades involved much higher costs than simply repairing equipment. Most such efforts are funded through the O&M and procurement accounts. During this phase, O&M funding continued to account for roughly 60 percent of total funding.
• In 2006, DoD began widening its focus from resetting equipment to “reconstituting” the force, an effort that involved purchasing new equipment as well as repairing and replacing damaged systems. Whereas the reset program had required more O&M funding, the shift to reconstitution increased the need for procurement funds.
• In 2007, DoD expanded the list of expenses that could be included in the request for war-time appropriations. In addition to seeking funds to pay for the direct incremental costs of operations in Iraq and Afghanistan, the services were permitted to include costs related to the broader war on terrorism. DoD requested funds to replace damaged equipment with newer models, accelerate planned purchases of new systems, address emerging needs, and enhance the military’s capability not only to continue current operations but also to be better prepared for the longer war on terrorism. Achieving the goals of that expanded reconstitution program required significantly more procurement spending.
• Thus, in 2007 and 2008, procurement funding soared, averaging about 35 percent of total war funding in those years. While O&M funding continued to increase and funds for military personnel held steady, those accounts fell to an average of 52 percent and 10 percent of total war funding, respectively.
• If the Congress provides the remaining $101 billion that DoD has requested for the war in 2008, annual funding levels will have increased by 155 percent since 2004. Increases in procurement and in operation and maintenance account for almost all of that growth. Appropriations for military personnel have changed little, and other DoD appropriations contribute relatively small amounts to the total.

David Newman and Jason Wheelock of the defense unit in our Budget Analysis Division put the analysis together.

Among some parts of the health policy community, CBO appears to have acquired an undeserved (and unfortunate!) reputation for failing to incorporate offsetting savings into our cost estimates. Since I hear about this issue so frequently, I thought it would be worthwhile to do a brief post about it.

To begin, let me be very clear: We are firmly committed to reflecting savings in our cost estimates, whether in health care or any other area, when (a) there is relevant evidence suggesting such savings, (b) the savings accrue to the federal government rather than to other parts of the economy, and (c) the savings occur within the relevant budget windows. Unfortunately, many examples that advocates cite as offering potential savings don’t meet at least one of those three necessary conditions.

As a specific example of offsetting savings that do meet these conditions, consider the tobacco legislation (S. 625) that has been reported out of the Senate Committee on Health, Education, Labor and Pensions. Our score of that legislation includes savings to Medicaid from a reduction in low-birth weight babies triggered by reduced smoking among pregnant women. As the score notes:

“CBO anticipates that the decline in smoking due to FDA’s regulation of tobacco products also would reduce the number of women on Medicaid who smoke during pregnancy. This reduction would lead to lower spending by the Medicaid program - which covers about 40 percent of all pregnancies in the United States - because women who do not smoke are less likely to have miscarriages, experience complications during pregnancy, and give birth to children with low birth weights. A variety of research indicates that children with low birth weights have higher medical costs, particularly at birth, but also later in life. Savings of some such costs would be partly offset by higher costs for additional live births because of the decline in miscarriages. On net, CBO estimates that FDA’s regulation of tobacco products would reduce federal Medicaid spending by $78 million over the 2009-2018 period.”

Admittedly, the net offsetting savings are modest relative to Medicaid’s total outlays — but that is partly because of the 10-year budget window and also partly because, as the document notes, fewer miscarriages means more births and more kids, which can raise costs for Medicaid.

That raises an important point: improved health outcomes, even if they are desirable, do not always mean reduced costs for the federal government. Indeed, many health benefits would be associated with reduced smoking — but the impact on health care costs in other federal health programs beyond Medicaid is less clear, especially over a 10-year window. CBO therefore concluded, “Reduced smoking levels may have additional effects on other federal health care programs. However, CBO did not estimate any additional effects, because the magnitude and direction of those effects is less certain than the impact of reduced smoking levels on pregnancies covered by Medicaid.”

As CBO continues to expand its work on health care, we will continue to assess and explore possible sources of offsetting savings. We draw upon our panel of health advisers to help us identify new academic studies that may be relevant in this effort, and I would welcome additional suggestions from the policy community to make sure that we continue to produce the best possible cost estimates. And we all need to remember that although opportunities exist to improve health outcomes and reduce overall health care costs, many steps that would lead to better health outcomes don’t necessarily reduce federal budget costs – which is the focus of the scoring process — given the institutional structure of the federal health programs.

CBO released its monthly budget review this afternoon.  The deficit for the first four months of fiscal year 2008 was $90 billion — almost $50 billion higher than during the same period last year, although about a third of that increase was due to timing shifts.  Corporate tax receipts were down about 10 percent, and have declined in each of the past seven months on a year-over-year basis.